key: cord-0041919-kt5gt26t
authors: nan
title: Poster Session Abstracts
date: 2007-08-29
journal: Pediatr Pulmonol
DOI: 10.1002/ppul.20700
sha: f914f7d003858ae8cef6e260f4fd6ac449a5ae4c
doc_id: 41919
cord_uid: kt5gt26t

nan

causing mutation is ∆F508, making studies of NBD1 essential in understanding CFTR function.

Crystal structures have been solved of wild-type (WT) and mutant CFTR NBD1, including variants containing ∆F508 (Lewis et al. EMBO J. 2004; Lewis et al. J. Biol. Chem. 2005; Thibodeau et al. Nat. Struct. Mol. Biol. 2005) . CFTR NBD1 contains a regulatory insertion (RI) and a regulatory extension (RE) that contain PKA phosphorylation sites. Although the NBD core is very similar in the various structures, the RI and RE in some structures of human WT NBD1 differ by 180°rotations, indicating that they are mobile.

We have used NMR spectroscopy to study WT and ∆F508 NBD1 in the phosphorylated (phospho) and non-phosphorylated (non-phospho) states, and bound to different regions of CFTR. Despite their similar structures, differences are observed in the NMR spectra of WT and ∆F508 NBD1. These changes, possibly due to altered dynamics or interactions of the RI and RE with the NBD core, will be discussed. The inherent stability and function of NBD1 could be affected by these changes, accounting for part of the ∆F508 defect.

Different interactions between WT and ∆F508 NBD1 with the ICLs could also affect channel function. To identify residues comprising the ICLs, we generated a homology model of CFTR based on the crystal structure of Sav1866 in which NBD1 and NBD2 form a productive dimer (Dawson and Locher. Nature. 2006) . Our data on a peptide corresponding to the first intracellular loop (ICL1) indicate that binding to NBD1 requires phosphorylation of the RI. Our homology model shows that the ICL1 binding site is buried when the RI is bound to the NBD core. NMR data comparing phospho and non-phospho NBD1 indicates that phosphorylation of the RI and RE disrupts their interactions with the NBD core, likely exposing the site for ICL1 binding. The structure of the molybdate transporter ModB2C2, with the NBDs in an open conformation, shows ICL1 binding is cognate NBD at a different interface (Hollenstein et al. Nature. 2007) . ICL1 in ModB2C2 interacts with the NBD near Y93, the analogous residue to F508 in CFTR. We will show binding data of ICL1 with WT and ∆F508 NBD1 in different states and also present resonance assignments of NBD1 to map the ICL1 interaction sites. Since the Sav1866 and ModB2C2 structures were solved in different states, the different ICL interactions may represent structural changes during the reaction cycle. The ∆F508 mutation may thus affect intramolecular associations in certain conformations, which would account for another part of the ∆F508 defect. Probing differences in interactions of WT and ∆F508 NBD1 is critical for understanding the molecular basis of normal CFTR function and of the altered CFTR function in cystic fibrosis.

SHSPS TARGET ∆F508 CFTR FOR ERAD VIA THE SUMO PATHWAY Ahner, A. 1 ; Brodsky, J.L. 2 ; Frizzell, R.A. 1 1. Cell Biology and Physiology, University of Pittsburgh, Pittsburgh, PA, USA; 2. Biological Sciences, University of Pittsburgh, Pittsburgh, PA, USA The most frequent disease-causing mutation in cystic fibrosis (CF) is a deletion of phenylalanine at position 508 (dF508) in the first nucleotide binding domain of CFTR. Essentially all of the mutant protein, as well as 70% of wild type (WT) CFTR, is retained in the ER and targeted for ERassociated degradation (ERAD) by the 26S proteasome. Key mediators of CFTR folding and degradation are molecular chaperones, which help protein substrates fold, but can target them for degradation if folding efficiency is compromised. To identify factors modulating ERAD of CFTR we performed a microarray analysis, screening for transcription profile differences between yeast expressing CFTR and control strains, and identified enhanced transcript levels for the small heat shock protein Hsp26 in yeast expressing CFTR. We then demonstrated that CFTR degradation was blocked in strains lacking the genes encoding the two yeast sHsps, HSP26 and HSP42 (1) . To examine whether sHsps regulate CFTR biogenesis in mammalian cells, we asked which of the 10 human sHsp homologues are endogenously co-expressed with CFTR using RT-PCR and Western blot analysis. We detected 5 sHsps, including alphaA-crystallin and Hsp27, in Calu-3 and primary HBE cells. Co-expression of alphaAcrystallin or Hsp27 together with WT or DF508-CFTR in HEK293 cells selectively decreased the steady state levels of dF508-CFTR. Pulse-chase experiments showed that the rate of dF508-CFTR degradation was enhanced when either sHsp was over-expressed, but WT CFTR biogenesis was unchanged. Co-immuno-precipitation experiments in HEK 293 cells revealed that alphaA-crystallin and Hsp27 interacted preferentially with dF508-CFTR. Thus far, our results suggest that sHsps selectively increase dF508-CFTR's accessibility to proteasome-mediated degradation pathways. Previously, Hsp27 was reported to interact with Ubc9, the SUMO (small ubiquitin-like modifier) conjugating enzyme (2) . SUMO is covalently linked to its substrates, often at sumoylation consensus sites; similarly to ubiquitin, this occurs through a series of thiol transfer reactions. We confirmed this interaction by co-immuno-precipitation and found that Ubc9 as well as the SUMO specific protease, Senp1, are expressed in airway epithelial cells. Over-expression of Ubc9 decreased, while over-expression of Senp1 increased dF508-CFTR protein levels. Pulse-chase experiments indicated that these enzymes regulate selectively the degradation of dF508-CFTR, similarly to the sHsps. Preliminary studies in vitro and in vivo indicate that CFTR, and particularly purified NBD1, is sumoylated and that Hsp27 facilitates this process. Mutational analysis of the 3 sumoylation consensus sites within CFTR suggests a cycle of SUMO modification that facilitates WT CFTR biosynthesis but targets dF508-CFTR for degradation, probably because the balance between SUMO addition and removal is impaired by an excessive interaction of sHsps with the mutant protein.

[Supported by grants from the NIH and the Cystic Young, A. Physiology and Biophysics, Chicago Medical School, North Chicago, IL, USA Efficient retrieval of wtCFTR from the cell surface is mediated by components of the clathrin mediated endocytic pathway; however many aspects of CFTR endocytosis remain to be elucidated. We have previously shown that endocytosis of wtCFTR is dependent upon recognition of a tyrosinebased motif in the carboxyl tail of CFTR by the AP2 endocytic adaptor complex. In contrast to wtCFTR, less is known about the mechanisms whereby mutant CFTR undergoes internalization. According to one model, dF508 CFTR undergoes accelerated endocytosis compared to wtCFTR, whereas others have argued that endocytic rates of dF508 CFTR are comparable to those of wild-type CFTR. To further define the function of proteins involved in the initial steps of wt and mutant CFTR endocytosis, we investigated the role of the GTPase dynamin. Dynamin has been proposed to play a key role in endocytosis by facilitating the scission of nascent clathrin coated vesicles from the cell surface. Until recently, very few tools were available to modulate clathrin mediated events. We took advantage of Dynasore, a newly discovered small molecule, non-competitive inhibitor of the GTPase dynamin. Dynasore acts as a potent, rapid and reversible inhibitor of endocytic pathways known to depend upon dynamin, by blocking coated vesicle formation within seconds of dynasore addition. Endocytosis of wtCFTR was inhibited rapidly and efficiently upon addition of dynasore. Furthermore, within minutes of dynasore washout, endocytosis of wtCFTR resumed completely. Also upon dynamin washout, normal trafficking and recycling of wtCFTR was also observed. Temperature correction of dF508 causes expression of CFTR at the cell surface, however such 'rescued' dF508 CFTR is not stabilized upon returning cells to 37C. Application of dynasore prevented such rapid loss of mutant CFTR from the cell surface. Since maintenance of mutant CFTR at the cell surface is critical for effective therapeutic intervention in patients with CF, our results provide insight into additional potential targets for pharmacological manipulation for the treatment of CF.

Richardson, J.M.; Thibodeau, P.H.; Watson, J.; Thomas, P.J. Physiology, UT Southwestern Medical Center, Dallas, TX, USA Protein misfolding is the basis for a multitude of human diseases; however, the mechanisms underlying misfolding are not well understood. Most cases of cystic fibrosis are associated with mutations-including the most common, deletion of phenylalanine 508 (∆F508) in NBD1-that interfere with the folding of the cystic fibrosis transmembrane conductance regulator (CFTR). The resulting loss of functional CFTR causes the disease. Thus, elucidating how ∆F508, affects the folding of CFTR is critical to understanding the pathology of the disease. CFTR is composed of five domains: two integral membrane transmembrane domains (TMDs), a regulatory domain (R), and two nucleotide binding domains (NBDs). ∆F508 occurs in the N-terminal NBD1. Both the murine wildtype and ∆F508 NBD1 can be expressed in bacteria and purified to near homogeneity. While the soluble expression yield of ∆F508 is lower than the wild type under identical conditions, ∆F508 achieves a native conformation similar to wild type as monitored by a variety of hydrodynamic and spectroscopic characteristics such as analytical size exclusion chromatography, circular dichroism, and fluorescence. Recently, a non-native, but folded, species has been detected under mildly denaturing conditions. Far-UV CD reveals a time and temperature dependent conversion from a mixed alpha/beta native conformation to a less helical non-native conformation, while fluorescence measurements reveal a parallel blue shift in the peak emission intensity from 343 to 330 nm. This non-native species is in a more open conformation as determined by both limited proteolysis and the change in retention time on analytical size exclusion chromatography. The conversion to this state is inhibited by the native state ligand (ATP) and by the presence of the second site suppressors (G550E, R553M, and R555K). Notably, the ∆F508 NBD1 protein populates the non-native state under milder conditions than the wild type NBD1. These studies reveal the properties of the native state and its conversion to a partially folded state that is affected by the ∆F508 mutation. This data highlight the proximal affect of the disease causing mutation and provide a means of assessing strategies designed to correct the defect. This work is supported by the Cystic Fibrosis Foundation through a postdoctoral training grant and by the NIH We have previously demonstrated that the stability of CFTR is negatively modulated by the CFTR interacting protein, CAL. CAL competes with NHERF in binding to the PDZ motif at the C-terminus of CFTR and directs CFTR trafficking to the lysosome. Recently, silencing of CAL or overexpression of NHERF has been shown to restore chloride transport activity in ∆F508-CFTR cell lines. To identify the molecular machineries involved in CAL dependent degradation of CFTR, we investigated the role of the CAL interacting SNARE protein syntaxin 6 (STX6). STX6 protein expression and function were manipulated by siRNA silencing, over-expression, and a dominant-negative mutant. We also performed a functional study of chloride transport activity in cystic fibrosis bronchial epithelial cell line CFBE41o-stably expressing either wt-CFTR or ∆F508-CFTR.

The endogenous STX6 and CAL protein expressions were knocked down by siRNA techniques in multiple human cell lines including HEK293, Hela cell stably expressing ∆F508-CFTR and CFBE41o-stably expressing either wt-CFTR or ∆F508-CFTR. In HEK293 cells, knockdown of either STX6 or CAL augments the expression of transfected GFP-tagged CFTR (GFP-CFTR), which is consistent with the observation that overexpression of either STX6 or CAL enhances the degradation of CFTR. STX6 interaction with CAL was confirmed by a co-immunoprecipitation assay. Because of this interaction, we hypothesize that STX6 negatively regulates CFTR protein stability by interacting with CAL which in turn binds to the C-ter-minal PDZ motif of CFTR. Consistent with his hypothesis, STX6 knockdown has no effect on protein levels of a CFTR mutant lacking the PDZ motif (GFP-CFTR∆TRL). More importantly, CAL knockdown eliminates the inhibitory effect of STX6 overexpression on GFP-CFTR. A cell surface biotinylation assay and confocal microscopy showed significant increases in plasma membrane expression of CFTR in STX6 knockdown cells without gross changes in the expression pattern. Furthermore, consistent with its Trans-Golgi localization, STX6 knockdown has no effect on GFP-∆F508-CFTR expression. However, ∆F508-CFTR increases significantly in STX6 knockdown, temperature-rescued cells.

To assess the effect of STX6 in bronchial epithelial cells, STX6 was knocked down in CFBE41o-stably expressing wt-CFTR. Consistent with the observations in HEK293 and Hela cells, the expression of untagged wt-CFTR is also augmented. Knockdown of STX6 increases CFTR expression in a does-dependent manner, reciprocal to the expression level of STX6, without changing the level of expression of CAL or GAPDH. Furthermore, STX6 knockdown lead to a does-dependent increase in CFTR mediated short-circuit current that is stimulated by forskolin and inhibited by CFTR chloride channel inhibitor CFTRinh-172. More importantly, in temperaturerescued CFBE41o-stably expressing ∆F508-CFTR, STX6 knockdown tripled the CFTR mediated short-circuit chloride current.

These results not only delineate the effect of STX6 on post-Golgi trafficking of CFTR and its dependence on CAL-mediated PDZ-based interaction, but also points to it as a potential therapeutic target in conjunction with the ER-quality control based rescue of ∆F508-CFTR. Supported by the Cystic Fibrosis Foundation and the National Institutes of Health.

Seavilleklein, G. 1 ; Evagelidis, A. 2 ; Amer, N. 1 ; Chappe, F. 1 ; Hanrahan, J. 2 ; Chappe, V. 1 1. Physiology & Biophysics, Dalhousie University, Halifax, NS, Canada; 2. Physiology, McGill University, Montreal, QC, Canada CFTR channel activity is regulated by phosphorylation and de-phosphorylation of the R-domain. Direct PKC phosphorylation at specific sites in the R domain and NBD1 is essential and modulates CFTR activation by PKA. However, the molecular mechanism by which phosphorylation of the R domain induces channel activity remains unknown. We have reported biochemical evidence for the role of phosphorylation in domain-domain association using a deltaR-Split construct encoding the front (TMD1-NBD1) and back (NBD2-TMD2) halves of CFTR as separate polypeptides (1) . In agreement with the inhibitory role of the R domain, iodide effluxes and patch-clamp experiments confirmed that deltaR-Split channels were functional and constitutively active without cAMP stimulation. Co-transfecting a plasmid that directs expression of the R domain restored the PKA-dependence of re-assembled channels. Co-immunoprecipitations of the 2 halves of the deltaR-Split with a GST-R domain fusion protein in vitro or with the cotransfected R domain in BHK cells revealed that PKA phosphorylation enhances the R domain binding to the rest of the channel. We have now investigated the effect of PKC phosphorylation on the R domain binding. Our data demonstrate that PKC phosphorylation also induces a strong binding of GST-R domain in vitro or co-transfected R domain in vivo when BHK cells are stimulated with PMA. Interestingly, the combination of PKC+PKA phosphorylation induced a stronger binding than PKA alone. Mutation of all consensus PKC sites on the R domain to alanine (7CA) eliminated specific PKC phosphorylation but did not alter PKA phosphorylation in GST-RD. Western blotting and confocal microscopy experiments confirmed that expression levels of all 3 domains of the Split-7CA/RD are unchanged compared to Split-RD and co-localize at the plasma membrane of BHK cells. Basal activity of the Split-7CA/RD was similar to that of the Split-RD in iodide efflux experiments. However, activation by cAMP+IBMX was delayed and significantly reduced (36.92 % of Split-RD and 13.85% of WT-CFTR activation). These results confirm the functional re-assembly of the 7CA-RD with the front and back halves of the channel and the inhibitory role of the R domain. In co-immunoprecipitation experiments, the same level of interaction was observed between the 7CA/RD and RD with the Split halves suggesting that basal, inhibitory interaction of the R domain is independent of PKC phosphorylation. However, 7CA/RD binding with the Split channel was not increased by PKC or PKA stimulation. We conclude that PKC phosphorylation is essential for PKA-dependent binding of the R domain to the rest of the channel, and that the combined effect of these kinases further strengthens the R domain interaction. These structural changes are consistent with the essential role of PKC sites for CFTR activation by PKA as previously reported (2) . Supported by CIHR, NSHRF, DMRF & CCFF. G.Seavilleklein was supported by NSERC. (1) The in vivo folding mechanism of multidomain membrane proteins, including the cystic fibrosis transmembrane conductance regulator (CFTR), is poorly understood. CFTR, a member of the ABC transporter superfamily, contains two symmetrical halves, each consisting of a membrane spanning domain (MSD1 and MSD2) and a cytosolic nucleotide binding domains (NBD1 and NBD2), connected by the regulatory (R) domain. To address whether CFTR follows the archetypical domain-wise folding of multidomain soluble polypeptides and/or the cooperative domain folding mechanism, first we determined domain combinations that are necessary and sufficient to escape the endoplasmic reticulum (ER) quality control. One-, twoor three-domain assemblies failed to be processed efficiently in cells, measured by biochemical and morphological assays. The smallest folding unit that escaped the ER quality control was composed of a four-domain assembly, containing MSD1, NBD1, R or MSD2 as linear or split polypeptide. As a second approach we showed that CF-causing point mutations in the MSD1, NBD1 or MSD2 provoked the ER retention of the full-length CFTR with severe conformational defect in the NBD2, measured by limited proteolysis and immunoblotting, using domain spacific antibodies. The posttranslational folding kinetics of the four-domain folding unit was comparable to that of CFTR, suggesting that cooperative domain assembly is essential for the channel biogenesis. This mechanism provides an explanation for the processing defect caused by numerous CF mutations and outlines a possible paradigm for CFTR folding in living cells.

CFTR in post-Golgi compartments. Our results, jointly, uncover the role of N-glycans as a critical structural determinant of CFTR conformational maturation/stability in a chaperone-independent manner.

Karamyshev, A.L. 1 ; Patrick, A.E. 1 ; Johnson, A.E. 2 ; Thomas, P.J. 1 1. UT Southwestern Medical Center at Dallas, Dallas, TX, USA; 2. Texas A&M University, College Station, TX, USA Cystic fibrosis (CF) is caused by mutations in CFTR (Cystic Fibrosis Transmembrane Conductance Regulator). CFTR, a member of the ABC transporter family, is a multispanning membrane protein critical to chloride and water movement in secretory epithelia. Over one thousand mutations, in all parts of the CFTR sequence, have been identified, many of which disrupt the function of the protein to cause the disease. Some of the disease-associated mutations in the N-terminal portion of CFTR were predicted to alter cotranslational interactions with factors involved in the membrane integration of CFTR required for its proper function. To test this hypothesis, a photoreactive probe was incorporated into different positions of the N-terminal portion of CFTR by introducing amber stop codons and translating these messages in vitro in the presence of the amber suppressor tRNA N ε -(5-azido-2 nitrobenzoyl)-Lys-tRNA amb . Ribosome-associated nascent chains of a specific length containing the modified amino acid at a defined position were produced using truncated mRNAs. Some of these CFTR nascent chains photocrosslinked with proteins of approximately 50 and 100 kDa molecular mass. Crosslinking to the ~50 kDa protein was observed only after the TM1 span had emerged from the ribosomal tunnel. Moreover, truncation analysis demonstrated that TM1 is necessary for this interaction. In evaluating a range of nascent chains, the extents of photocrosslinking to the ~50 kDa and 100 kDa proteins often varied inversely. Immunochemical and knockdown/depletion studies indicate the ~50 kDa protein is SRP54, a subunit of the signal recognition particle involved in targeting ribosome nascent chain complexes to the translocon in the endoplasmic reticulum membrane. Highlighting the functional relevance of this interaction, in cells with reduced SRP levels, nonglycosylated forms of CFTR accumulate and total CFTR levels decrease. Notably, some CF-causing mutations in TM1 also reduce crosslinking to SRP. These data reveal that mutational inhibition of nascent CFTR association with SRP likely contributes to or mediates some types of CF.

Aleksandrov, L.; Aleksandrov, A.A.; Riordan, J.R.

The initial reversible steps of ATP binding to the first (NBD1) and second (NBD2) nucleotide binding sites of CFTR are divalent cation independent and dependent, respectively (Aleksandrov et al, J Biol Chem 277, 15419-25, 2002) . Subsequent to the initial binding, either Mg or Mn drive rapid hydrolysis at the second site while promoting trapping of the nucleotide at the first site. This occlusion at the first site of functional wildtype CFTR is somewhat similar to that which occurs when the catalytic glutamates in both of the hydrolytic sites of P-glycoprotein are mutated. This occluded state has been proposed to be the result of dimerization of the two NBDs and represent a transient intermediate formed during ATP hydrolysis (Tombline and Senior, J Bioenerg Biomembr 37, 497-500, 2005) . To test the possible relevance of this interpretation to CFTR, we have now characterized the process by which NBD1 occludes [ 32 P]-8N 3 ATP and [ 32 P]-8N 3 ADP. We have now found that only N 3 ATP and not N 3 ADP can be bound initially at NBD1 in the absence of Mg. Consistent with this Mg requirement for the NDP binding, N 3 ADP or ADP is able to compete for the N 3 ATP binding site only when the divalent ion is present. Despite the lack of a requirement for Mg for ATP binding, retention of the nucleotide triphosphate at the binding site at 37°C was dependent on the cation. However, at reduced temperature (4°C), N 3 ATP remains locked in the binding pocket with virtually no reduction over a 1 hour period even in the absence of Mg. These apparently paradoxical temperature effects hint that different molecular events can contribute to the occlusion of nucleotide at NBD1. That the occlusion process occurs exclusively at NBD1 and does not depend on its interaction with NBD2, was shown by experiments in which identical behavior was exhibited by ∆NBD2 constructs of CFTR. Thus, nucleotide trapping by wild-type CFTR, unlike that by mutant P-glycoprotein, is accomplished by events at a single domain (NBD1), rather than by dimerization of the two NBDs. (Supported by the NIH and CFF).

CFTR consists of five domains or regions, two regions that span the cell membrane (the TMDs), two cytoplasmic nucleotide-binding domains (the NBDs), and a regulatory (R) domain. The R domain is unique to CFTR: without phosphorylation of this domain, the CFTR chloride channel does not open. The TMDs and the NBDs of CFTR are similar to those found in many ATP-binding cassette proteins, the majority of which are transporters. The domains are fused in a single polypeptide chain, linked in the sequence: TMD1, NBD1, R domain, TMD2, NBD2. The first NBD is the site of the location of the mutation that is responsible for the development of cystic fibrosis in the majority of human patients (loss of phenylalanine 508). Although there is a structure for NBD1, we are still lacking a structure for the entire CFTR protein.

We have employed electron microscopy to provide low-to medium-resolution structures for the complete CFTR protein which was expressed in BHK cells and purified by affinity chromatography. Using single particle analysis combined with site-specific labelling using 1.8nm diameter Ni-NTA nanogold, we have identified the locations of NBD2 and the region around residue 696 of the R-domain within the low resolution 3D structure. By a process of elimination, this has also allowed the identification of NBD1 and the region occupied by the two TMDs. These low resolution studies have now enabled us to interpret medium resolution 3D structural data obtained for CFTR. Progress in obtaining two-dimensional arrays (crystals a single molecule thick) should allow higher resolution structural data to be obtained for the entire CFTR protein -sufficient for the identification of the transmembrane alpha helices. These data should allow a much clearer picture of the transmembrane channel formed by CFTR and will inform molecular modelling approaches aimed at the development of novel drugs for the treatment of cystic fibrosis.

We explored the gating process by introducing cysteines along the entire lengths of various transmembrane domains and extra-cellular loops, and studying their ability to chemically modified. We study the changes in their reactivity in closed and open channel states to characterize the changes in transmembrane domain conformations during gating. Using membrane impermeant MTS-reagents, we have characterized the conformational changes in TM1-ECL1, ECL3-TM6, ECL4-TM8, and ECL6-TM12 of CFTR. Our results suggest that ATP binding induces modest changes in the transmembrane domains and extracellular loops. Channel opening involves a rearrangement of the TM helices leading to decreased hydration of residues near the extracellular end. This conformational change is either a part of, or allosterically coupled to the gating mechanism. 14 PHYSIOLOGICAL ROLE OF NHERF1 IN THE REGULATION OF CFTR Raghuram, V.; Yang, Y.; Yaemsiri, S. Laboratory of Kidney and Electrolyte Metabolism, NHLBI/NIH, Bethesda, MD, USA In the apical membrane of epithelial cells, CFTR seems to exist within a multiprotein complex, in which its activity is regulated by interactions with other proteins. The c-terminus of CFTR contains a PDZ-binding motif, which binds to several PDZ domain-containing proteins, including members of Na+/H+ exchanger 3 regulatory factor (NHERF) family, NHERF1/EBP50, NHERF2/E3KARP, CAP70/NHERF3, NHERF4, and CAL/PIST/GOPC. PDZ domain proteins act as scaffolds to assemble signaling proteins to form signaling complexes. These signaling complexes are central to achieve specificity and efficiency of signal transduction. NHERF1 and NHERF2 have a c-terminal ERM domain, which interacts with cytoskeleton-associated ERM proteins. This interaction may anchor CFTR to the actin-cytoskeleton and restrict its movement within the membrane. In addition, several functional roles have been proposed for CFTR-PDZ domain interactions, including as an adapter to promote efficient PKA phosphorylation of CFTR, apical targeting and recycling of CFTR, regulation of single-channel activity, organization of β-adrenergic, and lysophosphatidic acid-2 receptor signaling complex, and trafficking of ROMK channels. Our current understandings of the role of CFTR-PDZ interactions are mostly obtained from heterologous expression studies. Very few studies have addressed the role of PDZ domain proteins in native epithelial cells or in organ physiology. We have used RNA interference techniques to produce a knockout or knockdown (KD) Calu-3 cell lines using a lenti-viral vector for stable suppression. Using four different shRNAs, we have generated four stable NHERF1-KD cell lines with varying levels of NHERF1 suppression. In these cells, NHERF1 expression ranged from 10-40% of Calu-3NT cells (control cells). In NHERF1-KD cells, steady-state CFTR protein level was greatly reduced, whereas its apical localization was unaffected. We also found structural defects in the apical microvilli. In polarized NHERF1-KD epithelia, membrane-permeable cAMP analogs were unable to stimulate CFTR mediated short-circuit currents. However, forskolin induced response was nearly normal despite reduced CFTR expression. Examinations of βadrenergic-, and adenosine-receptor signaling pathways suggest that NHERF1 plays a unique role in organizing the adenosine receptor signaling pathway. 

Abnormal retention of the dF508 CFTR mutated protein in lung epithelial cells underlies the pathology in a large proportion of individuals with cystic fibrosis. A drug discovery alliance between CFFT and BioFocus DPI was initiated with the aim to identify novel genes which upon shRNA-mediated knockdown were able to efficiently restore dF508 CFTR activity. A high-throughput assay based on an YFP halide reporter was developed in a human cystic fibrosis bronchial epithelial cell line (CFBE41o-), whereby the rate of halide influx directly correlated to dF508 CFTR activity.

BioFocus DPI's proprietary adenoviral shRNA libraries totalling 11,334 viruses were screened in this functional assay and yielded 753 hits. These 753 hits were repropogated to generate new adenoviral stocks and rescreened at three concentrations on the YFP halide reporter assay. 39 genes were eliminated due to potential cytotoxic effects, and hit calling was performed on the remaining 714 hits. A total of 315 duplicate hits were confirmed (44%). 7 of the confirmed hits had more than one shRNAs which targeted the respective gene. 36 confirmed hits were stronger in restoring dF508 CFTR function than the most potent positive control: syntaxin-8.

In order to elucidate which of these 315 hits induce halide influx via direct rescue and activation of dF508 CFTR, the YFP halide reporter assay was performed in the presence or absence of the CFTRinh172. These CFTR dependent hits will move forward into subsequent validation experiments whereby hits will be categorised into potentiators and correctors. The final goal of this program is to prioritize targets for entry into drug discovery. 

Restoring expression and/or function of mutated CFTR provides a means to treat Cystic Fibrosis. BioFocus DPI -A Galapagos Company -in collaboration with the Cystic Fibrosis Foundation Therapeutics (CFFT) has established a drug target discovery project that will enable the selection of proteins that rescue mutated CFTR.

By screening BioFocus DPI's proprietary adenoviral shRNA libraries in a high-throughput primary functional assay for CFTR activity in human cells, we have identified and confirmed 315 hits out of 11,334 that entered the primary screen.

To further validate these proteins as potential drug targets, we have set up a series of experiments and assays focused on the rescue of CFTR expression and its maturation by knock-down of specific targets. 1) In order to establish whether the knock-down of a specific target was able to promote dF508-CFTR plasma membrane localization, we performed immunofluorescence stainings of dF508-CFTR in a patient-derived human cell line carrying the dF508-CFTR mutation (CFBE41o-) grown in air-liquid interface culture.

A transduction protocol to deliver shRNA adenovirus to this polarized airway cell model system has been optimized and preliminary results showed increased levels of expression of dF508-CFTR at the plasma membrane level upon knock-down of syntaxin-8, our positive control. Experiments performed at 27 οC also demonstrated redistribution of dF508-CFTR to the apical plasma membrane at low temperature. We will show the results of all the shRNA hits in this assay.

2) Changes in the post-translational maturation levels of dF508-CFTR upon knock-down of specific proteins were evaluated by determining the accumulation of band B and C by using a Western Blot approach. In this way, we can select targets able to promote post-translational maturation of dF508-CFTR and/or its escape from ER in the CFBE41o-cell line. Accumulation of band B and C indicates maturation of CFTR and will be considered as a beneficial effect. Preliminary results show increased accumulation of band B upon knock-down of specific targets and overexpression of wild-type CFTR induces band C accumulation. Experiments performed at 27 οC also demonstrated increased band B expression. We will show the effect of all the shRNA hits in this assay on band B and C expression.

We believe that the results obtained by the combination of those two approaches will guide us in obtaining very strong drug targets for entry into a drug discovery phase.

Because gene therapy as a corrective measure for CF is possible in the foreseeable future, it is important to understand whether the global effects of gene introduction will have significant detrimental impact on cell function. To address this question, global gene mRNA expression profiles were created with the Affymetrix U133A & B chip sets. Wild type CFTR (N=1), ∆F508 (N=1) and G551D (N=1) were transfected in separate IB3 cell cultures using Lipofectamine 2000™. Control expression profiles were generated from untreated IB3 cells; one from this experimental set and five from previous control experiments. Differential gene expression was determined by ANOVA using a 10% false discovery rate threshold. GeneSpring was used for data analysis and visualization. GeneGo was used to determine which functional categories were over represented in the list and further assisted in interpretation of the results. RT-PCR was used to confirm differential expression of selected genes and examine the response in more detail.

All transfected constructs produced very similar gene expression profiles quite distinct from the untransfected cells. 256 probe sets representing 228 uniquely named genes were significantly different between transfected and untransfected cells. CFTR mRNA expression in all forms was increased an average of 525-fold. Transfection of IB3-1 cells resulted in upregulation of a variety of genes involved in several basic processes: inflammation, protein folding and cytoskeleton framework regulation. GeneGo canonical pathways with a significantly enriched number of genes from the data set involved glucocorticoid signaling and regulation, cytoskeleton remodeling and adhesion plus a variety of other signaling pathways including IL8, chemokines, g-protein and MAPK cascades. Cell processes overrepresented in the list were predominately associated with protein folding, inflammatory responses and cytoskeleton regulation. GeneOntology categories over represented in the list were regulation of signal transduction, receptor mediated endocytosis and protein folding.

The proinflammatory interleukins IL6 and IL 8 were upregulated 2.5 fold and 4 fold respectively. The genes most upregulated, SAMD9 (15-fold) and OASL (12 fold), are not well annotated.

As controls, RT-PCR analysis of cells treated with either Lipofectamine 2000™ alone or any form of CFTR DNA alone did not show upregulation of IL6, IL8, SAMD9 or OASL. In cells transfected with Lipofectamine containing DNA, upregulation of IL6 (N=3), OASL (N=3) and SAMD9 (N=3) was confirmed by RT-PCR.

OASL is known to interact with methyl CpG-binding protein 1, a DNA transcription repressor acting at methylated promoter regions. Thus it is possible that the global effects noted here leading to dramatic upregulation of two novel genes OASL and SMAD9 could be previously unrecognized consequences of gene therapies. Cystic fibrosis (CF) is a disease which is caused by mutations of the CF gene, which codes for the CFTR chloride channel. The ∆F508-CFTR mutation results in a mutant protein that undergoes transcription but fails to localize to the apical membrane, where it normally functions as a cAMP-regulated Clchannel. However, Clchannel activity can be detected at the plasma membrane when ∆F508-CFTR is overexpressed, synthesized at a reduced temperature, or in the presence of chemical chaperones. Thus, strategies that facilitate the translocation of ∆F508-CFTR to the plasma membrane by increasing CFTR transcription may be beneficial for the treat-ment of CF. Previous studies performed using fluorescence halide efflux measurements and short-circuit current voltage clamp have shown that treatment with PPARγ (peroxisome proliferator activated receptor gamma) agonists, such as pioglitazone and FLL (FMOC-L-leucine), resulted in an increased biosynthesis and trafficking of ∆F508-CFTR to the cell surface. This effect was at least partially due to increased ∆F508-CFTR expression. Further studies were designed to investigate the effect of PPARγ agonists on CFTR promoter activity. Two firefly luciferase-conjugated CFTR promoter plasmids, containing either 0.79 kb or 3.9 kb fragments, were transfected together with a control plasmid containing a Renilla luciferase in human alveolar basal epithelial (A549) cells. The ratio of the firefly to Renilla luminescence was used to measure CFTR promoter activity. PPARγ agonists increased CFTR expression acting both on 0.7 kb and 3.9 kb promoter fragments. These results were confirmed in quantitative RT-PCR studies. Transcription factor binding site analysis using current online programs and matrices showed three putative PPRE (peroxisome proliferator response element) consensus sequences in the CFTR promoter. The interaction of these sites with PPARγ was shown using EMSA (Electrophoretic mobility shift assay). Interestingly, none of these consensus sequences appear in the 0.7 kb construct, which still maintained a significant PPARγ activation response. Therefore, we hypothesized that a novel PPARγ agonist effector site(s) is involved in transcription activation of CFTR. To characterize this novel site(s), a series of deletion constructs to the promoter region were created, using mutation-introduced restriction sites in the original promoter construct. The new constructs ranged from 0.61 kb to 24 base pairs upstream of the transcription start site. In preliminary studies we found that at least one previously uncharacterized PPARγ effector site was present in the 0.61 kb promoter fragment that regulated CFTR expression. Additional experiments were performed using EMSA to determine the proteins and cofactors which controlled the optimal conditions for CFTR expression. The results of this study show that PPARγ agonists, currently prescribed to treat hyperglycemia and improve peripheral insulin resistance, may also have additional benefits for the treatment of CF. Our studies also show that pioglitazone and other PPARγ agonists, by increasing CFTR expression, are potent activators of Clflux in epithelial cells expressing ∆F508-CFTR.

This work was supported by grant from the Canadian Cystic Fibrosis Foundation decreased by 10 µM GSNO. The effect of GSNO on Hop expression was post-transcriptional. In CFBE41o-cells transfected with siRNA duplex specific for Hop (48 hr), Hop expression was modestly inhibited; however, GSNO dramatically augmented the inhibition. We next showed that knockdown of endogenous Hop by siRNA increased the steady-state levels of immature and mature forms of CFTR. Though overexpression of Hop did not significantly affect CFTR expression (immature and mature forms increased 1.3 ± 0.1 and 1.2 ± 0.3 fold respectively), in the presence of 10 µM GSNO, expression of both immature and mature CFTR forms significantly increased in the Hop overexpressing cells (3.6 ± 0.4 and 2.7 ± 0.2 fold respectively). Next, we treated CFBE41o-cells with a pulse of GSNO (10 µM; 4 hr); isolated cytosol, ER and Golgi fractions; immunoprecipitated CFTR from these fractions; and studied whether Hop, co-immunoprecipitated with CFTR, was S-nitrosylated (using a biotin substitution technique followed by streptavadin isolation). We found that SNO-modified Hop coimmunoprecipitated with ∆F508 CFTR in the cytosol and ER at baseline; and that GSNO decreased the association between SNO-modified Hop and ∆F508 CFTR. We speculate that the effect of GSNO to decrease Hop-CFTR association may 1) result from increased Hop S-nitrosylation; and 2) be permissive for ∆F508 CFTR maturation. These data may be of relevance to the development of NO donor-based therapies for CF. Supported by CF Foundation and the NIH. Mouse models for cystic fibrosis (CF) provide new possibilities to study disease pathogenesis, to correlate genotype and phenotype, and to test novel CF therapies. We use the Cftr tm1Eu mice homozygous for the F508del mutation in a congenic FVB background as a tool in pre-clinical testing of therapeutic strategies aimed at the correction of the F508del folding and trafficking defect. These mice show characteristic CF ion transport abnormalities such as reduced cAMP-and cGMP activated intestinal chloride secretion and an increase of the amiloride-sensitive sodium absorption in nasal epithelium.

We developed an ex vivo assay for the rescue studies of the F508del trafficking defect in murine intestine by proteasome inhibitors based on periodical measurements of cAMP-activated transepithlial chloride currents in the Ussing chamber. Muscle-stripped ileum was incubated at 37 0 C in William's E Glutamax medium supplemented with insulin (10µg/mL) and dexamethasone (20µg/mL), in the presence of vehicle or proteasome inhibitor. A partial gain of chloride secretory function was achieved by incubating the tissue for 2-4h at 37 0 C in the presence of the proteasome inhibitors (PI) epoxomicin (1µM), MG132 (10µM), NIP-(Leu)3-vinyl and AdaLys(bio)Ahx3L3VS (both 2µM) and bortezomib (3nM). Exposure of the tissue for 6h at 37 0 C to ALLN (20µM) revealed a functional recovery up to WT levels. Immunohistochemical and Western blot analysis of all PItreated tissues showed the appearance of Cftr positive crypt cells and mature band C Cftr protein. Rescue by ALLN remained detectable following inhibition of the ER to Golgi transport by brefeldin A. Our ex vivo experiments demonstrate that proteasome inhibitors can enhance the amount of mature, functionally active murine F508 Cftr mutant protein in intact tissue by preventing proteolytic degradation and increasing its expression at the cell surface.

In vivo treatment of homozygous F508del mice with bortezomib at a dose of 0.1mg/kg/day for 3 days resulted in a high morbidity. Administration of a lower dose (0.04mg/kg/day ) was well tolerated and no mortality or significant weight loss could be observed. As outcome parameters for the in vivo treatment with PIs we used a salivary secretion assay and bioelectric measurements of intestinal and nasal epithelium in the Ussing chamber. There was no difference in isoproterenol-induced salivary gland secretion or and in Cftr-mediated intestinal chloride currents between control -and bortezomib-treated animals. However we found a reduction (1.7 fold) of the amiloride-sensitive short circuit current (Isc) in the nasal epithelium of bortezomib-treated mice, indicating a moderate correction of the ENACmediated sodium hyper absorption, that was not paralleled by a gain of the forskolin-mediated transepithelial chloride secretion.

Current in vivo experiments focus on the effects of PIs upon short term (2h-6h) rather than chronic treatment, to allow a direct comparison with the ex vivo treatment data.

Supported by grants from the CF Foundation, Bethesda, the Sophia Foundation and the Dutch Stomach-Liver-Intestine Foundation. Rotterdam, Netherlands; 2. Physiology and Biophysics, University of Alabama Birmingham, Birmingham, AL, USA; 3. Physiology, School of Medical Sciences, University of Bristol, Bristol, United Kingdom Recent progress in the development of small molecule correctors and potentiators capable of restoring CFTR function have increased the need for pre-clinical test models including cultured airway epithelial cells from human CF patients as well as CF mouse models. The validity of CF mice as a surrogate for human CF patients depends on the assumption that human and mouse CFTR, despite their limited sequence homology (78%) and differences in open probability (P o ) and channel subconductance state, behave similarly in their response to pharmacological correctors and potentiators.

To verify this assumption, we used the 125 Iodide efflux technique to compare the CFTR chloride channel activity in (i) CHO cells stably transfected with mouse F508del-CFTR, (ii) NIH-3T3 cells stably expressing human F508del-CFTR, and (iii) their WT-counterparts. Forskolin (10µM)-stimulated iodide efflux was very low in both mouse and human F508del CFTR cells (0.5 respectively 2 % of WT values) and was stimulated to a similar extent (3.5 and 3.8fold) by the CFTR potentiator genistein (100 µM). Preincubation for 14 h at 26 0 C, or at 37 0 C in the presence of the Vertex corrector CFcor-325 (3 µM) caused a 3-4 fold increase in forskolin-activated 125 I-efflux in both cell types that was further enhanced by 2.6-3 fold upon addition of genistein. Functional correction was paralleled by the appearance of mature, complex-glycosylated mouse or human CFTR on Western blots. Surprisingly, two other, structurally unrelated compounds known to correct the gating defect in hCFTR, i.e. the Vertex potentiator CFpot-532 (10 µM) and the uncharged NPPB analogue NPPB-AM (20 µM; cf. Wang W et al. 2005 ; J Biol Chem 280: 23622) mimicked the effect of genistein on hF508del-CFTR but failed to potentiate mF508del-Cftr in the 3T3 cells. Both CFpot-532 and NPPB-AM likewise failed to potentiate forskolin-stimulated, CFTR-mediated transepithelial chloride currents ex vivo in muscle-stripped ileal mucosa from homozygous F508del-Cftr tm1Eu mice (congenic FVB) mounted in Ussing chambers under conditions in which genistein (100µM) enhanced the current response by 2-3 fold. In contrast, preincubation of the tissue in William's E Glutamax medium for 14h at 26 0 C, or for 5 h at 37 0 C in the presence of CFcor-325 (10µM) enhanced the forskolin+genisteininduced transepithelial Clsecretion up to 70-100% of the response in Cftr+/+ littermates, and was associated with the appearance of band C on Western blots.

These results indicate that mouse and human F508del-CFTR (1) respond to low temperature incubation or to a small molecule corrector with a similar gain in surface expression and function, supporting the use of CF mouse models for in vivo tests of CFTR correctors; (2) respond differently to various classes of CFTR potentiators, in line with their pronounced differences in gating behaviour, and emphasizing the specificity of CFTR-potentiator interaction. Further studies focussing on human/mouse CFTR chimera and aimed to identify the domain(s) accounting for the differential response to the CFTR potentiators are in progress. cell apoptosis is a highly regulated process which is central to the maintenance of pulmonary vascular homeostasis. Recent investigations implicated CFTR in the regulation of intracellular sphingolipid levels. Our laboratory demonstrated that an imbalance in endothelial sphingolipid levels triggers apoptotic cell death. We therefore hypothesized that a dysregulated CFTR function in human lung endothelial cells inhibits programmed cell death in response to stress.

Methods: Patch clamp analysis was used to verify the expression of CFTR on human lung endothelial cells and the effect of specific inhibitors. We inhibited CFTR with DPC, CFTR inh -172, and NPPB or specific siRNA followed by treatment with prototypical stressors that induce endothelial cell apoptosis, staurosporine, H 2 O 2 , or TNF-α. Apoptosis was quantified by caspase-3 activity or annexin/ propidium iodide staining. Lipid extraction was performed by a modified Bligh and Dyer method and intracellular sphingolipids were measured by tandem mass spectrometry.

Results: Patch clamp analysis confirmed functional CFTR expression on endothelial cells and a complete inhibitory effect of DPC and NPPB but no effect of DIDS. As expected, treatment with staurosporine, a general protein kinase inhibitor and H 2 O 2 , as an oxidative stress stimulus resulted in significant endothelial cell apoptosis. Similarly, TNF-α treatment (20 ng/ml) in combination with cycloheximide caused activation of caspase-3 associated with a significant upregulation of the pro-apoptotic sphingolipid ceramide. Knock-down with specific CFTR siRNA was validated with Western blot and whole cell patch clamp analysis. DPC, NPPB, and CFTR inh -172 consistently and significantly inhibited apoptosis in human lung and primary mouse lung endothelial cells. Caspase-3 activity was decreased by 38-95% (% decrease in Caspase units/µg/min) in HLMVEC treated with H 2 O 2 after pretreatment with DPC (90%), CFTR inh -172 (95%), and NPPB (38%). However, we observed that DIDS, a non-specific chloride channel inhibitor also inhibited staurosporine induced caspase-3 activation (93%), suggesting the possible involvement of other chloride channels.

In conclusion, our data suggest a role of CFTR in modulating endothelial cell apoptosis in response to various stresses. The absence of functional CFTR impaired stress-induced apoptosis, a process which may be mediated via alterations in sphingolipid trafficking. This abnormal CFTR function in CF leading to aberrant cellular responses to stress may perpetuate an activated, pro-inflammatory phenotype of the lung vasculature. In turn, aberrant vascular activation may have a significant impact to the abnormal airway remodeling and persistence of a chronic inflammatory state characteristic of CF.

Supported by: Cystic Fibrosis Foundation clinical fellowship training grant

Hegedus, T. 1, 2 ; Serohijos, A.W. 1, 3 ; Dokholyan, N.V. 1 ; He, L. 1,2 ; Riordan, J.R. 1, 2 1. Department of Biochemistry and Biophysics, UNC at Chapel Hill, Chapel Hill, NC, USA; 2. Cystic Fibrosis Treatment and Research Center, UNC at Chapel Hill, Chapel Hill, NC, USA; 3. Department of Physics and Astronomy, UNC at Chapel Hill, Chapel Hill, NC, USA CFTR (Cystic Fibrosis Transmembrane Conductance Regulator) is a cAMP dependent chloride channel. The phosphorylation of its 200 amino acid R domain by protein kinase A (PKA) is obligatory for channel gating under normal conditions. R domain contains multiple PKA phosphorylation sites, which participate in activation, but no specific site is essential for channel function. In spite of numerous studies of the role of R domain in CFTR regulation, the mechanism is largely unknown. One reason for this is the lack of information on R domain structure and its interactions with other parts of CFTR. It has been shown that R domain lacks well-defined secondary structural elements and is intrinsically disordered. In this study, we used computational methods to explore the structure and function of the disordered R domain. Our results show a specific disorder-order pattern in R domain, which is conserved among species even though the primary structure is not conserved. This result implies a role of the disorder-order pattern in R domain function and provides a basis of a novel working model. To gain some insight into the three dimensional structure of R domain, we per-formed ab initio folding of R-domain using discrete molecular dynamics with an all-atom force field. As disordered domains can not be characterized with a single structure, an ensemble was generated by selection of structures with the lowest energy from a large number of folding simulations. These structures were clustered to determine the dominant R domain conformations. They reveal features of secondary and tertiary structure, the positions of phosphorylation sites in 3D, and residue contacts within the R region that might be important for structure and function. To computationally confirm the mechanism proposed in our working model, an ensemble of phosphorylated R domain structures was also produced and characterized.

Supported by NIH (R01DK051619), CFF (DOKHOL07I0), and Novartis (06123815).

physiology system that utilises the planar patch clamp technique (Finkel et al 2006) . This format utilises perforated patch clamp recording of 384 wells in parallel, permitting the generation of up to 4000 data points per day. We have employed this system to develop assays to detect modulators of mutant CFTR.

Human CFTR-G551D and dF508, over expressed in BHK and CHO cells respectively, were cultured to approximately 80% confluence before harvesting. dF508 cells were temperature corrected at 26°C for 72 hours prior to experimentation to facilitate channel trafficking to the plasma membrane. Cells were placed on the Quattro system and whole cell chloride currents measured using asymmetric chloride selective solutions such that the chloride reversal potential was significantly different from that of leak currents (0mV). Currents were assessed using a voltage ramp from -100 to +100mV applied prior and 10 min subsequent to compound addition. To validate the assays we used the prototypical CFTR potentiator, genistein, which was added simultaneously to cells under study with 10 µM forskolin. After assessing the effects of genistein, the specific CFTR inhibitor GlyH101, was added at 30 µM to confirm the currents were mediated via activation of CFTR. The EC 50 values determined for genistein were 3.8 ± 0.7 µM (mean ± S.E.M, n=7) for dF508 and 16.3 ± 2.6 µM (mean ± S.E.M, n=4) for G551D. These values agree with literature values (Bulteau-Pignoux et al 2002 ) . Two alternative potentiators, Phenylglycine-01 (PG-01) and VRT-532 were also tested in the presence of 10 µM forskolin and were active against both dF508 and G551D. The EC 50 values for PG-01 were 1.1 ± 0.1µM (mean ± S.E.M, n=10) for dF508 and 3.2 ± 0.5µM (mean ± S.E.M, n=8) for G551D. The EC 50 values for VRT-532 were 2.8 ± 1.2µM (mean ± S.D, n=3) for dF508 and 20.8 µM (mean, n=2) for G551D. These values agree with published data from ion transport studies (Pedemonte et al 2005; Van Goor et al 2006) .

In summary the IonWorks Quattro can be used to identify compounds which modulate the activity of CFTR. This assay has a number of advantages over other formats in that it is direct, sensitive and employs voltage control of the cells under study. These assays can be used for primary screening of small to medium compound sets or as a secondary screen to triage the output from higher throughput indirect formats such as membrane potential.

References It is very well agreed that ∆F508 CFTR is recognized as a misfolded protein by the ER quality control and targeted for degradation by the proteasome. During the course of experiments for AAV gene therapy for CF we created an AAV2/5 vector expressing a truncated CFTR insert (∆264) driven by a chicken beta actin promoter (rAAV-CB-∆264-CFTR). ∆264-CFTR contains the normal coding sequence of WT CFTR except that it is missing the first 264 amino acids.

We observed that Cos7 cells transiently transfected with rAAV-CB-∆264 CFTR express CFTR at significantly lower levels as compared to cells similarly transfected with a wt-CFTR construct. To examine why, cells were transfected with rAAV-CB-∆264-CFTR and treated for 16 hours with 5 µM of proteasome inhibitor, MG132. In the cells treated with MG132, ∆264-CFTR protein expression was significantly increased. Because lysosomal inhibitors did not have a significant effect, the data suggest that a large frac-tion of ∆264-CFTR protein is degraded in the proteasome. To evaluate the rate of degradation, cells were treated with cycloheximide. Time course experiments showed that ∆264 is more rapidly degraded than ∆F508-CFTR.

Importantly, in Cos7 cells cotransfected with ∆F508 and ∆264-CFTR, there was a significant increase in the mature C-band of ∆F508 CFTR compared to cells transfected with ∆F508 alone. This suggests that ∆264-CFTR enhanced the maturation of ∆F508 from B to C bands. Enhanced maturation of CFTR from B band to C band also occurred when ∆264-CFTR was transfected into HeLa cells stably expressing ∆508 CFTR (HeLa ∆508) and CFBE 41o-cells stably expressing ∆508 (CFBE 41o-∆508).

A number of proteins participate in the ER to determine the fate of CFTR. ∆F508-CFTR is known to associate with retrograde translocation protein complex on the ER membrane containing VCP (valosin-containing protein) and Derlin 1. VCP is also found in aggresomes containing ∆F508-CFTR in association with cytosolic histone deacetylase, HDAC6. We hypothesized that since ∆264 is more efficiently degraded as compared to ∆F508-CFTR it may have higher affinity with VCP protein complex. We verified that ∆264-CFTR interacts more readily with VCP and also has a higher degree of association with HDAC6 as compared to ∆F508-CFTR. Our data suggest that ∆264 is more readily degraded by the proteasome as compared to ∆F508-CFTR and lacks biosynthetic arrest in the ER. ∆264-CFTR binds more readily to VCP and associates avidly with HDAC6, two outcomes that are expected to enhance proteasomal degradation. By engaging key quality control proteins, ∆264-CFTR allows more ∆F508-CFTR to mature.

Our data suggest that truncated forms of CFTR (∆264) may be useful for CF gene therapy by affecting the maturation of endogenous CFTR. ∆F508, the most common disease causing CFTR mutation, has impaired conformational maturation during its biosynthesis, and thus is unable to exit the endoplasmic reticulum (ER). A well-defined di-acidic ER exit code has been identified within the first nucleotide binding domain (NBD1) of wildtype CFTR (JCB 167: 65-74, 2004) , the disruption of which prevents CFTR from efficiently exiting the ER. This is consistent with the role of the diacidic code in the COPII-mediated cargo concentration process. However, it is not known whether other types of ER-to-Golgi trafficking signals play a role in CFTR trafficking in the early secretory pathway. Even less is known about the trafficking signal(s) that are responsible for the rescue of ∆F508 CFTR by a number of means. To achieve a better understanding of the trafficking signals dictating the export of CFTR from the ER, we constructed a series of CFTR mutants to probe the potential functional targeting signals within both wild-type and ∆F508 CFTR. It has been shown that NBD2 is not required for the export of wild-type CFTR from the ER. We further found that an internal deletion mutant lacking membrane spanning domain 1 (MSD1) and NBD1-R but retaining the N-terminal cytoplasmic region, MSD2 and NBD2 does not have sufficient signal to allow ER export, suggesting that the signal(s) that are responsible for CFTR export largely reside within the NBD1-R. Unlike ∆F508 CFTR, the di-acidic code mutants are not temperature-sensitive. Dissecting the di-acidic code revealed that both acidic residues contribute to ER export, and that simultaneous substitution of both leads to a much decreased export of CFTR. We found that the temperature-dependent export of ∆F508 relies on the presence of the di-acidic code within the NBD1, suggesting that this code has a functional role in the temperature-dependent rescue of ∆F508 CFTR. In our effort to further examine the role of the di-acidic code in the rescue of ∆F508 CFTR by second site mutations, we found that ∆F508 CFTR rescue by the simultaneous disruption of two of the arginine-framed tripeptides (AFTs) is also dependent upon the di-acidic code. Surprisingly, we found that while NBD2 is not necessary for the export of wild-type CFTR, it is required for the AFT-mediated rescue of ∆F508 CFTR, suggesting a role for cytoplasmic inter-domain interactions in the AFT-mediated rescue of ∆F508 CFTR. A detailed biochemical analysis of these mutants provides valuable insights into the roles of different CFTR "trafficking" motifs in the ER-to-Golgi transport of CFTR, the specific conformational defects of ∆F508 CFTR and the potential molecular mechanism underlying the rescue of ∆F508 CFTR. Supported by CFF.

Sonawane, N.D. 1 ; Zhao, D. 1 ; Galietta, L. 2 ; Zegarra-Moran, O. 2 ; Verkman, A. 1 1. Medicine, University of California San Francisco, San Francisco, CA, USA; 2. Genetica Molecolare, Istituto Gianina Gaslini, Genova, Italy CFTR inhibitors are predicted to prevent intestinal fluid secretion in enterotoxin-mediated secretory diarrheas such as cholera. CFTR inhibitors that are not absorbed across the intestinal wall are attractive for diarrhea therapy because they may provide safe oral therapy. We previously discovered low affinity glycine hydrazide (GlyH, IC50 ~5 µM) CFTR inhibitors that block CFTR at its external pore (Muanprasat et al., J. Gen. Physiol. 2004; 124:125-137) . In order to increase inhibitor potency and prevent washout during severe secretory diarrhea, we synthesized a series of glycocalyx interactive CFTR inhibitors containing a malonic acid hydrazide (MalH) CFTR pore blocking moiety linked to a lectin (Sonawane et al., Gastroenterology 2007; 132:1234 -1244 . Lectin conjugation improved CFTR inhibitory potency by ~100-fold (IC50 to 50 nM). High-affinity CFTR inhibition was abolished by MalH-lectin heat denaturation, protease digestion, or competition by mannose or unconjugated lectin. Fluorescently labeled MalH-lectin remained membrane-bound for >6 hours after washout, whereas washout occurred in a few minutes without the lectin. MalH-lectins blocked cholera toxin-induced intestinal fluid secretion in closed intestinal loops in mice with EC50 50-100 pmol, and greatly reduced mortality in a suckling mouse model of cholera. We recently synthesized mono-and divalent CFTR inhibitors consisting of MalH coupled via a disulfonic stilbene linker to flexible mono and bifunctional polyethyleneglycols (PEGs) of molecular size 0. 2, 0.75, 2, 3, 5, 10, 20, 40 and 100 kDa, with calculated solution lengths of 1-23 nm, with the larger size PEGs potentially spanning CFTR dimers or inducing their formation. IC50 for inhibition of CFTR chloride current was 10-15 µM for monovalent MalH-PEGs, but substantially lower and size-dependent for divalent MalH-PEGs, decreasing from 1.5 µM to 300 nM with increasing PEG size. The mechanisms responsible for the improved and size-dependent potency of divalent MalH-PEGs were studied by whole cell, single-channel patch-clamp and by functional analysis of multivalent MalH-conjugated dextrans and asymmetric divalent PEGs. Whole-cell experiments revealed reversible voltage-dependent block of CFTR currents, with outward (positive) currents being more strongly blocked. In outside-out patch-clamp experiments, inhibitors caused a reduction of the mean open time. For MalH-PEG20kDa-MalH (2 µM) , the mean open time decreased from 189 ± 44 to 97 ± 29 ms. The effect on currentvoltage relationship and channel kinetics are consistent with a mechanism involving occlusion of the CFTR pore from the extracellular side. Luminally added divalent MalH-PEGs blocked by > 90% cholera toxin-induced fluid secretion in mouse intestinal loops with IC50s of <10 pmol/loop, and greatly reduced mortality in a suckling mouse model of cholera. Nonabsorbable, multivalent CFTR inhibitor-macromolecule conjugates may be useful as anti-secretory agents in the treatment of enterotoxin-mediated diarrheas. Supported by CFF and NIH. Activation of the cystic fibrosis transmembrane conductance regulator (CFTR) Clchannel is primarily controlled by PKA-dependent phosphory-lation of the R domain. Once the R domain is phosphorylated, ATP binding and enzymatic activity at two nucleotide-binding domains (NBDs) open and close the CFTR Clchannel. We hypothesized that R domain phosphorylation regulates CFTR activity by modulating ATP interactions within two NBDs. To test this hypothesis, we studied the activity of wild-type CFTR and five variants with deletions of portions of the R domain between residues 708 and 835. To alter ATP-dependent channel gating, we tested three CFTR stimulators, pyrophosphate (PP i , 2 mM), 2'deoxy-ATP (2'dATP, 1mM) and CaATP (1mM) . Consistent with previous studies, PP i , 2'dATP and CaATP all increased current of wild-type CFTR about two-fold. Each stimulator produced a similar increase in the R domain variants. The similar effects of stimulators on the channel activities of wild-type CFTR and R domain variants suggest that the R domain does not have a major role in regulating ATP-dependent channel gating. The R domain deletions did not alter the single-channel current amplitude of the R domain variants. However, R domain variants missing the sequence between residues 784 and 835 markedly reduced the open state probability, suggesting that this region is required for normal gating. These data suggest that the R domain does not control CFTR activity by modulating ATP interactions with NBD binding sites. Instead, we speculate that the C-terminal part of R domain might participate in the channel-gating machinery downstream of ATP regulation.

Supported by the Cystic Fibrosis Foundation and Howard Hughes Medical Institute.

Haggie, P.M.; Verkman, A. Medicine and Physiology, CVRI, U.C.S.F., San Francisco, CA, USA It was recently reported that phagolysosomes of alveolar macrophages from CF mice acidify in a CFTR-dependent manner and that defective phagolysosomal acidification impairs bactericidal activity (Di et al., Nat. Cell Biol. 2006, 8:933-944) . These findings suggested a unifying hypothesis for CF disease progression: defective phagolysosomal acidification in CF macrophages permits the initiation and promotes progression of bacterial infection in the lungs. To reassess the central finding of that study we measured phagolysosomal pH using a fluorescent pH indicator containing Oregon Green® 488 (pKa ~4.7) and tetramethylrhodamine covalently bound to zymosan. Phagolysosomal pH was insensitive to CFTR inhibition (10 µM CFTRinh-172) in J774 macrophages (pH 5.08 ± 0.08 vs. 5.10 ± 0.03), alveolar macrophages from mouse lung (pH 5.34 ± 0.05 vs. 5.37 ± 0.07), and alveolar macrophages from human lung (pH 5.30 ± 0.08 vs. 5.35 ± 0.07). Phagolysosomal pH in alveolar macrophages from ∆F508 CF mice was not significantly different from that in alveolar macrophages from wild-type mice (pH 5.34 ± 0.05 vs. 5.40 ± 0.06). To account for their finding of defective phagolysosomal acidification in alveolar macrophages, Di et al. reported that lysosomal acidification was CFTR-dependent and that fusion of lysosomes to phagosomes (at ~20 min after phagocytic uptake) was responsible for phagosomal acidification. We measured lysosomal acidification in J774 macrophages using a dextran-conjugate containing Oregon Green® 488 and tetramethylrhodamine and found that acidification was not impaired by CFTR inhibition (pH 4.48 ± 0.03 vs. 4.42 ± 0.03) . We also measured the kinetics of phagosomal acidification using a zymosan-conjugate containing 5-(and-6)-carboxyfluorescein (pKa ~ 6.4) and tetramethylrhodamine. Phagosomal acidification in J774 macrophages and murine alveolar macrophages began within 3 min of phagocytosis and reached steady-state by 10-15 min, in agreement with prior data in murine peritoneal and bone marrow-derived macrophages. Acidification kinetics in J774 macrophages was not altered by CFTR inhibition, nor was acidification kinetics different in wild-type vs. CF alveolar macrophages. Our findings do not support the conclusion that phagolysosomal acidification in alveolar macrophages is CFTR-dependent, nor that it is impaired in CF. The mechanism of phagosomal acidification proposed by Di et al. is not in accord with our data or precedents in the literature. Because phagolysosomal acidification is central to the proposed mechanism linking defective CFTR chloride channel function with CF lung disease, our results do not support the involvement of CFTR in defective macrophage function in the pathogenesis of CF lung disease. Supported by CFF and NIH. pared to that of wild-type CFTR. The Hsp70 family of molecular chaperones play important roles in the protein quality control process within the ER. Hsp70 ATPase activity is regulated by multiple co-chaperones such as Hsp40, BAG-1 and HspBP1. Hsp105, a high-molecular-weight member of the Hsp70 family was recently shown to display nucleotide exchange activity for Hsp70s in vitro. Furthermore, Hsp105 was identified as a component of a CFTR-associated multiple protein complex using a global proteomic approach (Cell 127: 803-815, 2006) . In an attempt to explore the role of Hsp105 in CFTR conformational maturation in the ER, we over-expressed the co-chaperone in HEK293 cells and quantitatively analyzed its effect on the maturation and degradation of CFTR. Consistent with its role as a nucleotide exchange factor for Hsp70s, over-expression of Hsp105 inhibits the ER export of wild-type CFTR and promotes its degradation. However, in striking contrast, over-expressing Hsp105 stabilizes ∆F508 CFTR and promotes its ER export at reduced temperature. This effect is less pronounced at physiological temperature. The apparently opposite effects of Hsp105 on wild-type and ∆F508 CFTR maturation and quality control, suggests distinct conformational maturation pathways for the two CFTR molecules, and reveals a specific role for Hsp105 in regulating ∆F508 CFTR refolding. Such conclusion is reinforced by RNAi experiments and further supported by quantitative co-immunoprecipitation. RNAi-mediated down-regulation of Hsp105 expression destabilizes ∆F508 CFTR and reduces its export at reduced temperature, and Hsp105 shows more extensive association with the ER form of ∆F508 CFTR than that of wild-type CFTR. Further analysis revealed that Hsp105 not only modulates the chaperone activities of Hsp70s but also alters their steady state levels within the cell, creating secondary effects on CFTR maturation and quality control. Further studies are necessary to achieve a better understanding of the machinery, pathway and mechanism of ∆F508 CFTR conformational maturation at reduced as well as physiological temperatures, and this in turn will provide critical insights and key factors that are of potential value to the rescue of the trafficking defect of ∆F508 CFTR.

Farinha, C.M. 1, 2 ; Pissarra, L.S. 1 ; Amaral, M.D. 1, 2 1. Department of Chemistry and Biochemistry, Faculty of Sciences, University of Lisboa, Lisboa, Portugal; 2. Ctr Hum Genet, Nat Inst Health, Lisboa, Portugal The most frequent mutation in the cystic fibrosis (CF) gene, F508del, causes retention of its protein product, F508del-CF transmembrane conductance regulator (CFTR) in the endoplasmic reticulum (ER) as a core-glycosylated intermediate that is rapidly degraded. Therefore, most F508del-CFTR fails to traffic to the plasma membrane, where wild-type (wt) CFTR normally functions as a chloride (Cl-) channel. Retention, however, is not due to lack of function, since the mutant still retains some function if it reaches the membrane. Instead, it results from misfolding which is recognized by the ER quality control (ERQC) in which many intervenients, including molecular chaperones, participate. Identification of molecular partners involved in the disposal of F508del-CFTR to the proteasome is therefore crutial.

It was recently shown that casein kinase II (CK2), a pleiotropic constitutively active protein kinase involved in several processes, such as neoplasia, cell survival and viral infections, binds wt-CFTR near the F508 residue, phosphorylating its first nucleotide binding domain (NBD1) at S511 [1] . Interestingly, deletion of F508 abrogates this interaction, which is the first described F508del-dependent protein-protein interaction. Our aim here was to identify whether CK2 interaction affects the early steps of CFTR biogenesis, turnover and processing.

To this end, novel BHK cells were produced which stably express wt-or F508del-CFTR in which the consensus residue S511 was substituted by either a neutral (alanine -S511A) or an acidic residue (aspartate -S511D). Initial biochemical analyses of these cell lines revealed that: 1) CFTR proteins bearing D or A at position 511 are processed; 2) despite producing equivalent levels of CFTR transcripts, S511A expressing lines consistently show lower levels of CFTR protein.

Metabolic labelling and pulse-chase experiments followed by CFTR immunoprecipitation were also performed in these lines. After quantification of bands B (immature form) and C (mature form) of CFTR, these preliminary results show that substitution of S511 (by A or D) does not affect the turnover or processing of either wt-or F508del-CFTR.

The effect of CK2 inhibition on the turnover and processing of CFTR was also studied, by incubating cells with 20 µM of the CK2 inhibitor tetrabromobenzotriazole (TBB) for 90 min. Pulse-chase experiments under these conditions show that: 1) the steady-state levels of both wt-and F508del-CFTR are reduced; 2) the turnover of wt-CFTR (but not of F508del-CFTR) is increased; and 3) processing of wt-CFTR is decreased.

Such data are consistent with a putative stabilizing role for CK2 upon wt-CFTR [1] . However, in our preliminary results this effect does not appear to be dependent on residue 511 (nor on the putative charge added by the kinase on this residue), thus suggesting that this effect may be indirect. Further studies are underway to identify the mechanism by which CK2 affects the turnover and processing of CFTR.

1. Treharne et al (2007) 

Pissarra, L.S. 1 ; Xu, Z. 2 ; Farinha, C.M. 1, 3 ; Sheppard, D.N. 2 ; Amaral, M.D. 1, 3 1. Department of Chemistry and Biochemistry, Faculty of Sciences, University of Lisboa, Lisboa, Portugal; 2. Dep Physiology, Univ Bristol, School Med Sciences, Bristol, United Kingdom; 3. Ctr Hum Genet, Nat Inst Health, Lisboa, Portugal G550E and 4RK (the simultaneous mutation of four arginine-framed tripeptides (AFTs): R29K, R516K, R555K and R766K) are second site mutations that rescue the processing and function of F508del-CFTR [1, 2] . These revertant mutations rescue F508del-CFTR from retention within the endoplasmic reticulum by distinct mechanisms: G550E likely alters the conformation of the first nucleotide-binding domain (NBD1), whereas 4RK plausibly allows F508del-CFTR to escape ER retention/retrieval mediated by AFTs [3] . Both G550E and one of the AFTs (R555K) lie close to the residue G551, where a common CF-causing mutation G551D occurs, generating a correctly localized Clchannel with a severe gating defect.

Our aim here was to assess whether the revertants G550E and 4RK influence the folding, processing and gating behaviour of the G551D mutation, by employing biochemical and functional approaches.

To test this idea, we introduced the G551D mutation into CFTR cDNAs containing either G550E or 4RK in the pNUT vector and stably expressed these constructs in BHK cells. Preliminary results from biochemical studies indicate that the mature fully-glycosylated form of CFTR protein (band C) of both G550E-G551D-CFTR and 4RK-G551D-CFTR were present at similar levels to those of G551D-CFTR. Analysis of CFTR-mediated iodide efflux from these cells revealed that G550E is unable to rescue the functional defect of G551D. However, 4RK generated an efflux of iodide larger than that elicited by cells expressing G551D, albeit smaller than that of cells expressing wild-type (wt) CFTR. Consistent with these data neither 4RK nor G550E rescued the defective channel gating of G551D-CFTR. However, both revertants caused a small increase in G551D-CFTR activity by attenuating the prolonged interburst interval of G551D. We conclude that F508del and G551D disrupt CFTR channel gating by distinct mechanisms.

Altogether, our data suggest that at least when in cis with the G551D mutation, the AFTs (together or individually) might have a direct effect on CFTR channel gating. This raises the possibility that 4RK, in addition to its well-described effect on trafficking, may act on CFTR structure and/or folding, as previously suggested for R555K [4] .

Work supported by grant POCTI/SAU/MMO/58425/2004 and pluriannual funding of CIGMH (FCT, Portugal) and the UK CF Trust. L. Pissarra was a recipient of BD/9095/2002 doctoral fellowship (FCT, Portugal) . [1] Carvalho AC, Gansheroff LJ, Teem JL (2002) J Biol Chem 277, 35896-35905. [2] Chang XB, Cui L, Hou YX, Jensen TJ, Aleksandrov AA, Mengos A, Riordan JR (1999) Mol Cell 4, [137] [138] [139] [140] [141] [142] [3] Roxo-Rosa M, Xu Z, Schmidt A, Neto M, Cai Z, Soares CM, Sheppard DN, Amaral MD (2006) Proc Natl Acad Sci USA 103, 17891-17896. [4] Teem JL, Carson MR, Welsh MJ (1996) improves ∆F508-CFTR intracellular trafficking in CF epithelial cells, although the mechanism by which this occurs is not clear. To identify gene products with atlered abundance in response to 4PBA, we performed a differential display RT-PCR screen on RNA isolated from IB3-1 CF epithelial cells treated with 4PBA for 0-24 hours. We isolated a cDNA encoding StIP-1, a putative human STAT-3 (signal transducer and activator of transcription-3) interacting protein and confirmed that 4PBA causes transiently increased StIP-1 mRNA and protein abundance after 8 hours of exposure. StIP-1 was originally described as a scaffold protein that is required for ligand-dependent activation of STAT-3. StIP-1 is also identical to elongator protein 2 (Elp-2), a subunit of the multicomponent Elongator complex that stimulates RNA polymerase II activity. Interestingly, recent data suggests that Elongator may also regulate polarized secretion (Rahl, et al. (2005) , Mol. Cell, 17:841-853). We therefore hypothesized that StIP-1/Elp-2 would regulate ∆F508-CFTR intracellular trafficking. In IB3-1 cells, StIP-1/Elp-2 associates with ∆F508-CFTR in coimmunoprecipitation experiments after 4PBA treatment. In IB3-1 cells overexpressing StIP-1/Elp-2, immunofluorescence experiments suggested that ∆F508-CFTR trafficked to the plasma membrane even in the absence of 4PBA treatment. StIP-1/Elp2 co-localized with markers of the Golgi (58 kDa Golgi protein) and exocytic vesicles (SNAP-25) when overexpressed in these cells. In contrast, overexpression of a deletion mutant of StIP-1/Elp-2 lacking the WD domain, or STAT-3 binding region, blocks the improvement of ∆F508-CFTR trafficking in response to 4PBA in immunofluorescence and surface biotinylation experiments. This StIP-1/Elp-2 deletion mutant did not co-localize with Golgi or exocytic vesicle markers. Furthermore, immunofluorescence experiments also suggested that 4PBA causes colocalization of StIP-1/Elp-2 and Elp-1, another member of the Elongator complex that is often defective in Familial Dysautonomia. These data are consistent with StIP-1/Elp-2 regulating ∆F508-CFTR intracellular trafficking in response to 4PBA. This regulation may occur in the context of 4PBA stimulating assembly of Elongator, which in turn may promote trafficking of exocytic vesicles carrying ∆F508-CFTR to the plasma membrane.

Supported by grants from NIDDK.

fold [1, 2] . However, functional studies demonstrate significant differences in the gating behaviour of hCFTR and mCFTR [3] . A powerful approach to investigate CFTR structure and function is to examine interspecies differences and identify regions of conservation and divergence. To understand the structural basis for the functional differences between hCFTR and mCFTR, we generated hmCFTR chimeras containing mCFTR domains on an hCFTR backbone.

For this purpose, we replaced all or part of NBD1, NBD2 or the Rdomain of hCFTR with the equivalent regions of mCFTR and investigated their biochemical and functional properties. The in vivo folding of these hmCFTR chimeric proteins was indirectly evaluated from their maturation status, after their stable expression in novel BHK cell lines. Like wt-hCFTR, most chimeric proteins were processed within the cell. However, two chimeras failed to mature: clone 12b (mNBD1, amino acid (aa) residues 518-585) and clone 114c (mNBD2, aa 1260-1412). We compared the murine sequence of these two chimeras with that of hCFTR to determine the respective physico-chemical distances (PCDs) of their aa changes. Changes with higher PCD values were selected and in vitro mutagenesis performed to introduce these aa alterations into hCFTR cDNA. For clone 12b the selected changes were: E527Q, E528Q (PCD = 29); S531T (PCD = 58); K536Q (PCD = 53), I539T (PCD = 89) and K584E (PCD = 56). For clone 114c the changes were: P1290T (PCD = 38), K1302Q (PCD = 53), Y1307N (PCD = 143), S1311K (PCD = 121), C1344Y (PCD = 194), D1394G (PCD = 94) and E1409D (PCD = 45). Biochemical analyses of these mutants stably expressed in BHK cells revealed that for clone 12b, E527Q, S531T, K536Q and I539T were processed, whereas K584E was not. For clone 114c, P1290T, K1302Q, Y1307N, D1394G and E1409D were processed. Iodide efflux showed that S531T, K536Q, I539T, E1409D and D1394G were functional, but not K584E.

We consider that with additional functional analyses, our approach will identify critical residues responsible for conformational changes and hence, functional differences between hCFTR and mCFTR.

Work supported by the BBSRC and grant POCTI/MGI/47382/2002 (FCT, Portugal) and pluriannual funding of CIGMH (FCT, Portugal) . AC DaPaula is a recipient of PhD fellowship SFRH/ BD/17475/2004 (FCT, Portugal We recently reported that the UPR decreases endogenous wild-type (WT) CFTR expression. As a follow up of these studies, we investigated the role of the folding deficient, ∆F508 CFTR on UPR induction and identified a mechanism by which the UPR decreases CFTR expression. For these studies, we developed a cell line expressing recombinant ∆F508 CFTR on the endogenous WT background (Calu-3 ∆F) and established individual clones with different ratios of endogenous (WT) to recombinant (∆F508 CFTR) expression. Two clones which express 1:1 (Calu-3 ∆FC3) and 1:8 (Calu-3 ∆FC5) ratios of WT to ∆F508 CFTR and CFPAC-1 cells expressing endogenous ∆F508 CFTR were tested as models. The UPR was constitutively active in Calu-3 ∆FC5 cells only. In Calu-3 ∆FC3 induction of ∆F508 CFTR expression resulted in UPR activation, indicating that high expression of ∆F508 CFTR is required for UPR induction. Furthermore, following pharmacological induction of the UPR, endogenous CFTR mRNA decreased to undetectable amounts both in Calu-3 ∆F and CFPAC-1 cells. The decrease in CFTR mRNA levels was not the result of shortened mRNA half-life. In contrast, using a human CFTR promoter driven reporter (pCFTR-Pluc), we demonstrate suppression of the CFTR promoter when the UPR is activated. Considering that correction of ∆F508 CFTR is the main therapeutic approach for CF, it is important that ∆F508 CFTR expressed at low levels, such as in vivo, does not activate the UPR. However, our results also reveal that ∆F508 CFTR correctors have to be tested for UPR activation, since transcriptional inhibition in this setting may contribute to inefficient rescue in native cells or in vivo. Genistein, a naturally occurring tryrosine kinase inhibitor, at low concentrations (50µM) stimulates Wt -CFTR-mediated chloride secretion in a variety of cell and tissue types by an adenosine 3',5'-cyclic monophosphate (cAMP)-independent pathway. Genistein has been shown to potentiate the channel function of ∆F508-CFTR and G551D-CFTR channels by prolonging channel open time (Hwanget al., Am J Physiol Cell Physiol 273: C988-C998, 1997; Illeck et al. Am J Physiol Cell Physiol 277: C833-C839, 1999) Patch clamp studies have provided indirect evidence that genistein may bind directly to CFTR and potentiate channel open probability by modifying the function of the nucleotide binding domains (Weinreich Et al., Pflügers Arch 434: 484-491, 1997) . The goal of the present study was to determine whether genistein activated purified and reconstituted CFTR protein as an initial step toward defining its molecular mechanism of action.

The effect of genistein on purified and reconstituted Wt-CFTR was first studied following its reconstitution into planar lipid bilayers. In the presence of 100µM ATP, phosphorylated CFTR exhibited a low open probability (Po= 0.12). The addition of 50µM genistein, caused a significant increase to the open probability of CFTR (Po=0.76). Using an electrogenic flux assay on a population of purified and phosphorylated CFTR molecules, we determined that 50 µM genistein caused a potentiation (70% increase) of chloride channel activity in the presence of 100 µM ATP. These results suggest that this macroscopic assay of purified CFTR function is sufficiently sensitive to detect the effect of this potentiator. Currently, the consequences of genistein on the channel activity and ATPase activity of purified reconstituted ∆F508-CFTR and G551D-CFTR are being assessed. Preliminary results suggest that in the presence of low ATP concentrations (10-100 µM),50µM genistein inhibits the ATPase activity of ∆F508-CFTR, suggesting that it interacts directly with the nucleotide binding domains to alter their function.

Previous studies indicated an acute coordination between the activities of cystic fibrosis transmembrane conductance regulator (CFTR) Chloride channel and the amiloride sensitive epithelial sodium channel (ENaC) so that both these channels are either activated or deactivated in a synchronous fashion in the human sweat duct (Reddy et.al, 1999) . However, the mechanisms responsible for such cooperativity between these ion channels are unknown. Previous studies indicated that: cytoplasmic pH controls CFTR activity through effects on phosphorylation (Reddy et.al. 1998) . At pH 6.0, CFTR chloride conductance was reduced to 10±5 mS/cm 2 , but incresed to 42±7 mS/cm 2 at pH 8.0 (n=number of ducts=5). ENaC channel activity is also a function of cytosolic pH in heterologous expression systems (Chalfant et.al, 1999) . The objective of this study was to test the hypothesis that the cytosolic pH may mediate the cooperative effects that occur between CFTR and ENaC. We used basolaterally a-toxin permeabilized apical membrane preparations of native human sweat duct which expresses ENaC and CFTR robustly as an experimental system. We showed that while luminal pH had no effect, changes in cytosolic pH acutely affected ENaC activity. That is, acidic pH inhibited, while basic pH activated ENaC activity. Alkalinizing cytosolic pH from 6.8 to 8.5 increased ENaC conductance (gENaC) by 35±1.8 mS/cm 2 (n=6). pH regulation of ENaC activity appears to be independent of CFTR and endogenous kinase activities because basic pH stimulated ENaC: a.) after deactivating CFTR by removing cAMP and ATP in normal ducts, b.) in the absence of CFTR in CF ducts, and c.) after blocking endogenous kinase activity with the non-specific kinase inhibitor, staurosporine. Na + /H + exchanger (NHE) may mediate changes in cytosolic pH as a function of intracellular Na + activity. NHE mRNA is expressed in the sweat duct and cytosolic pH responds to changes in Na + gradients across the basolateral membrane. Conceivably, when transport conditions are favorable and intracellular Na + is low, alkaline pH would allow ENaC and CFTR to cooperatively admit Na + and Clthrough the apical membrane. When conditions are unfavorable and intracellular Na + is excessive, acid pH would limit Na + and Clentry to protect the cell from disruptive changes in cell volume. Thus, changes in cytosolic pH may play a crucial role in coordinating the activities of ENaC and CFTR during transepithelial salt transport.

Acknowledgements: Kirk Taylor and Sucharitha Madireddi for expert technical assistance. Funded by NIH-RO1 DE14352, NIH-RO1HL08042 and the Nancy Olmsted Trust. The cystic fibrosis transmembrane conductance regulator (CFTR) is an anion channel that is normally expressed in the plasma membrane but is mislocalized in CF. To monitor trafficking we inserted super ecliptic pHluorin, a pH-sensitive variant of green fluorescent protein (GFP) (Miesenbock G, De Angelis DA, Rothman JE (1998) Nature 394, 192-5) into the fourth extracellular loop of CFTR. This construct, which we call CFTR-pHluor, has weak fluorescence intracellularly that increases ~10-fold when it is inserted into the plasma membrane and the pHlour becomes exposed to more alkaline extracellular pH . Thus CFTR-pHluor should be less fluorescent than CFTR with the normal GFP fused to the N or C termini when situated in intracellular vesicles. Despite the insertion of an entire GFP in the middle of the channel protein, CFTR-pHluor trafficked to the membrane and patch clamp studies revealed little change in unitary conductance or open probability compared to wild-type CFTR. CFTR-pHluor enabled visualization of vesicle insertion events at the plasma membrane by total internal reflection fluorescence (TIRF) microscopy. Faintly fluorescent vesicles containing CFTR were observed under the plasma membrane and eventually fusing with it to generate a burst of fluorescence. pHluorin inserted at the same position in ∆F508-CFTR, the predominant mutant in CF, was retained in the ER where the pHluor remained partially quenched. When the folding defect was corrected by culturing cells at 29°C or treating them with known correctors, ∆F508-pHluor trafficking was partially restored and total cell fluorescence increased. This enabled a quantitative analysis of ∆F508-CFTR correction based on the increase in fluorescence in individual cells by flow cytometry. CFTR-pHluor may be most useful for microscope-based high-content screening studies of CFTR and ∆F508-CFTR trafficking. Support: the Breathe program, Canadian CF Foundation, Canadian Institutes of Health Research and CF Foundation Therapeutics (USA).

Tukaye, D.N.; Guggino, W.B. Physiology, Johns Hopkins University, Baltimore, MD, USA Type III Secretion System (T3SS) toxins of P. aeruginosa are important virulence factors in P. aeruginosa infections. One of the T3SS toxins, ExoS, has been shown to facilitate uptake and invasion of P. aeruginosa at airway surface epithelium. In mouse models of Pseudomonas pneumonia, infection with P. aeruginosa ExoS+ strains caused increased levels of fluid in lungs as determined at autopsy in contrast to infection with P. aeruginosa ExoSstrains. The exact molecular mechanism underlying these observations is not known. ExoS is a bifunctional protein with GTPase Activating Protein (GAP) activity at the N terminal domain and ADP ribosyl transferase (ADPRT) activity at the C terminal domain. We found that ExoS-GAP activity increases total cellular levels of mature (C band) Wild Type (WT) CFTR in CFBE41o-cells as measured by western blot analysis. This increase was mediated by decreased delivery of WT CFTR for lysosomal degradation. In contrast, ExoS-GAP failed to increase total levels of ∆508 CFTR (B band). Interestingly, ExoS-GAP increased total levels of ∆508 CFTR, bands C and B, following rescue with 4-phenylbutyrate. This indicates that targets of ExoS-GAP exist in CFTR trafficking beyond the ER degradation pathway. ExoS-GAP also brought about a corresponding increase in surface levels of mature WT CFTR as measured by surface biotinylation. In conclusion, we have shown that P. aeruginosa T3SS ExoS-GAP, upregulates total and surface levels of mature WT CFTR by modulating CFTR trafficking beyond ER, most likely by decreasing lysosomal degradation. The decrease in lysosomal degradation could be occurring in part by inactivation of small molecular weight G-proteins involved in delivery of WT CFTR to lysosomes. Increases in total and hence surface levels of CFTR could in part explain increased amounts of fluids seen in P. aeruginosa pneumonia mouse model treated with ExoS+ strains.

The post-maturational trafficking and localization of CFTR is regulated by a wide variety of proteins. Among the most prominent are several PDZ domain proteins that bind to the C-terminal residues of CFTR. In particular, two of these proteins, NHERF1 and CAL, have been shown to mediate opposing effects on the apical membrane levels of the disease-associated ∆F508-CFTR mutant, by controlling the balance between endocytic recycling and lysosomal degradation. In particular, whereas overexpression of NHERF1 rescues ∆F508-CFTR at the cell surface, overexpression of CAL has the opposite effect on CFTR. CAL-mediated downregulation of CFTR requires a functional PDZ binding site. We have now shown that suppression of endogenous CAL cooperates with temperature rescue to stabilize functional ∆F508-CFTR channels at the apical membrane in polarized bronchial epithelial cells. This work supports the hypothesis that the CAL PDZ binding site is a therapeutic target for treatment of cystic fibrosis.

A more direct test of this hypothesis will require competitive inhibitors that can efficiently block the CAL:CFTR interaction. In addition, given the large-number of protein-protein interactions formed by both the CAL PDZ domain and the CFTR C-terminus, these inhibitors should ideally exhibit bidirectional selectivity, neither disrupting the favorable NHERF1:CFTR interaction, nor strongly displacing other membrane proteins that interact with CAL. Using a fluorescence polarization assay and peptide-array technology, we have now identified a high-affinity, CAL-selective peptide inhibitor. Compared to CFTR, this sequence binds CAL more tightly, but NHERF1 more weakly. Furthermore, we have shown that the CAL:CFTR interaction is the weakest among the known CAL-binding proteins, and that CFTR should thus be susceptible to selective displacement from the CAL binding site. Finally, we have designed cell-permeable peptides that allow us to test the hypotheses (a) that such compounds will specifically disrupt the CAL:CFTR interaction and (b) that such targeted stereochemical inhibition will stabilize the functional cell-surface expression of ∆F508-CFTR in airway epithelial cells.

The pathways for the endoplasmic reticulum associated protein degradation (ERAD) of misfolded proteins in the mammalian cells are incompletely understood. We investigated the role of molecular chaperones in ERAD for the cystic fibrosis transmembrane conductance regulator (CFTR) and its common folding mutant, ∆F508CFTR. We found that Hsp70 and its cochaperone Hdj-2 interacted significantly with ∆F508CFTR and wt-CFTR in airway epithelial cells. Hsp70 interacted with the immature form of CFTR while Hdj-2 associated with both immature form and ubiquitylated CFTR.

Structure-function studies showed that Hdj-2 recognized ubiquitylated CFTR via its Zn-binding domain. Immunoprecipitation and GST-fusion protein pulldown experiments revealed that Hdj-2 interacted with ubiquitin. In steady-state, Over-expression of Hdj-2 elicited increasing both immature and mature forms of wt-CFTR, but it resulted in decreasing immature form of ∆F508CFTR. Pulse-chase studies showed that co-expression of Hdj-2 promoted ∆F508CFTR degradation, reducing its half-life from 85 to 50 min. In contrast, Hdj-2 expression did not significantly affect wt-CFTR biogenesis. These data are consistent with our finding that Hdj-2 shows a selective physical interaction with ∆F508CFTR, which translates into a functional discrimination between the mutant and its wt counterpart. The HPD motif within the J-domain of Hdj-2, which is necessary for the co-chaperone to bind Hsp70 and stimulate its ATPase activity, was required for Hdj-2 mediated ∆F508CFTR degradation. Mutation of the HPD motif retarded the degradation of ∆F508CFTR and increased the steady-state levels of ∆F508CFTR 3-fold. A role for Hsp70 in Hdj-2 mediated ∆F508CFTR degradation was tested in which Hsp70 knockdown increased ∆F508CFTR expression 3-4 fold. Hjd-2 regulates ∆F508CFTR maturation and knockdown of endogenous Hdj-2 promoted ∆F508CFTR and its cAMP-dependent anion conductance in airway epithelial cells. In contrast, we were unable to detect maturation of ∆F508CFTR in Hsp70 knockdown experiments, indicating that the maturation of ∆F508CFTR mediated by reduced Hdj-2 expression may be independent of Hsp70 function. Taken together, our data demonstrate that Hdj-2 is a molecular sensor that can detect differences in the folding related to the ∆F508CFTR maturation. These data also highlight a novel pathway for Hdj-2 linked, ubiquitin-dependent degradation of ∆F508CFTR. Reduction of Hdj-2 expression or its association with ∆F508CFTR promotes maturation of this mutant and therefore represents a potential therapeutic approach for cystic fibrosis. [Supported by NIH and CF Foundation] How the loss of CFTR function results in cholesterol accumulation within the cell is currently unknown. CFTR activation is driven by the cAMP and we propose that CF cells respond to the loss of CFTR activity by increasing the cAMP pathway in order to increase CFTR expression. To test this hypothesis, epithelial cells were treated with the phosphodiesterase-3 (PDE3) and PDE4 inhibitors milrinone and rolipram, respectively. Inhibition of PDE4 function with rolipram in wildtype cells leads to peri-nuclear free cholesterol accumulation identical to what is observed in CF epithelial cells as viewed by filipin staining, strongly implicating a cAMP-dependent pathway in the regulation of cholesterol trafficking. The PDE3-selective inhibitor milrinone had no effect on cholesterol trafficking suggesting specificity for the PDE4 pathway. Our preliminary data support this hypothesis in that both CF-model 9HTEo-pCEPR cells and mouse nasal epithelium (MNE) from Cftr -/-mice exhibit reduced protein expression of the cAMPspecific phosphodiesterases PDE4 compared to wt controls. The proposed consequence of a chronic amplification in cAMP signaling is altered cholesterol transport regulation. To address this potential role of PDE4 in cholesterol trafficking, free cholesterol was visualized using filipin staining in wild type cells that were treated with rolipram, a PDE4-specific inhibitor. Conversely, treatment of CF-model pCEPR cells with the PKA inhibitor Rp-cAMPs significantly improves cholesterol processing, further pointing to the involvement of the cAMP pathway. We propose that the cAMP pathway influences cholesterol processing through the regulation of β-arrestin-2 (βarr2) according to the premise that chronic increase in the cAMP pathway would initiate an elevation in βarr2 expression. βarr2 is an important regulator of adrenergic receptor recycling and organelle trafficking. Because βarr2 is pivotal in regulating endocytotic recycling pathways, it could also impact cholesterol processing. This predicted increase in βarr2 protein expression is observed in both CF-model pCEPR cells and Cftr -/-MNE compared to respective controls. Over expression of βarr2 in wt epithelial cells leads to CF-like peri-nuclear cholesterol accumulation further implicating a role for βarr2 in the development of this phenotype. Altered cholesterol trafficking in CFTR would lead one to expect different βarr2 localization throughout the cell. Further understanding of the implications of altered cAMP signaling and its relationship to aberrant cholesterol accumu-Rationale: We recently reported that VCP (valosin containing protein) physically interacts with gp78/AMFR (autocrine motility factor receptor) to couple retrograde translocation of ∆F508-CFTR to proteasomal degradation (Vij et al. JBC 2006) . Recent studies have revealed an alternative system to the proteasome for degradation of polyubiquitinated misfolded proteins termed the aggresome. Histone deacetylase-6 (HDAC6) is a unique cytoplasmic deacetylase capable of interacting with ubiquitin and mediating the accumulation of ∆F508-CFTR in aggresome bodies.

Hypothesis: VCP competes with HDAC6 to dissociate ∆F508-CFTR perinuclear aggregates.

Methods: IB3-1 (∆F508/W1282X) CF bronchial epithelial cells were transiently transfected with ∆F508-CFTR and VCP-GFP constructs and treated with a HDAC6 inhibitor (Tubacin), proteasome inhibitor (PS-341 or MG-132), lysosome inhibitor (baflomycin A1) or Nil-Tubacin (control) for 4, 8 or 12 hrs. The effect of these inhibitors on ∆F508-CFTR and VCP:∆F508-CFTR interactions were quantified by immunoprecipitating these protein complexes followed by immunoblotting with VCP, HDAC6 or Clathrin antibody. ∆F508-CFTR levels were measured by metabolic labeling using TRAN-35S-label (250 µCi/ml) for a 30 min pulse and 2 hrs chase. The subsequent effects of these inhibitors on VCP localization was detected by fluorescence microscopy of GFP moiety.

Results: The co-immunoprecipitation experiments showed that HDAC6inhibition by 10µM Tubacin (12 hrs) promotes VCP:∆F508-CFTR and prevents HDAC6:∆F508-CFTR interaction. The proteasome inhibition (MG132 10µM) resulted in maximal HDAC6:∆F508-CFTR at 4hrs with minimal at 12 hrs while VCP:∆F508-CFTR increased overtime (4 to 12hrs). Adding Tubacin (10µM; 4 hrs) reversed the proteasomal inhibitor effects. In pulse-chase experiments, MG-132 increased the accumulation of CFTR Bform and poly-ubiquitinated-∆F508-CFTR as compared to the untreated control. Tubacin suppressed the levels of MG-132 induced poly-ubiquinated-∆F508-CFTR as well as ∆F508-CFTR B-form. The VCP localization by microscopy showed the accumulation of VCP-GFP in perinuclear aggregates in the presence of proteasome inhibitor (PS-341 10µM). The 10µM Tubacin blocked these PS-341 induced perinuclear aggregates. We observed that VCP is associated with endocytic protein-complexes containing clathrin in the presence of PS-341 indicating cytosolic re-localization of VCP. We further confirmed the cytosolic re-localization of VCP-GFP in the presence of lysosome inhibitor (bafilomycin A1 10µM) by fluorescent microscopy.

Conclusion: We found that a small molecule inhibitor of aggresome function prevents HDAC6:∆F508-CFTR interaction, promotes VCP:∆F508-CFTR interaction, and suppresses the accumulation of poly-ubiquinated-∆F508-CFTR.

Wang, Y. 1 ; Jiang, Y. 1 ; Zhu, N. 1 ; Feng, X. 1 ; Yang, H. 1, 2 ; Ma, T. 1, 2 1. Membrane Channel Research Laboratory, Northeast Normal University, Changchun, China; 2. Biopharmaceutical Center, Liaoning Normal University, Dalian, China Deletion of the codon encoding the phenylalanine residue at position 508 (∆F508) in the cystic fibrosis transmembrane conductance regulator (CFTR) is the most common mutation causing cystic fibrosis (CF). The ∆F508 mutation results in a CFTR protein with impaired folding, trafficking and gating in human and rodents. The consequences of ∆F508-CFTR mutation in other species have not been studied. The purpose of the present study was to characterize the ∆F508 mutant of porcine CFTR in cell culture model. cDNA sequence encoding full-length porcine CFTR (pCFTR) was cloned from the lung by RT-PCR using primers designed according to porcine genomic sequences in two BAC clones (GenBank Accession No. AC092497 and AC092478) containing all exons of the porcine CFTR. pCFTR encodes a 1482-amino-acid protein that shows 92.3% identity to human CFTR. The phenylalanine residue at position 508 is conserved in pCFTR. Functional analysis of pCFTR stably expressed in Fisher Rat Thyroid (FRT) epithelial cells by short-circuit current assays indicated a cAMPactivated Cl-channel similar to human CFTR except that the sensitivity of pCFTR to the specific CFTR inhibitor CFTRinh-172 is 25-fold lower than that of human CFTR. A ∆F508 mutation of porcine CFTR was generated by site-directed mutagenesis. Surprisingly, ∆F508-pCFTR expressed as a func-tional chloride channel activated by forskolin in stably transfected FRT cells. Western blot analysis of COS7 and BHK cells transiently transfected with cMyc-tagged ∆F508-pCFTR indicated a protein pattern identical to that of wildtype pCFTR. Immunofluorescence analysis of the transiently transfected cells demonstrated similar plasma membrane expression pattern of ∆F508-pCFTR and wildtype pCFTR. These results clearly demonstrated unimpaired cellular processing of ∆F508-pCFTR. However, the sensitivity of ∆F508-pCFTR to forskolin activation (IC50~2.5 µM) was dramatically lower than that of wildtype pCFTR (IC50~0.5 µM), indicating a defect of channel gating by PKA phosphorylation. The present study provided the first evidence that the ∆F508 mutation in mammalian CFTR does not result in impaired cellular processing that is closely associated with CF phenotype. We have optimized airway epithelial cell spheroid cultures for functional measurements of CFTR chloride conductance in order to screen smallmolecule activators and inhibitors. Spheroids, which are sealed spheroidal monolayers of airway surface epithelial cells with diameter 75-250 micron, were generated within 2 to 4 days after nasal brushing and suspension culture. Yields of up to 2000 spheroids were obtained from nasal brushing of a non-CF or CF subject. Approximately 3,000 spheroids could be prepared from one million freshly isolated human bronchial epithelial cells. We compared several electrophysiological and fluorescence methods to assay CFTR function in spheroids. Whole-cell current was measured in single spheroids following pipette immobilization and micropipette puncture of the solutionexposed apical cell membrane. Large, forskolin-stimulated whole-cell chloride currents were measured in non-CF spheroids. As an independent approach, transepithelial potential difference was measured by micropipette insertion into the spheroid lumen. For fluorescence detection, cell cytoplasm was stained with the halide indicators MQAE or LZQ. Fluorescently stained spheroids were immobilized on a coated coverglass or in a custom microfluidic chamber. Halide flux across the apical surface was measured in response to chloride-nitrate or iodide-chloride solution exchange, in the absence and presence of cAMP activators. The fluorescence assay is suitable for medium-throughput screening of ∆F508 correctors and potentiators, with definitive verification of CFTR chloride channel function by whole-cell current analysis. An airway spheroid cultures provide a powerful approach to prioritize the efficacy of ∆F508 correctors because of their rapid generation without prolonged culture, and their suitability for quantitative analysis of CFTR function. Supported by CFF and NIH. Background and Aims: CFTR has been shown to be expressed and functional in ventricular cardiomyocytes. However, to date, no physiological role for CFTR in the heart has been established. The aim of this study was to determine if CFTR plays a role in the regulation of cardiomyocyte contraction rate. Methods: Cardiomyocyte contraction rate was assessed using the neonatal mouse beating assay. Ventricular myocytes were isolated from neonatal mice and cultured for 2-3 days until they formed a functional in sitium. Contraction rates were captured by video imaging and analyzed by MetaMorph in the presence and absence of beta-adrenergic receptor stimulation by isoproteronol (10 µM). Assessment of CFTR activity was determined by pharmacological inhibition by glibenclamide (100 µM), DPC (500 µM), and CFTR inh-172 (10 and 20 µM). The role of K ATP channels was assessed by 5-HD (100 µM). Results: Baseline contraction rate of vehicletreated (DMSO or H 2 O) cardiomyocytes was unchanged. Addition of isoproteronol caused a significant increase in contraction rate, which was sustained for at least 30 minutes. In contrast, application of glibenclamide to beating ventricular myocytes significantly, but transiently, inhibited cardiomyocyte contraction rate, with recovery occurring within 20 minutes. Glibenclamide did not affect the initial increase in contraction rate caused by isoproteronol stimulation, but inhibited sustained increases in isoproteronol-stimulated contraction rate. To determine if this was an effect of K ATP channel inhibition we examined the effect of 5-HD on contraction rate. In contrast to glibenclamide, 5-HD had no inhibitory affect on basal or isoproteronol-stimulated contraction rate, indicating glibenclamide's inhibitory effect was due to its inhibition of CFTR. To further verify this, we examined the effect of two other CFTR inhibitors, DPC and CFTR inh-172 on ventricular cardiomyocyte contraction rate. Both DPC and CFTR inh-172 elicited similar responses as glibenclamide, with a quick, but transient inhibition of baseline contraction rate and inhibition of sustained isoproteronol-stimulated increases in cardiomyocyte beating. Summary and Conclusions: Using neonatal murine ventricular myocyte cultures inhibition of CFTR leads to a significant, but transient inhibition of baseline ventricular cardiomyocyte contraction rate. Additionally, while inhibition of CFTR did not affect initial increases in beta-adrenergic receptor-stimulated contraction rate, CFTR inhibition did prevent sustained increases in contraction rate. The data suggest that CFTR plays a role in modulating ventricular cardiomyocyte contraction rate during resting and adrenergic-stimulated conditions.

Drevillon, L. 1,2 ; Tanguy, G. 1, 2 ; Hinzpeter, A. 1,2 ; Arous, N. 1,2 ; Goossens, M. 1,2 ; Fanen, P. 1, 2 1. Génétique, INSERM U.841, Créteil, France; 2. Faculté de Médecine, Université Paris 12, Créteil, France Cystic fibrosis is mainly caused by mutations that interfere with the biosynthetic folding of the CFTR protein. The aim of this study was to find cellular proteins capable of interacting with CFTR and modifying its processing or its trafficking. We have used a genetic screen in yeast to identify such proteins and identified COMMD1 that preferentially interacted with the third cytoplasmic loop of CFTR. COMMD1 is a regulator of copper transporter ATP7B in hepatocyte and sodium epithelial channel ENaC, but its exact biochemical function and physiological relevance remain elusive.

Here, we report that first, COMMD1 interacted with wild-type and F508del-CFTR in epithelial cells and second, COMMD1 subcellular distribution was not only both nuclear and cytoplasmic, but also in cytoplasmic vesicular compartments. We wondered if COMMD1 was involved in the processing and/or the trafficking of CFTR protein in epithelial cells. We demonstrated that COMMD1 was not involved in CFTR processing but that cell surface expression of wild-type CFTR at the plasma membrane was cut by half when COMMD1 expression was 90% reduced by RNA interference. We drew the conclusion that COMMD1 is a major component of CFTR trafficking and/or recycling at the plasma membrane, precise characterization of COMMD1 in these vesicular compartments is under process to decipher its function in CFTR trafficking.

COMMD1 has been labelled as the prototype of a newly described protein family that plays a role in inhibiting NF-kappaB signalling. Moreover, subcellular distribution of COMMD1 was different in CF and non CF cells, nuclear form of COMMD1 being increased in F508del cells compared to wild-type cells. Thus, this opens up a new area of investigation about COMMD1 nuclear function.

Finally, these data indicate that COMMD1 is involved in multiple cellular processes of outstanding interest in CF pathophysiology. Understanding how its modulation modifies transepithelial transport and inflammation in CF versus non CF cells should give new therapeutic clues to reduce exaggerated inflammation and improve fluid secretion in CF patients. Supported 1 ; Fanen, P. 2 ; Galietta, L.J. 1 1. Lab. of Molecular Genetics, Istituto Giannina Gaslini, Genova, Italy; 2. Inserm U.841, Creteil, France Class III cystic fibrosis mutations cause reduced Cltransport by impairing the process of CFTR channel opening. Typical class III mutations include G551D, G1349D, and F508del, all of them affecting nucleotide binding domains. It has been reported that mutations residing in other regions of the CFTR protein may also reduce channel activity. We investigated the extent of loss of function caused by such mutations, in particular by missense mutations localized in CFTR intracellular loops, and the sensitivity to pharmacological stimulation with CFTR potentiators.

To determine CFTR activity, plasmids carrying the coding sequence for wild type or mutant CFTR were co-transfected in COS-7 cells together with the halide-sensitive yellow fluorescent protein YFP-H148Q/I152L. The rate of CFTR-dependent anion transport was measured after short-term stimulation with forskolin alone or in combination with potentiators (felodipine, phenylglycine PG-01, or sulfonamide SF-01). Processing of CFTR protein was assessed by examining glycosylation status of CFTR by immunoprecipitation experiments.

When stimulated with forskolin alone, CFTR mutants showed the following anion transport compared to the wild type protein: I148T (45 %), I175V (153 %), Q179K (54 %), E193K (9 %), G970R (6 %), D1152H (29 %). Interestingly, all mutants having a reduced anion transport responded to stimulation with potentiators PG-01, SF-01, and felodipine (5 µM) with an increase in anion transport that approached the level of normal CFTR protein. The correction of defective CFTR activity by potentiators was confirmed in FRT cells expressing E193K and D1152H mutants using short-circuit current measurements.

Our results indicate that some missense CF mutations, like E193K and G970R, really cause a severe reduction in CFTR activity. Others, like D1152H, cause only a partial loss of function which may explain their association with a milder form of the disease. Conversely, some other amino acid changes (with the extreme example of I175V that seems to cause a gain of function) do not impair CFTR activity and therefore may represent polymorphisms and not CF-causing mutations. Interestingly, all mutants having a mild or more severe anion transport defect were sensitive to potentiators. This finding indicates that CFTR potentiators have a wide efficacy on many class III mutants and therefore may represent a promising therapeutic strategy for a significant number of CF patients.

Supported by CFFT and Telethon-Italy. CFTR, a Cl-channel membrane protein, is a member of the ATP-Binding Cassette (ABC) super-family. Mutations in CFTR result in a misfolded protein which fails to mature, leading to impaired flow of Cl-ions across the membrane of the epithelial cells lining the airway, and to CF. The most common mutation leading to CF is DF508.

Currently available structural information relevant to CFTR modeling includes low resolution structures of CFTR (20Å) and P-gp (8Å) and high resolution structures of other ABC transporters as well as nucleotide binding domains (NBD) . Of particular relevance is the newly published structure of the bacterial multidrug ABC transporter, Sav1866, whose topology can serve as a starting point for CFTR modeling.

As a first step towards the development of CF therapeutics, Epix has undertaken the modeling of the 3D structure of CFTR (excluding the R domain) in its conducting (i.e., open-channel, NBD-dimer) state, using our proprietary membrane-protein modeling technology (PREDICTTM) in combination with other modeling tools.

To date, we have modeled the cytosolic part of the receptor, namely the NBDs:ICLs construct and work is ongoing on modeling the transmembrane (MSD1:MSD2) region.

The now available partial model has been scanned for putative binding sites for DF508-CFTR correctors/potentiators. Several potential sites have been located, and Epix has initiated its in silico screening technology in order to identify potential drug candidates. This work is supported by the CFFT. The deletion of a phenylalanine residue at position 508 (F508del) in the first nucleotide-binding domain (NBD1) of CFTR is the principal cause of cystic fibrosis (CF). The altered interaction of F508del CFTR with endoplasmic reticulum (ER) quality control proteins, primarily chaperones, promotes its proteasomal degradation. However, it is believed that crucial CFTR-interacting proteins (CIPs) remain unknown [1] . Moreover, there is little information currently available on the strength of CIP-CFTR interactions. Our goals here are two-fold: 1) to quantify CFTR-CIP interactions; and 2) to isolate and characterise unidentified CIPs.

To address our first goal, we used surface plasmon resonance (SPR; Biacore TM ) to quantify real-time binding of Hsc70 (a well-known CIP) to bacterially expressed wt-and F508del-NBD1. Hsc70/Hsp70 was covalently immobilised onto the surface of carboxymethyl dextran (CM5) sensor chips (500 µM) and the real-time binding of purified NBD1 and control proteins quantified. In control experiments, anti-Hsc70 antibody 1B5 (20 nM; against residues 373-650) bound specifically to immobilised Hsc70 with high affinity (K D , 0.46 ± 0.07 nM; n = 3) whereas bovine serum albumin (15 µM) did not interact (n = 10). Given the difficulties associated with expression of hNBD1 carrying the F508del mutation [2] it was not used in our SPR analyses. Instead, we used purified murine (m) NBD1 to quantify the impact of F508del on the strength of the Hsc70-NBD1 binding. In the presence of low concentrations of ATP (≤5 uM), the affinity of NBD1 binding to immobilised Hsc70 was strengthened when F508 was deleted (wt, K D app , 1.20 ± 0.16 µM; F508del, K D app , 0.43 ± 0.11 µM; n = 3; P < 0.05). Interestingly, raising the concentration of MgATP weakened the binding of NBD1 (0.5 µM) to Hsc70 (wt, K I , 29 µM ATP; n = 2). Furthermore, deletion of F508 dramatically increased the concentration of MgATP required to destabilise the NBD1-Hsc70 interaction (F508del, K I , 221 µM ATP; n = 2). In summary, we used SPR to demonstrate that: (i) NBD1 of CFTR binds specifically to Hsc70, and (ii) the F508del mutation enhances the association of NBD1 with Hsc70. We are presently investigating the effect of co-chaperones and small molecules on the interaction of Hsc/Hsp70 with wt-and F508del-NBD1.

To identify novel CIPs, purified NBD1 was immobilised onto metalaffinity resin and used to capture CIPs from epithelial cell lysates. By this approach, CIPs were captured from human respiratory cell (Calu-3) lysates and analysed by 2D-electrophoresis. Relevant protein spots so far identified by mass spectrometry include: 1) Raichu404X (Thr/Ser kinase); 2) Profilin 2 isoform b; 3) Annexin A5; 4) Ifapsoriasin (intermediate filament-associated Ca 2+ regulatory protein); 5) MGC35308 (member of the ER reticulon family). Current analyses are underway to determine their functional roles in CFTR trafficking and function. [1] Amaral ( 

Billet, A. 1 ; Melin, P. 1 ; Norez, C. 1 ; Bilan, F. 2 ; Vandebrouck, C. 1 ; Mettey, Y. 3 ; Becq, F. 1 1. physiologie, CNRS UMR 6187, Poitiers, France; 2. GMAS UMR 6187, Poitiers, France; 3. faculté medecine et pharmacie, Poitiers, France In order to gain a better insight into the structure-activity relationship (SAR) of CFTR protein, we have functionally characterized a cystic fibrosis mutation: the CFTR-G622D, identified in patients with oligospermia (Vankeerberghen et al., 1998) . We examined CFTR chloride channel activity by patch-clamp analysis, in whole cell configuration, using Cos-7 cells transiently transfected with GFP-CFTR-wt or with the mutant GFP-CFTR-G622D. CFTR channels were activated by 10 µM Forskolin (Fsk), and, after stable activation 10 µM CFTRinh-172 was added. As GFP-CFTR-wt, G622D chloride channels elicit a time and voltage independent current in presence of 10 µM Fsk. GFP-CFTR-G622D current density at +40mV (52.08 ± 5.8 pA/pF) is 1.5 fold less than GFP-CFTR-wt current density at +40mV (76.67 ± 9.3 pA/pF). G622D mediated currents are blocked by 10 µM of the specific CFTR-inhibitor, CFTRinh-172 (Ma et al., 2002) . As expected, GFP-CFTR-wt channels are stimulated by 100 µM of the potent activator 5-butyl-6-hydroxy-10-chlorobenzo[c]quinolizinium chloride (MPB-91, Derand et al., 2001 ) with a current density at +40 mV of 127.44 ± 25.49 pA/pF. Interestingly, no activation of G622D channels is recorded in presence of MPB-91: difference between currents recorded in basal condition or in presence of MPB-91 is not significant. Using iodide efflux experiments, we showed that CFTR-G622D mutant does not respond to several benzo[c]quinolizinium derivatives. However, the mutant channel can be activated by forskolin, IBMX and genistein. Thus, the mechanisms of activation by xanthine and isoflavone are not affected by the mutation. These results show that:

(1) CFTR-G622D channels are functional: activation by cAMP pathway, inhibition by CFTRinh-172.

(2) cAMP chloride currents elicited by CFTR-G622D protein are weaker than CFTR-wt.

(3) CFTR-G622D channels are refractory to benzo[c]quinolizinium activation.

The replacement of the glycine by a negative charged amino acid in position 622 affects CFTR chloride channel activity and prevents activation by benzo [c] quinolizinium. This amino acid G622, localized in the interface of NBD1 and R domain between two β sheets (Callebaut et al., 2004) seems to have a crucial role in pharmacological activation of CFTR by MPBs. Further experiments will be performed to evaluate the role of the charge and of the nature of the amino acid at position 622 in CFTR protein.

Supported by Vaincre la Mucoviscidose and CNRS. Cysteine String Protein (Csp) is a J-domain containing protein that regulates the extent of wild type CFTR maturation. Over-expression of Csp blocks the formation of mature, glycosylated (band C) CFTR and increases the level of immature (band B) CFTR, whereas knockdown of Csp elicits a marked increase in CFTR maturation (Zhang et al., J. Biol. Chem. 281: 11312, 2006) . Thus, Csp appears to block the exit of CFTR from the ER. When the ER exit of CFTR is blocked by a different mechanism, i.e. the expression of mutant Sar1-GTP (Sar1 H79G interferes with the formation of COPII vesicles), the steady-state level of immature CFTR is two-fold greater than that resulting from Csp over-expression. This finding suggests that Csp, in addition to blocking ER exit of CFTR, may facilitate the degradation of the immature protein. In agreement with this concept, Csp expression increased CFTR ubiquitylation.

A mutation within the conserved HPD motif of the J-domain (H43Q) disrupts the ability of Csp to interact with and stimulate the ATPase activity of Hsp70. Csp H43Q does not block CFTR maturation (above ref); in addition, the level of immature CFTR was ~60% higher when H43Q Csp was over-expressed than the level observed with mutant Sar1. This finding suggests that the pro-degradative effect of Csp on immature CFTR requires Hsp70 binding/activation. CHIP (carboxy terminus of Hsp70-interacting protein) is an E3 ubiquitin protein ligase that can target WT and ∆F508 CFTR for proteasome-mediated degradation (Meacham et al., Nature Cell Biol. 3: 100, 2001) . We sought to determine weather CHIP was involved in the Csp-mediated degradation of CFTR. Indeed, co-immunoprecipitation (co-IP) experiments showed that CHIP associated with Csp. Also, Csp H43Q, which has reduced interaction with Hsp70, showed a similar level of CHIP interaction by co-IP, suggesting that their association did not depend on Hsp70 as a linker between the two proteins. A Csp mutant that is truncated after the cysteine string domain, which is its site of membrane attachment, blocked CFTR maturation as did WT Csp; however, this mutant did not decrease CFTR band B levels like WT Csp did. In addition, the Csp mutant missing its Cterminus did not associate with CHIP. These findings further implicate CHIP in the Csp-mediated degradation of immature CFTR, and they suggest that the C-terminus of Csp is a CHIP binding domain.

When a dominant-negative mutant of CHIP, which lacks the U-box needed for CHIP-mediated ubiquitylation (CHIP∆Ub), was co-expressed with CFTR, Csp was no longer able to facilitate the degradation of CFTR in the ER. These data show that it is possible to separate the effect of Csp on CFTR maturation from its ability to mediate the degradation of immature CFTR. Thus, Csp blocks the exit of CFTR from the ER, and this requires a J-domain mediated activation of Hsp70 but not the Csp C-terminus. The action of Csp on CFTR degradation requires both a functional J-domain and the Csp C-terminus, and this may require a trimeric complex involving Csp, Hsp70 and CHIP.

[ The cystic fibrosis transmembrane conductance regulator (CFTR) represents the main Clchannel in the apical membrane of epithelial cells for cAMP-dependent Clsecretion. Mutations of this channel causes cystic fibrosis disease ; thus discovery of pharmacological activators of CFTR is crucial to design future medicament for protein therapy. Recently, we reported on the synthesis and screening of a small library of 6-phenylpyrrolo [2,3b] pyrazines (named RP derivatives) evaluated as activators of wild-type CFTR, G551D-CFTR and F508del-CFTR Clchannels (Noël et al, 2006) . This preliminary structure-activity relationship study identified 4-hydroxyphenyl and 7-n-butyl as determinants required for activation of CFTR (RP-107 and RP-108). Here we studied structure-function relationship of more than 190 compounds prepared by chemical synthesis, and the subsequent activation of CFTR channels. Within the 6-phenylpyrrolo [2,3-b] pyrazines family, we observed by iodide efflux technique that RP-173 bearing a 2hydroxyphenyl substituant is more potent (EC 50 = 16 nM) than RP-107 having 4-hydroxyphenyl substituant (EC 50 = 150 nM). By whole-cell patch clamp recording analysis, we confirmed that nanomolar concentrations of RP-173 activate linear chloride current in CHO cells stably transfected with human wild-type CFTR. This current was inhibited by 10 µM of CFTR inh -172. We also found significant stimulation of short circuit current (I sc ) by RP-173 (EC 50 = 9 nM) on colon of Cftr +/+ but not of Cftr -/mice mounted in Ussing chamber. Stimulation of I sc by RP derivatives was inhibited by glibenclamide. The structural analogue , was less potent (EC 50 = 347 nM). As for RP-107 compound, we found that the 7-nbutyl chain is crucial for RP-146 and RP-173 activity, and that the 2-hydroxyphenyl compounds without 7-n-butyl chain are unactive on CFTR. In this study, we showed that the presence of an hydroxyl group at position 2, 3 or 4 of the pyrrolopyrazine cycle determined the highest activity on CFTR. The most potent compound is the 7-n-butyl-6-(2-hydroxyphenyl)5Hpyrrolo [2,3b] pyrazine . The potency of these agents indicates that compounds in this class may be of therapeutic benefit in CFTR-related diseases, including cystic fibrosis. The most common mutation causing cystic fibrosis (CF) is the deletion of F508 in the CFTR gene, which results in inefficient trafficking of the CFTR protein from the endoplasmic reticulum to the plasma membrane. The surface expression of the CFTR is usually quantified biochemically with Western blot or functionally with electrophysiological assays. However, these methods are labor intensive. We explore using fluorescence-based techniques for monitoring CFTR surface expression. Wild-type CFTR was tagged at the amino terminus with green fluorescent polypeptides (GFP-CFTR). The GFP-CFTR was either transiently or stably expressed in Chinese hamster ovary (CHO) cells. The cells were then treated with a red fluorophore that specifically labels plasma membranes. The fluorescent images were acquired with an Olympus laser scanning confocal microscope. Two image channels were taken simultaneously. The green channel was excited with a 488nm laser, and signals within 500-545nm were collected. The red channel was excited with a 559nm laser, and signals within 570-670nm were collected. For a single focal plane, each image was composed of 1024*1024 pixels. Threshold values were given to both green and red channels. Pixels with an intensity value above the threshold were marked as green/red pixels. The green pixels in the first channel represent the signal of GFP-CFTR, and the red pixels in the second channel represent the location of the plasma membrane. Thus, pixels that are both green in the first channel and red in the second channel are the CFTR surface expression, and denoted as yellow pixels. The percentage of successfully trafficked CFTR is thus quantitatively estimated by the ratio of yellow to green pixels in the single plane. Alternatively, a stack of images from the same cell at different depth was collected and analyzed. The total amount and surface portion of CFTR expression are then determined by summing the correspondent pixels. The percentage of CFTR surface expression was then calculated and will be verified both biochemically and functionally. The approach is modified to study deltaF508 CFTR trafficking. We anticipate that the method can be used to screen CFTR correctors from a small or intermediate sized chemical database. Supported by CFFT and NIH. The most common mutation of the cystic fibrosis transmembrane conductance regulator (CFTR) gene (deletion of Phe-508 (∆F508) in the first nucleotide binding domain (NBD1)) causes retention of the ∆F508 protein in the endoplasmic reticulum (ER). ∆F508 mutation prevents conformational maturation of CFTR protein without altering profoundly the local structure of NBD1, but possibly by disrupting the interaction between NBD1 and NBD2. However, the individual role of NBD1 and NBD2 in biogenesis, folding, maturation and membrane stability of a full length CFTR is still under debate. Using splicing by overlap extension method (SOE by PCR), we generated six recombinant CFTR proteins with inverted or suppressed NBDs. The native boundary domains of NBD1 and NBD2 were respected for all mutants (Riordan J.R. et al. 1989 ) and their C-terminal tails preserved. The constructs were transiently or stably expressed in Cos-7 and BHK cells respectively. Immmunodetection and immunolocalization assays confirmed that deletion of NBD2 domain (N1-CFTR) did not alter the trafficking or the plasma membrane expression of CFTR. Both processes were abolished in the presence of ∆F508 mutation. However, using whole cell patch-clamp configuration, we did not detect any cAMP-stimulated Clcurrent. Pulse-chase experiments established that the turnover of N1-CFTR is 3-4 fold faster (T1/2~ 5h) than its wild type counterpart (T1/2~ 18h). When we replaced NBD2 with NBD1 (2N1-CFTR), we could detect a faint plasma membrane expression associated with a weak Cl-channel activation. More surprisingly, the T1/2 of 2N1-CFTR coreglycosylated form was around 12h. When NBD1 was deleted (N2-CFTR) or substituted with NBD2 domain (2N2-CFTR), the expression and functional pattern were similar to CFTR ∆F508. Furthermore, we showed that the N1-CFTR folding was not attenuated as demonstrated by protease susceptibility and SDSresistant thermoaggregation tendency, supporting the notion that N1-CFTR membrane stability requires NBD1-NBD2 interaction. Altogether, our results suggest that NBD1 domain is essential to form a foldable CFTR that satisfies ER quality control (ERQC) but correct inter-domain assembly is mandatory for its stability at plasma membrane. The results are also in agreement with previous suggestion that ∆F508 mutation alters the interactions between NBDs and transmembrane domains of CFTR.

Supported 1 1. Inserm U845, Inserm, Paris, France; 2. INSERM U845, Inserm, Paris, France; 3. INSERM U845, Inserm, Paris, France; 4. INSERM U845, Inserm, Paris, France; 5. INSERM U845, Inserm, Paris, France; 6. INSERM U845, Inserm, Paris, France; 7. INSERM 845, Inserm, Paris, France; 8. INSERM U845, Inserm, Paris, France The physiological role of CFTR in renal epithelium is not known. In the proximal tubule, CFTR co-localizes with the sodium-phosphate cotransporter NPT2a, a protein involved in phosphate (Pi) reabsorption. The aim of our work is to determine if there is a functional interaction between CFTR and NPT2a. To this end, the ARNm coding for CFTR and the ARN coding for NPT2a (bearing a Myc tag in C-ter) were injected in Xenopus laevis oocytes. Electrophysiological or biochemical experiments were performed 3 days after ARNs injection. Currents induced by 1mM Pi (IPi) were measured in voltage-clamped (-50 mV) oocytes that expressed NPT2a alone, or in oocytes co-expressing CFTR + NPT2a. For immunoprecipitation experiments, 24-1 CFTRAb and Myc Ab were used.

Results IPi was significantly reduced in oocytes expressing CFTR + NPT2a compared to that measured in oocytes expressing NPT2a alone. Using a two-bath electrodes configuration during voltage-clamping did not change this observation. This result suggests that when expressed in oocytes, CFTR is in functional interaction with NPT2a. Kinetic analysis (IPi as a function of Pi concentration) of NPT2a showed a reduction in Vmax but not in phosphate Km. This suggests that NPT2a membrane expression and/or activity is altered when CFTR is co-expressed. Co-immunoprecipitation experiments performed on protein lysate derived from oocytes co-expressing CFTR and NPT2a suggested an interaction between the 2 transporters. Using a PKA activating cocktail as a PKC activator induced CFTR-mediated currents in CFTR+NPT2a expressing oocytes. In these oocytes, IPi is decreased by PKC activation (as expected from NPT2a regulation) but is increased shortly after exposure to the PKA activating cocktail. The increase of IPi observed under this condition is at variance to that expected from classical NPT2a regulation. In oocytes expressing NPT2a alone, both PKA activation and PKC activation reduced IPi. These observations suggest that in oocytes, CFTR expression modifies NPT2a regulation.

Our study suggests that, after expression in Xenopus laevis oocytes, CFTR interacts functionally with NPT2a, and modifies its regulation. An interaction is also suggested by the co-immunoprecipitation of CFTR and NPT2a. These results cannot discriminate between a direct or an indirect (via a third protein) interaction. To determine if an intracellular chloride concentration change may participate to the functional interaction, we are now measuring in CFTR+NPT2a expressing oocytes intracellular chloride activity (using chloride-sensitive microelectrodes) during the use of the PKA activating cocktail, and measuring IPi when PKA is stimulated in the presence of a specific CFTR inhibitor. Cystic fibrosis (CF), the most commonly inherited lethal pulmonary disorder in Caucasians, is caused by mutations in the cystic fibrosis transmembrane conductance regulator gene (CFTR). To date, more than 1000 mutations were identified in CFTR and were associated with clinical disease. A phenylalanine deletion at position 508 accounts for 70% of CF genotypes in Caucasian populations, and determines CFTR misfolding and degradation by proteasome. Consequently, limited CFTR abundance leads to multiorgan disease, affecting the lung, pancreas, gut, liver, sweat glands and the reproductive organs. The past several years have witnessed an explosion of information regarding the identification and elucidation of molecules and pathways that are regulated by CFTR or that regulate CFTR activity. Genomics and proteomics technologies now offer the opportunity to examine global alterations in the mRNA and protein expression patterns of CF cells and tissues to elucidate the pathways linking defective CFTR to clinical disease. Here we describe systems biology methods that integrate heterogeneous datasets, including protein-protein interaction networks, gene expression and mass-spectrometry profiles, and mutation and genetic variation information, in order to identify the regulatory circuits and active subnetworks that are responsible for the progression and the severity of CF disease. The study will focus on correlations that define the key events in the CFTR folding and trafficking pathways that lead to pathogenesis and dysfunctions, and could provide insights and targets for intervention and therapy. The number of CFTR Cl-channels in the plasma membrane, and thus the transepithelial Cl-secretion is controlled, in part, by the endocytic trafficking of CFTR. Rab GTPases regulate the endocytic trafficking by acting as molecular switches that cycle between the GDP-bound (i.e. inactive) and the GTP-bound (i.e. active) state, associate with target membranes, and recruit downstream effectors. Rab5 facilitates CFTR endocytosis and Rab11a facilitates the recycling of internalized CFTR to the plasma membrane. Furthermore, Rab7 and Rab9 facilitate CFTR trafficking to the lysosome. The mechanism of sorting internalized CFTR for either recycling to the plasma membrane or degradation in the lysosome is not well understood. Recent evidence demonstrates that protein sorting occurs in Rab4 specific sorting endosomes. Presence of CFTR in Rab4 compartment has been previously demonstrated in human airway epithelial cells by electron microscopy. Yet, the role of Rab4 in CFTR trafficking has not been determined in these cells. Rab4 did not control CFTR trafficking in fibroblasts (BHK cells). However, it negatively regulated the plasma membrane expression of CFTR in the colorectal cells . The goal of this study was to elucidate the role of Rab4 in the endocytic trafficking of CFTR in polarized human airway epithelial cells (Calu-3 cells and CFBE41o-cells stably expressing WT-CFTR or deltaF508-CFTR). Endogenous Rab4 coimmunoprecipitated with CFTR in Calu-3 cell and CFBE41o-cells. To determine if Rab4 plays a role in CFTR trafficking, CFBE41o-cells were transfected with either the FLAG-tagged, dominant negative (DN) Rab4 mutant deficient in GTP binding (GDP-locked; FLAG-Rab4-N121I) or with a non-specific cDNA control (GFP). If Rab4 sorts internalized CFTR to the recycling pathway, the DN Rab4 should inhibit recycling, and, thus, should decrease the plasma membrane expression of CFTR. If, on the other hand, Rab4 sorts internalized CFTR for degradation the DN Rab4 should inhibit CFTR degradation, increase CFTR recycling and thus, should increase the expression of CFTR in the plasma membrane. The DN Rab4 increased CFTR expression in the plasma membrane as determined by cell surface biotinylation. These data are consistent with the role of Rab4 in sorting internalized CFTR for degradation. Similar to overexpression of the DN Rab4, silencing endogenous Rab4 with double stranded small interfering RNA specific for the human Rab4 sequence increased the plasma membrane expression of CFTR and increased the CFTR mediated Cl-secretion across polarized CFBE41o-cells in Ussing chamber experiments. The plasma membrane half-life of deltaF508-CFTR is decreased compared to WT-CFTR. However, after silencing Rab4, the plasma membrane half-life of deltaF508-CFTR was similar to that of WT-CFTR, as determined by CFTR disappearance from the plasma membrane following incubation with cycloheximide. Our data are consistent with the role of Rab4 in sorting the internalized CFTR for degradation and are consistent with the changing paradigm for the role of Rab4 in protein trafficking (supported by the NIH Grant 1 P20 RR018787 and the CFF SWIATE03QO). The conformational changes underlying activation of phosphorylated CFTR by ATP remain unclear. In the present study we assessed the utility of labeling endogenous cysteine(s) in CFTR using an environmentally-sensitive fluorescent probe Alexa-488 in monitoring such changes in structure.

These studies were performed using purified and functionally reconstituted wild-type CFTR. We labeled PKA-phosphorylated CFTR with 138 µM Alexa-488 maleimide prior to reconstitution in phospholipids liposomes. Changes in fluorescence intensity were monitored in a suspension of proteoliposomes following addition of 1 and 5 mM Mg-ATP or 5 mM Mg-AMP-PNP in order to evaluate nucleotide dependent changes in conformation. In order to identify the labeled cysteines, the protein was subjected trypsin-mediated proteolysis and analysis by mass spectrometry.

We have shown that CFTR is labeled using Alexa-488 maleimide. A significant increase in fluorescence emission occurred in Alexa-488 labeled CFTR of 17 ± 5 (AU) relative to empty liposomes one minute after the addition of activating nucleotide, 1 mM Mg-ATP (n = 3). The non-hydrolyzable analogue, 1mM Mg-AMP-PNP, failed to cause an increase in fluorescence of Alexa-488-labeled CFTR. However, the higher concentration of 5 mM Mg-AMP-PNP evoked an increase in fluorescence similar in magnitude to that of 5 mM Mg-ATP suggesting that the fluorescence response reflects structural changes relating to ATP binding to CFTR. Analysis by mass spectrometry identified the labeled residue as C647. This cysteine resides at the junction between NBD1 and the "R" domain in a region which has been suggested to be highly dynamic on the basis of multiple crystal structures. Our studies suggest that this flexible region around C647 of CFTR undergoes conformational change upon Mg-ATP binding such that it moves the Alexa 488 fluorophore into a relatively more polar environment. This finding presents a first case where conformational changes in the CFTR NBDs induced by ATP binding could lead to changes in the conformation of the R domain.

Further, these studies suggest that fluorescence methods can be used to probe dynamic conformational changes in purified reconstituted CFTR.

Acknowledgements Cystic Fibrosis is caused by mutations in the apical chloride channel CFTR with 90% of patients carrying at least one allele of the ∆F508 mutation. This mutant form of CFTR is characterized by a trafficking defect in which it fails to exit the endoplasmic reticulum (ER). We have previously reported that ∆F508 CFTR is found in complex with Hsp90 and its co-chaperones in cell extracts suggesting that either this mutant form of the protein accumulates in on-pathway folding intermediates which result in its accumulation in the ER or that these intermediates include inhibitory factors leading to retention of ∆F508 CFTR in this compartment. We have examined the role of one of these Hsp90 co-chaperones, FK506 binding protein 8 (FKBP8), in the life cycle of CFTR. FKBP8 is a unique member of the immunophilin family of peptidyl-prolyl isomerases (PPIases) which mediate the cis/trans interconversion of peptidyl-prolyl bonds, which is thought to represent a rate limiting step in protein folding. In order to establish a functional role for FKBP8 in CFTR stability and trafficking, we modulated the expression level of this protein in human bronchial epithelial cells endogenously expressing either wild type or ∆F508 CFTR. A knockdown of FKBP8 by siRNA resulted in a destabilization of both wild type and ∆F508 CFTR as well as a concomitant loss in channel activity of wild type and temperature corrected ∆F508 CFTR. Conversely, over expression of FKBP8 resulted in a stabilization of the ER glycoform of ∆F508 CFTR in both mammalian and yeast systems. These data suggest that FKBP8 is an essential component of the Hsp90-mediated folding machinery for both wild type and ∆F508 CFTR, which supports the hypothesis that ∆F508 CFTR accumulates in the ER in an on-pathway folding intermediate.

DMH CFTR folding involves a series of sequential, although potentially overlapping steps that involve i) formation, orientation and integration of (TMs), ii) helical packing, iii) folding of cytosolic and extracytosolic domains, and iv) establishing proper domain-domain contacts. This process begins cotranslationally and is facilitated by the Sec61 ER translocation machinery and a diverse group of lumenal and cytosolic chaperones. The most common inherited CFTR mutation, deltaF508, disrupts one or more early steps along this folding pathway. A major obstacle in understanding how deltaF508 causes CF is that the native folding environment, comprised of lipids, cytosolic and ER proteins, is not amenable to traditional biochemical and biophysical approaches. To overcome this problem we are developing a flu-orescence based analytical approach that enables one to directly access nascent ribosome-attached folding intermediates at virtually any stage of synthesis. The strategy requires incorporation of fluorescent probes into the substrate protein, which provide a highly selective means to monitor structural features at sub-nanomolar concentrations in highly complex biological mixtures. Polypeptides are synthesized in vitro from truncated mRNA templates to generate uniform cohorts of ribosome-bound nascent chains. Progressive stages of folding can then be evaluated using mRNA templates of different length. As proof-of-principle we have shown that fully synthesized GFP variants can be readily trapped in an unfolded state and complete their folding only after the entire nascent peptide has exited the large ribosomal subunit. This approach enabled us to measure real-time folding kinetics of extended peptide domains following synchronous ribosome release and compare fluorophore maturation of four fluorescent protein variants with different spectral properties CFP, GFP, Venus and mCherry. To extend this approach to CFTR and other general protein substrates, we are employing fluorescence resonance energy transfer (FRET) using "donor" and "acceptor" probes incorporated simultaneously into the substrate protein. An in frame N-terminal fusion to CFP provides the "Donor" fluorophore, while the "acceptor" fluorophore is incorporated at an engineered stop codon using a synthetic modified aminoacyl-suppressor tRNA. We are currently developing a 'molecular ruler' to correlate the efficiency of energy transfer with changes in the relative distance between probes in ribosome-bound and ribosome-released CFTR folding intermediates. FRET-based experiments are underway to evaluate folding of both N-terminal (1-80aa) and NBD1 (387-674aa) domains. These studies will provide new insight into the mechanism of cotranslational CFTR domain folding and provide a potential means to identify agents that correct the deltaF508 folding defect. Relationships between hypoxia, Ado release, and prostenoid production in ariway cells are not defined. In the present studies, we examined mechanisms governing hypoxia-induced production of prostenoids in CFBE41o-and Calu-3 monolayers. Using ELISA-based detection of prostenoid and leukotriene production, both Calu-3 and CFBE41omonolayers treated with Ado (10 µM) or arachidonic acid (AA, 10 µM, the parent molecule of prostenoids and leukotrienes) for 16 hrs led to increases in PGE 2 levels (1198-1283 ± 64-145 pg/mg of total proteins vs controls 732 ± 89 pg/mg in Calu-3 cells, 756-839 ± 8-12 pg/mg vs controls 408 ± 42 pg/mg in CFBE41o-cells, p<0.05 in both airway cell types compared with control). In contrast, LTB 4 and LTC 4 release were not stimulated by both maneuvers. We next used siRNA techniques to knockdown A2B adenosine receptor expression, and demonstrated durable knockdown of A2B AR expression to~25% of scrambled siRNA-treated controls following lipid-based transfection with siRNA directed against the A2B AR in CFBE41o-cells. Following siRNA knockdown of A2B AR expression, cells were exposed to hypoxic conditions (mucosal volume = 2 ml in 12-well plate), normoxic conditions (mucosal volume = 0.2 ml, FiO 2 = 0.21), and Ado (10 µM). Hypoxic stress led to high levels of mucosal PGE 2 production (1613 ± 102.8 pg/mg vs controls 495 ± 19.6 pg/mg, p<0.01) that was sensitive to A2B AR knockdown (743 ± 70.6 pg/mg, p<0.01). Exogenous Ado also stimulated PGE 2 production, but this effect was small relative to hypoxia. We next examined basolateral effects of Ado and AA on transmucosal Clsecretion in CFBE41o-and Calu-3 cells grown as monolayers on permeable supports and conducted Ussing chamber analysis under voltage clamp conditions. Addition of Ado and AA to the basolateral membrane (10 µM) activated transmucosal Clsecretion [18.94 ± 4.07 µA/cm 2 (Ado) and 14.30 ± 2.6 µA/cm 2 (AA) for Calu-3 cells; and 12.65 ± 2.31 µA/cm 2 (Ado) and 18.96 ± 1.24 µA/cm 2 (AA) for CFBE41o-cells expressing wtCFTR] that was sensitive to basolateral blockade with barium Cl -(10 mM). Isc by either agonist was stimulat-ed in the presence and absence of apical glybenclamide (200 µM), adenosine deaminase (ADA -2U/ml) and hexokinase (HEXO -2U/ml), confirming that the effects were specific for the basolateral membrane and could be accomplished with or without CFTR activity. These results demonstrate that hypoxia is sufficient to stimulate PGE 2 production through A2B ARs, and that both Ado and AA activate basolateral K + channels to promote transepithelial Clsecretion. The findings further support a model in which Ado (and prostenoids) regulate Cltransport through effects on both the apical and the basolateral membrane, and elucidate molecules that couple the two membranes as part of the normal Clsecretory response. Hypoxia would be predicted to increase local prostenoid production through Ado release and stimulation of A2B ARs, and further highlight a unifying model by which local hypoxia promotes inflammation in airway epithelia. Supported by the NIH and CFF.

Low levels of tissue oxygen -observed in lung diseases such as cystic fibrosis -initiate a signaling cascade resulting in altered transcription of genes possessing a hypoxia response element (HRE). We recently used gene chip mRNA array, quantitative RT-PCR, protein analysis, and functional (short circuit current) assays to show that CFTR expression is strongly repressed by hypoxia in vitro and in vivo. Among several thousand human mRNAs suppressed by low ambient oxygen, CFTR was among the most strongly inhibited (by 10-20 fold) in a cell model system. However, CFTR and the vast majority of other repressed genes lack a traditional HRE. Because hypoxia inducible factor (HIF) acts primarily as a transcriptional activator, the mechanisms underlying hypoxia mediated suppression of mammalian genes are not well understood. We therefore tested the hypothesis that microRNAs play a central role during transcriptional regulation of genes such as CFTR. MicroRNAs are evolutionarily conserved, short noncoding RNA sequences believed important for repression of gene expression. Some reports have estimated that at least 30% of protein-coding genes are regulated by microRNAs. Thousands of microRNAs have been predicted in the human genome by available algorithms (e.g. PalGrade, MiRscan, and ProMiR) and several hundred experimentally validated. The CFTR gene contains 16 predicted target sites for 13 microRNAs in the 3'UTR, including hsa-miR-383 (nucleotides 1309-1331 of the CFTR 3'UTR), hsa-miR-449 (nt 1443-1463) and hsa-miR-509 (nt 1041-1063). However, no study to date has examined the effect of microRNAs on CFTR message levels, or the influence of microRNAs during the CFTR response to environmental perturbations such as hypoxia. We therefore performed a global, genome-wide analyses of both mRNAs and microRNAs expressed during hypoxia (using air/liquid or liquid/liquid interface culture systems) in HT29 (colonic epithelial) cells that robustly express CFTR. We found that approximately 9% of microRNAs were significantly altered by hypoxia, as judged by miRNA profiling (Ambion mirVana™ Bioarray 1566v2, Asuragen microRNA Expression Profiling Service). Comparison of these hypoxia-related microRNAs with expression profiles of their predicted targets indicated a much lower level of correlation than has been previously hypothesized by others. We show that none of the microRNAs predicted to regulate CFTR are altered in HT29 (in response to hypoxia). These findings are therefore consistent with the notion that highly expressed genes in a particular tissue type are not co-expressed with their predicted regulatory microRNAs. Our data also indicate that microRNAs do not mediate the hypoxic repression of CFTR via the 3' UTR. Supported by NIH and CFF. 1 1. Physiology, McGill University, Montreal, QC, Canada; 2. Biochemistry, McGill University, Montreal, QC, Canada The ∆F508 mutation impairs CFTR maturation and trafficking to the plasma membrane and results in a partially functional chloride channel that is retained in the endoplasmic reticulum and degraded. Using our highthroughput trafficking assay, we previously showed that the quinoline KM60, a structural analogue of sildenafil, corrects the ∆F508-CFTR processing defect and leads to a significant increase in CFTR surface expression after 2 h treatment with 10 µM and after 24 h treatment with 10 nM according to detection of mature CFTR in Western-blots. Additional studies have now been carried out with other preparations to assess the functionality of ∆F508-CFTR following rescue by KM60 treatment.

Different time and concentration exposures were tested on the CFTR function using iodide efflux assays with both BHK and human bronchial epithelial cells over-expressing ∆F508-CFTR. Exposure to 10 µM KM60 for 2 h partially restored iodide efflux responses to 10 µM forskolin + 50 µM genistein in both cell line. After 24 h incubation, rescue of the mutant protein in BHK cells still required 10 µM, whereas 100 nM was sufficient in the human cell line CFBE. Iodide efflux assays were performed at different times following KM60 wash out to functionally assess the stability of the rescued mutant protein. cAMP-stimulated iodide efflux was still detectable 6 h after removing KM60, but responsiveness was lost after 24 h in both cell types. Electrophysiological studies were performed to examine the channel activity of ∆F508-CFTR rescued by KM60 treatment. Incubation with 10 µM KM60 for 48 h increased cAMP-responsive ∆F508-CFTR current density in HEK293 whole-cell patches ~6-fold. The conductance was time-and voltage-independent, anion selective, and strongly inhibited by glibenclamide. Incubating polarized CFBE monolayers with 20 µM KM60 increased the forskolin + genistein-stimulated short-circuit current after 48 h, and this response was inhibited by CFTRinh172 (10 µM). Permeabilization of the basolateral membrane with nystatin confirmed the apical location of the forskolin + genistein-stimulation. Ussing chamber experiments also revealed trafficking correction in ileum that had been excised from Cftrtm1Eu mice homozygous for ∆F508. When the ileum was pre-incubated with 20 µM KM60 for 4-6 h, the forskolin + genistein-stimulation of short-circuit current was significantly increased when compared to untreated intestine. These results confirm partial correction of ∆F508-CFTR by KM60 in unpolarized cells and in human polarized epithelial cell monolayers and ex-vivo intestine isolated from ∆F508-CFTR mice. This work was supported by the BREATHE program funded by the CCFF and CIHR, CIHR, CFFT, and Génome Québec. Primary sequence and overall structure of human CFTR overlaps significantly with murine, rabbit and other homologues, a feature that has limited generation of conformation specific reagents to address CFTR folding. Because regions of ∆F508 CFTR exposed extracellularly may exhibit aberrant topology (JBC 279:39620-7), probes that identify CFTR from the extracellular surface could be useful to 1) verify plasma membrane localization of ∆F508 CFTR, 2) discriminate wild-type from mutant protein, and 3) provide a means of determining whether small molecule "correctors" of the ∆F508 processing lead to a native (wildtype) CFTR configuration. Our laboratories are investigating phage-display techniques suitable for detecting properly folded extracellular domains of CFTR. Two large phage libraries encoding 7 or 12 amino acid sequences (approximately 10 9 phage per library) were panned against extracellular regions of CFTR in HeLa cells.

Following incubation in blocking solution containing PBS and 1% BSA at 4°C, unbound fraction was incubated with recombinant HeLas expressing wild-type or temperature corrected ∆F508 CFTR (protein expression verified biochemically and functionally). Bound bacteriophage was dislodged after multiple washings by exposure to an acidic elution buffer and amplified to increase copy number (five rounds of antigen selection for each library, Eur J Biochem 268:2004-12). Supernatants from microtiter wells expressing phage were analyzed for binding to parental (no CFTR), wt or ∆F508 CFTR expressing HeLa cells. Using this method we obtained enrichment of phage at the surface of cells expressing wt or ∆F508 CFTR. An example using a 7-amino acid sequence (10 10 phage) with specificity for ∆F508 CFTR (representative of four repeat experiments) is shown in a binding assay with 10 6 HeLa cells/condition. Reagents such as these can be further optimized by constructing second generation libraries using core recognition sequences (derived from the initial screen) and surrounding the core epitopes with random amino acid diversity. These experiments are intended to identify peptides with high binding affinity (nM binding constants) that recognize native CFTR epitopes exposed extracellularly. Because there is evidence that low temperature corrected ∆F508 CFTR in the plasma membrane is misfolded, these probes may ultimately distinguish wild-type CFTR from the surface targeted ∆F508 mutant and be useful as drug discovery reagents. Supported by CFF and NIH. Bridges, R.; Nagubadi, S.; Thakerar, A.; Jia, Y.; Bradbury, N.A. Physiology and Biophysics, Rosalind Franklin University of Medicine and Science, North Chicago, IL, USA The goal of this project was to compare several reported correctors of ∆F508-CFTR biosynthesis for their efficacy on chloride channel activity under a standard set of experimental conditions. Fisher rat thyroid (FRT) cells were transfected with ∆F508-CFTR and a stable cell line selected with G418. Cells were grown on transwell filters and studied under short circuit current (I SC ) conditions in Ussing chambers. I SC measurements were performed at 27°C. The basolateral membrane was permeabilized with α toxin and chloride currents were measured with a serosal to mucosal chloride gradient. ATP (1 mM) and GTP (100 µM) were added to the serosal bath. CFTR channel activity was stimulated with cAMP (100 µM), IBMX (100 µM) and genistein (30 µM). Cells were treated with correctors or vehicle (DMSO) at 24 and 48 hours before the I SC measurements. A subgroup of cells were temperature corrected at 27°C for 48 hours for comparison. Because only modest currents were observed even in the temperature corrected cells (3 -6 µA/cm 2 ) cells were treated for 48 hours with 5 mM sodium butyrate and 30 nM dexamethasone to enhance the observed currents. Under these conditions temperature corrected cells displayed stimulated chloride currents of ~100 µA/cm 2 that were 6-8-fold greater than cells maintained at 37°C. Correctors C4 (N- 5' ]bithiazolyl-2'-yl]-benzamide) and C1 (6-(1H-Benzomidazol-2-ylsulfanylmethyl)-2-(6-methoxy-4-methyl-quinazolin-2-ylamino)pyrimidin-4-ol) caused a modest rescue of ∆F508-CFTR chloride currents compared to cells maintained at 37°C (C4, 1.7-fold; C1, 1.3-fold) but far below the effect of temperature correction. Rescue by C4 was concentration dependent with a significant observable effect at 300 nM and a maximum effect at 1 µM. Rescue by C1 was observed at only 3 µM, lower concentrations had no effect and higher concentrations caused an inhibition of the stimulated chloride currents. Compounds C2 -piperazin-1-yl]-ethyl}-4-piperidin-1-yl-quinazoline) (1 to 10 µM), C3 (4-Cyclohexyloxy-2-{1- [4-(4- methoxy-benzensulfonyl)-piperazin-1yl]-ethyl}-quinazoline) (1 to 10 µM) and 4 phenylbutyrate (3 mM) did not significantly increase the stimulated chloride currents. All four compounds C1 -C4 caused significant decreases in the transepithelial resistances at the higher concentrations (3 -10 µM) suggesting some degree of cytotoxicity.

Our results indicate only C1 and C4 demonstrate efficacy in FRT cells and that these effects are modest compared to temperature correction. Supported by the Cystic Fibrosis Foundation. CFTR contains two membrane-spanning domains (MSD), two nucleotide-binding domains (NBD), and a regulatory domain that require proper assembly for chloride channel activity. The most prevalent disease causing mutation, CFTR∆F508, arises from deletion of phenylalanine 508 in the NBD1 domain. The CFTR∆F508 mutant is translated and inserted into the endoplasmic reticulum (ER) membrane, but is unable to attain its native state and accumulates in a kinetically trapped conformation that is degraded by the ubiquitin proteasome system (1) . The nature of the misfolded CFTR∆F508 intermediate that is selected for premature degradation is not clear and is a topic of great interest (2) . Recent work in the Cyr lab indicate that defects in CFTR and CFTR∆F508 folding are sensed by two different E3 ubiquitin ligase complexes that are located in the ER membrane and cytosol (3) . The membrane-associated Derlin-1/RMA1 E3 complex appears to recognize early folding defects of CFTR that may involve assembly of MSD2 into a complex with the amino-terminal regions of CFTR. In contrast, the cytosolic Hsc70/CHIP E3 complex appears to sense folding defects that occur after synthesis of the NBD2 domain. Misfolded CFTRs recognized at either checkpoint are retained and degraded via the proteasomal pathway. The presence of dual protein quality control (QC) checkpoints suggests a mechanism by which the folding status of CFTRs membrane and cytosolic domains are sequentially monitored prior to its escape from the ER. Currently little is known about how these ERQC complexes sense misfolded substrates.

Small-molecule chemical correctors have recently been identified which rescue the folding defect of CFTR∆F508 in model cell culture systems. In the presence of these chemical correctors a subset of CFTR∆F508 is able to localize to the cell surface and retain channel function. However, how the chemical correctors rescue CFTR∆F508 folding is not clear. We have examined the extent to which the molecules correct folding defects that are recognized by the Derlin-1/RMA1 or Hsc70/CHIP QC checkpoints. In addition, because it is unknown at what point of CFTR biogenesis that the chemical correctors act to facilitate folding, we have determined how the chemical correctors affect the folding of different biogenic intermediates. Our results indicate that the correctors act in a post-translational manner to influence assembly of CFTR sub-domains. Interactions with members of the QC pathway have also been used to monitor the folding and assembly process of CFTR in the presence of chemical correctors. Ultimately the results from these studies will enhance our knowledge of how chemical correctors permit passage of the CFTR mutants past different ERQC checkpoints.

Supported by -Cystic Fibrosis Foundation Combination therapy has proven successful in treating a wide variety of human diseases, including cancer, infectious disease, and diabetes. Historically, combination treatments have been discovered through trial and error using drugs with proven disease modifying effects. We have developed a combination high-throughput screening (cHTS™) technology to systematically and efficiently find synergistic combinations that target multiple pathways underlying disease biology, and that may be developed into new therapeutics. We are applying cHTS technology to discover and develop combination therapeutics for the treatment of cystic fibrosis.

We have optimized two high throughput screening assays for identifying correctors and potentiators of the CFTR (Cystic Fibrosis Transmembrane protein Regulator) ∆F508 mutation. One assay uses the FLIPR membrane potential dye (FMP) and the other uses genetically modified yellow fluorescent protein YFP H148Q/I152L to measure changes in halide flux through ∆F508 CFTR. Both assays were optimized and validated on the FLIPR TETRA using the panel of CF correctors, potentiators and inhibitors provided by the CF Foundation. Using our automated cHTS, we systematically evaluate pairwise combinations of a library of ~2,000 approved drugs and other bioactive molecules to identify novel disease-relevant biological interactions. Candidate single agents and combinations are tested in secondary assays and prioritized for testing in animal efficacy models.

Combination therapies can act on multiple pathways in a coordinated way. cHTS may prove to be a useful tool for identifying combination therapeutic candidates with novel multi-target mechanisms and a way to enhance co-therapy regimens for the treatment of cystic fibrosis.

Supported The highly variable clinical phenotypes of CF airway disease suggest that a number of genetic factors, other than CFTR, play a role in its pathophysiology. Some of these modifier genes are expected to play a role in the endoplasmic reticulum quality control (ERQC), since the major defect caused by F508del is misfolding, retention and degradation of the mutant protein by this cellular surveillance system. The nematode Caenorhabditis elegans is an excellent multicellular genetic model, which has been successfully used in studies of human diseases. Its ~20,000-genes genome has been fully sequenced, it has short life-span (2-3 weeks) and life cycle (~ 3.5 days) and it is easily cultured and amenable for gene disruption, both by knockout or RNAi.

Our goal is to generate a C. elegans model for the CFTR folding defect, so as to take advantage of this model for the identification of genes involved in CFTR folding and/or degradation.

To this end, and because no CFTR orthologue has been described in C. elegans, we constructed two previously described human P-gp-/CFTR chimeras (P-gp/wt-CFTR and P-gp/F508del-CFTR) [1, 2] to be used as an in vivo folding substrate in this organism. The two chimeric cDNAs and intact human P-gp cDNA were cloned into the C. elegans ubiquitous expression vector pS235 and injected into mutant nematode strains for multidrug resistance genes [3] . The effect on the nematodes phenotype was evaluated by an assay of heavy metal sensitivity (2.0 mM arsenite), as described [3] . Our quantitative results show that the P-gp/wt-CFTR chimera increases the resistance to arsenite when injected into the pgp-1/pgp-3 C. elegans double mutant, whereas P-gp/F508del-CFTR causes no effect. These preliminary results indicate it is possible to generate two distinct nematode phenotypes caused by each of the transgenic chimeras (P-gp/wt-CFTR and Pgp/F508del-CFTR). They also suggest that the same folding defect impairing F508del-CFTR function in CF may be responsible for the loss of heavy metal resistance function by the P-gp/F508del-CFTR chimera. These chimeras are currently under test in pgp-1/pgp-3 double and in pgp-1/pgp-3/mrp-1 triple C. elegans mutants. Analysis of these strains by RT-PCR showed that the mRNAs from the two P-gp/CFTR chimeras have low expression levels. This led us to produce more robust constructs, including: 1) to test chimera expression under the C. elegans endogenous pgp-1 or intestinal-specific (potent) promoters; and 2) to develop P-gp chimeras with a full CFTR-NBD1.

These C. elegans models will be used in genome-wide RNAi screens to identify genes involved in the ERQC of the P-gp/CFTR chimeras.

Work Channel gating of CFTR is regulated by ATP binding and hydrolysis at the NBDs and by PKA phosphorylation at multiple sites, primarily within the intrinsically disordered regulatory (R) region. Phosphorylation at multiple sites is required for full activation of channel function, but no one specific phosphorylation site is required. The phospho-regulation of R region interactions with NBD1 likely contributes to the regulation of CFTR channel function by modulating R region interactions with NBD1 at the NBD1/NBD2 dimerization interface, consistent with NBD1 crystal structures that include the regulatory extension (RE) comprising the first 20 residues of the R region (1). This regulation likely occurs via a dynamic complex where multiple segments of nonphosphorylated R region with αhelical propensity transiently engage NBD1 and are released. PKA phosphorylation disrupts R region α-helical propensity and decreases binding of each interacting segment (2) .

Regulation of the ∆F508 CFTR by PKA is altered, with the rate of channel activation by PKA 7-fold slower than for wild-type (3). We hypothesize that the dynamic interactions between the R region and NBD1 are affected by the ∆F508 mutation. To characterize the differences in R region binding between wild-type and ∆F508 NBD1 at residue-level resolution, we use nuclear magnetic resonance (NMR) techniques. Experiments are performed using either R region or NBD1 samples labeled with NMR-active nuclei and monitoring their structural changes upon the addition of unlabeled binding partner, allowing us to confirm R region binding to NBD1 at the NBD1/NBD2 dimerization interface and to identify other potential interaction surfaces on both the R region and NBD1. Together, these experiments will allow us to characterize the dynamic associations between the R region and NBD1 in order to better understand the molecular basis for the pathogenesis associated with ∆F508 CFTR.

Funded by the Canadian Institutes for Health Research, the Canadian Cystic Fibrosis Foundation, and the US Cystic Fibrosis Foundation 1. Lewis, H.A., et al, EMBO J. 23, 1-12 (2004) . 2. Baker, J.M.R., Hudson, R.P., Kanelis, V., Choy, W-Y, Thibodeau, P.H., Thomas, P.J., and Forman-Kay, J.D. submitted.

The ∆F508-CFTR mutation, the most common gene mutation in cystic fibrosis (CF), results in diminished plasma membrane expression of CFTR, leading to loss of functional CFTR and altered mucociliary clearance. This impairment promotes chronic infection of CF patients by the opportunistic pathogen, Pseudomonas aeruginosa. We previously reported that a secreted protein from P. aeruginosa (CFTR Inhibitory Factor (CIF)) reduces the wt-CFTR and ∆F508-CFTR-mediated chloride secretion and the plasma membrane expression of CFTR by decreasing endocytic recycling of CFTR. The aim of the current study was to investigate the mechanism by which CIF is secreted by P. aeruginosa and delivered to human airway epithelial cells (CFBE41o-). In this study, we show that CIF is packaged in outer membrane vesicles (OMV) for secretion from P. aeruginosa. Interestingly, CIF secretion via OMV was increased when the bacteria were exposed to airway epithelial cells. Changes in CIF protein level or bacterial growth did not account for the increase in CIF secretion. Purification and application of P. aeruginosa OMV to airway epithelial cells caused a decrease in plasma membrane expression of CFTR. The decreased plasma membrane CFTR expression was followed by increased ubiquitination and lysosomal degradation of CFTR. CIF delivery to the airway epithelial cells via OMV showed a more robust decrease in plasma membrane CFTR expression, in comparison with CIF delivery as purified protein. To investigate the mechanisms whereby CIF enters epithelial cells, Optiprep gradient fractionation experiments were conducted in polarized airway epithelial cells. In airway epithelial cells treated with P. aeruginosa OMV, CIF was associated with airway cell lipid rafts. Pharmacological inhibition of lipid raft formation with Filipin III blocked the decrease in CFTR plasma membrane expression upon treatment with OMV. These studies demonstrate that CIF is released by P. aeruginosa via OMV and enters the airway epithelial cells through lipid rafts. Upon entry into the epithelial cell, CIF increases the ubiquitination and degradation of CFTR and redirects CFTR trafficking from the recycling pathway to the lysosomal-degradative pathway. These data suggest that chronic infection of P. aeruginosa in the CF lung may reduce the efficacy of therapeutics developed to increase plasma membrane expression of the mutant ∆F508-CFTR (supported by the NIH 5 RO1HL074175-06 and 5 T32-DK007301-29). 1 1. Physiology, Dartmouth Medical School, Hanover, NH, USA; 2. Microbiology and Immunology, Dartmouth Medical School, Hanover, NH, USA We have identified and cloned a secreted protein from Pseudomonas aeruginosa (P. aeruginosa PA14 and clinical isolates), designated CFTR inhibitory factor (CIF). CIF reduces the apical membrane expression of CFTR and inhibits the CFTR-mediated chloride ion secretion by human airway epithelial cells expressing wt-CFTR and dF508-CFTR. CIF does not have general effects on protein trafficking, as the localization and expression of gp114, Na + -K + -ATPase, and the transferrin receptor were not affected. CFTR is a member of the ATP-binding cassette (ABC) transporter superfamily. ABC proteins are ATP-dependent transporters involved in exporting a wide variety of cytotoxic agents across the plasma membrane. P-glycoprotein (Pgp; ABCB1, MDR1 gene product), also an ABC transporter, is one of the major drug-efflux pumps expressed in normal tissues, as well as in many human cancers. Over-expression of Pgp results in reduced intracellular drug concentration and reduced cytotoxicity, conferring multi-drug resistance (MDR) to cancer cells. The aim of the current study was to examine whether CIF also affects Pgp expression in the plasma membrane, which could be exploited as a potential therapeutic strategy for restoring the sensitivity to Pgp-transported drugs in cancer chemotherapy. CIF significantly reduced the apical expression of Pgp in MDCK cells stably transfected with GFP-tagged Pgp (MDCK-GFP-MDR1 cells) and in Caco-2 and Calu-3 cells, which express endogenous Pgp in the apical plasma membrane. The drug sensitivity of MDCK-GFP-MDR1 cells to doxorubicin, a Pgp substrate, was evaluated in the absence and presence of CIF by measuring the EC 50 . CIF reduced the cytotoxicity of doxorubicin by 100-fold. Neither CIF nor vehicle alone had a cytotoxic effect. The drug sensitivity of the parental cell line, MDCK-C7, that expresses 4-fold less Pgp than the transfected MDCK cells, to doxorubicin was also increased by 20-fold. By contrast, CIF had no effect on the EC 50 values of Cisplatin (a substrate of MRP2) and Etoposide (a substrate of MRP1), suggesting that the alteration of the sensitivity to doxorubicin was due to decreased apical expression of Pgp, but not other ABC transporters. These results suggest that although CIF may be an obstacle to therapeutic attempts to restore the apical expression of CFTR in the CF lung, it could be useful for the development of a novel class of inhibitors of Pgp aimed at increasing the sensitivity of tumors to chemotherapeutic drugs (Supported by the NIH (5 RO1 HL074175)).

Hamai, H. 1 ; Keyserman, F. 1 ; Worgall, T.S. 1, 2 1. Pathology, Columbia University, New York, NY, USA; 2. Pediatrics, Columbia University, New York, NY, USA Cystic fibrosis (CF) is characterized by an inflammatory state and susceptibility to chronic lung infections. A common clinical finding is dyslipidemia characterized by altered plasma free fatty acid patterns, low plasma HDL levels and increased phospholipase A2 (PLA2) activity. It is not clear how CFTR mutations relate to lipid abnormalities. It was shown that defective CFTR is associated with decreased uptake of sphingolipids (SL). SL homeostasis is tightly regulated and metabolites are relevant to lipid metabolism and CF. SL synthesis (SLS) correlates with activity of SREBP, a key transcription factor of lipid metabolism; ceramide-1-phosphate, a SL metabolite, is a regulator of PLA2 that is highly activated in CF. We investigated the hypothesis that SLS is increased in CF. Experiments were carried out in airway epithelial cell lines that express defective CFTR (C38 / IB3), no CFTR (16HBE sense / antisense) and overexpress CFTR (A549 cells infected with AdCFTR or AdControl). SLS was assessed using radioactive tracers 3H-serine (de-novo SLS) and 3H-sphinganine (recycling pathways). Ceramide and sphingomyelin mass were determined enzymatically. Neutral and alkaline sphingomyelinase activity was assessed using 3H-sphingomyelin as substrate. SREBP mediated gene transcription was assessed using SRE-promoter assays. Cholesterol synthesis was evaluated by incorporation of 3H-acetic and 3H-mevalonic acid. Cholesterol mass was determined by gas-chromatography. ABCA-1 expression was determined by western blot analysis. PLA2 activity was measured using fluorescent phospholipid substrates. SLS was post-transcriptionally increased in cells expressing defective or no CFTR through the de-novo pathway (133 % ± 15% for IB3; 78%± 12% for 16HBE-antisense; p<0.05 ) and the recycling pathways (38% ± 11% or IB3; 27% ± 12 % for 16HBE-antisense; p<0.05). Ceramide mass was increased by 50 % (± 18 %) in 16HBE antisense, 36 % ± 9% in IB3 cells. Sphingomyelin mass was increased 2-fold in IB3, up to 3-fold in 16HBE antisense cells. Neutral and alkaline sphingomyelinase activity was increased up to 3-fold in both cell lines. SRE-mediated gene expression was increased up to 2-fold in IB3 and by 50 % in 16HBE-antisense cells. Overexpression of CFTR in A549 decreased SLS and SREBP activity up to 40%. Free cholesterol synthesis was increased by 60 % in IB3 and by 30 % in 16HBE antisense. Expression of the cholesterol efflux receptor ABCA-1 was decreased in IB3 and 16HBE antisense compared to controls. SL mass, SRE-mediated gene transcription and PLA2 activity were decreased following incubation with inhibitors of SLS in IB3 and 16HBE antisense cells. Conclusion: Defective CFTR or lack of CFTR expression correlates with 1) increased SLS synthesis and SL mass; 2) increased SREmediated gene transcription and free cholesterol synthesis; 3) decreased expression of the ABCA-1 cholesterol efflux receptor. Inhibition of SLS decreases SRE-mediated gene transcription and PLA2 activity. Dysfunctional SLS is a newly recognized pathway associated with defective CFTR and a possible therapeutic target in CF. The ability to make accurate, reproducible measures of mucociliary and cough clearance (MCC/CC) in CF patients is critical to assessment of new therapies designed to improve MCC/CC function, and thus decrease pulmonary infections and decline in lung function. For example, using MCC/CC methods developed in our laboratory, we showed that 2-week treatment with aerosolized hypertonic (7%) saline (HS) led to a sustained increase in MCC and improved lung function in CF patients (Donaldson et al, N Eng J Med 2006; 354:241) . The MCC/CC technique involves patient inhalation of radiolabeled particles followed by gamma camera scanning to determine the rate of particle movement from the lungs. While a few CF centers have shown the capability to measure MCC/CC, the techniques across these centers are not standardized, making comparison of results difficult. It is vital to develop standard, simplified techniques for these measures to make larger CF patient populations available for such studies.

To standardize a radiolabeled particle inhalation technique that can be easily used by other CF centers, we have compared three protocols in healthy and CF subjects. Protocol #1 was identical to that used in recent studies, and incorporates a Devilbiss 646 jet nebulizer to generate 5 µm MMAD (GSD = 2.0), aqueous droplets containing suspended Tc99m-sulfur colloid particles. Tidal volume and inhaled/exhaled flow rates were guided by a volume signal displayed on an oscilloscope. Protocol #2 was further standardized and simplified to allow use at other centers. This protocol utilized the same nebulizer, but triggered by a commercial dosimeter (Spira Electro 2, Finland) during inhalation as the subject breathed in time to a metronome (30 breaths/min) at flow rate of 0.5L/sec. Subjects controlled inspiratory flow rate by feedback from a digital flow readout on the dosimeter. Repeat measurements of MCC/CC were made in 8 healthy subjects and 5 mild CF patients to compare MCC/CC variables following radioaerosol inhalation using Protocols #1 and #2. There were no differences in particle deposition (i.e. central to peripheral ratio, or "C/P"), or clearance of particles through 90 minutes and 24 hours in either subject group. These data suggest that the new methodology (protocol #2) will be useful and easily employed by other CF centers to test effectiveness of new therapies for CF.

The third protocol (#3) was then tested in five of the healthy subjects. This protocol used a nebulizer that generated very large droplets (9.5 um MMAD, GSD =2.0) and subjects inhaled at very slow flow rates (~80ml/sec). It was designed to provide greater bronchial airway deposition and thus an "improved signal" for MCC/CC evaluation. Indeed, this methodology produced a larger average C/P and significantly greater particle clearance from the lung compared to the other standardized methodology (protocol #2) (C/P = 1.77 vs. 1.60; 90min clearance = 25% vs. 18%; 24hr clearance = 54% vs. 35%, respectively). Interestingly, the greater fraction of clearance between 90min and 24hr may also reflect heightened ability to assess clearance from smaller airways, which could be particularly important in CF. Protocol #3 will be tested further in CF patients. Supported by CF Foundation.

Previous clinical trials in cystic fibrosis (CF) indicate that anti-inflammatory therapy probably will not result in improvement in lung function, but will slow the rate of decline. This imposes constraints on study design for new anti-inflammatory agents, requiring that they use many patients over long periods. It is highly desirable to design a strategy for evaluation of antiinflammatory agents that will allow for the selection of only the most promising agents for Phase III trials. Sputum induction (SI) samples lower respiratory tract secretions and permits measurement of inflammatory markers. The purpose of this study was to assess the measurement of inflammatory markers in induced sputum as one such strategy by using ibuprofen as the test agent because it has demonstrated clinical benefit in CF. If changes in inflammatory markers are detectable in ibuprofen treated patients, then SI might serve as a quick, noninvasive method by which to select the most promising anti-inflammatory agents for further study. Methods: In this twoarm (ibuprofen and no treatment), open-label, parallel group study, 114 CF patients >/= 10 years of age with mild to moderate lung disease were screened at 16 sites. SI was performed on days 0, 14, 42 and 56. Patients in the ibuprofen group received drug on days 14-42. 82 subjects met eligibility criteria. 40 were randomized to ibuprofen and 40 to no treatment. 63 subjects (32 ibuprofen and 31 no treatment) completed the study through day 42 and comprise the intent-to-treat efficacy population. A maximum of 60 subjects (32 ibuprofen and 28 no treatment) and a minimum of 47 subjects (23 ibuprofen and 24 no treatment) across all markers had acceptable inflammatory marker values at both Days 14 and 42. Within group and between group comparisons were made. Results: With respect to the within group comparisons (before and after ibuprofen), most inflammatory markers slightly decreased in the ibuprofen group and increased in the no treatment group. For ibuprofen patients, mean changes were most apparent for IL-6 (-0.13 log pg/ml, p=0.04) and % neutrophils (PMNs) (-4%, p=0.073). With respect to the between group comparisons (ibuprofen vs. no treatment), differences were strongest for IL-6 (0.03 log pg/ml, p=0.0237), percent PMNs (7%, p=0.085), and absolute PMN count (0.14 cells/ml, p=0.166). Fourteen days after discontinuing ibuprofen, inflammatory mediators tended to increase. There was no difference in AEs between the two groups, and SI was generally safe. Conclusions: 1. Overall, SI is well-tolerated in patients >/= 10 years, 2. A 4-week treatment period may not be long enough to study anti-inflammatory drugs, 3. With respect to changes in inflammatory markers associated with anti-inflammatory therapy, the effect size is small and the variance of the effect is small in clinically stable patients. 4. Measuring inflammatory markers in induced sputum has potential as a screening tool for studying anti-inflammatory drugs in CF, 5. Further studies of SI in CF are warranted. Supported by Cystic Fibrosis Foundation Therapeutics, Inc.

In preparation for our gene therapy clinical trial programme we are currently assessing a number of sputum biomarkers including viscosity and elasticity, total solids and DNA content and 24 hr sputum weight. We tracked and correlated these biomarkers in CF patients (12 years and over) during a course of IV antibiotics (Ab) by collecting samples on several occasions [visit (V)1: at the start of Ab treatment, V2: at the end of Ab treatment (generally after 2 weeks) and interim periods]. To ensure adequate reproducibility of the results viscosity/elasticity measurements were carried out in triplicate using a CSL 100 Rheometer which required a comparatively large volume (5 mls) of spontaneously expectorated sputum. Because of this requirement paired samples could only be obtained from approximately 50% of the patients. There was no change in viscosity/elasticity (n=13), solid or DNA (n=15) content when comparing samples at the beginning and end of IV Ab. In contrast, 24 hr sputum weight was significantly (p<0.05) lower at the end of the Ab treatment (visit 1: 57.74±14.58 g, visit 2: 49.41±24.01 g, n=16). There was a strong and significant correlation between solid content and viscosity/elasticity (r=0.8, p<0.0001) and a modest correlation between 24 hr sputum weight and % predicted FEV1 (r=-0.42, p<0.05) and 24 hr sputum weight and patient scored symptom severity (r=0.48, p<0.001). Surprisingly, sputum DNA content did not correlate with viscosity/elasticity, despite being generally thought of as a contributor to viscosity, which may in part be related to the assay not being able to discriminate between free and cell-enclosed genomic DNA.In summary, after a course of IV antibiotics which lead to significant subjective and objective (FEV1) improvement the overall quantity of expectorated sputum was significantly reduced but, based on analysis available to date, none of the other parameters (viscosity, elasticity, solid and DNA content) changed significantly. Considering the difficulties we encountered in collecting sufficient sputum during this period of an exacerbation, sputum viscosity/elasticity measurements may not be feasible parameters to measure in a gene therapy trial to which stable patients re likely to be recruited. This study was funded by the CF Trust. Chronic neutrophilic inflammation is a feature of cystic fibrotic (CF)related inflammation and serves as a significant contributor to morbidity and mortality associated with the condition. Recently, our group has described a novel collagen-derived neutrophil chemoattractant, proline-glycine-proline (PGP), in a murine model of lipopolysaccharide-induced inflammation. The purpose of this study was to explore both the presence of this peptide in sputum from CF patients and the determination of a proteolytic system involved in PGP generation. We show that lower airway secretions from CF subjects have 30-fold increase in PGP levels compared to normal controls and that CF sputum (p<0.05) is able to generate PGP from intact type I and II collagen 100-fold above that seen from normal control sputum ex vivo (p<0.01). We further demonstrate that production of PGP is dependent on two specific matrix metalloproteases (MMPs), MMP-8 and MMP-9. Finally, we show that the generation of PGP can be significantly inhibited (80-90% inhibition) by the use of either MMP-8 or MMP-9 specific inhibitors (p<0.01 versus no inhibitor); a nonspecific MMP inhibitor (doxycycline) also demonstrates significant attenuation of PGP production. The determination of the requirements of these proteases in PGP generation allows for the identification of logical targets for disease-modifying therapeutics in CF. 1 1. CFRC, UAB, Birmingham, AL, USA; 2. CFRC, University of North Carolina, Chapel Hill, NC, USA Cystic Fibrosis (CF) lung disease is characterized by chronic neutrophilic inflammation. The discrete collagen breakdown product prolineglycine-proline (PGP) is a potent neutrophil chemoattractant thought to be generated by a specific proteolytic cascade present during CF pulmonary exacerbation (Nat Med 12(3):317, 2006) . High-mobility group box 1 (HMGB1) is a potent inflammatory mediator found in sepsis, rheumatoid arthritis, and other inflammatory diseases characterized by neutrophilic inflammation. The role of PGP and HMGB1 as PMN attractants in CF is unknown. In this study we utilized specimens derived both from humans and Scnn1b-transgenic mice (βENaC mice) in which overexpression of the βENaC subunit in the airways leads to sodium hyperabsorption and airway surface liquid depletion, mimicking the CF muco-obstructive phenotype. PGP was quantified using tandem MS. Neutrophil chemotaxis was measured in vitro following preincuabtion of sputum or BALF with anti-HMGB1 neutralizing antibody. Murine BALF was assayed for HMGB1 by Western blot and ELISA. Human sputum was similarly evaluated after normalizing for total protein concentration. In vivo HMGB1 activity was measured after intratracheal instillation in wt mice. βENaC mice had 60% higher cell counts and a greater percentage of PMNs (37.3% vs. 0.6%) in BAL samples than wild-type (WT) littermates (p<0.005, n=15/genotype). PGP was elevated in BALF of Scnn1b animals (48.0 vs. below limits of detection in wt controls, p<0.001, n=11), increased in serum from CF subjects (834 vs. 340 pg/mL normal controls, p<0.005, n=5), and detected at high levels in CF sputum (54,700 pg/mL). BALF screened for HMGB1 by Western blot revealed elevated levels in βENaC mice (p<0.05 by densitometry, n=10/genotype); ELISA confirmed 52% greater HMGB1 concentration (61.8 vs. 40.6 ng/mL, p<0.02). HMGB1 was also detected by Western blot in human CF sputum (19 of 23) and at higher levels than secretions from normal subjects (p<0.05 by densitometry; n=23 CF, 4 control). Purified HMGB1 induced dose-dependent PMN chemotaxis in vitro (peak 50-250 ng/mL, p<0.05, n=5) and CF sputum caused potent chemotaxis (7-fold over control, p<0.005, n=12) that was inhibited by preincubation with HMGB1 antibody (p<0.005). BALF from βENaC mice (but not wt controls) was also chemotactic (2-fold greater than media control, p<0.01, n=5) and inhibited by anti-HMGB1 antibody (p<0.05). Exogenous HMGB1 (25 ng) administered intratracheally to LPS resistant mice (C3HeJ, Toll-4 receptor mutant) and BALB/c mice caused PMN influx in BALF at 24 hours (C3HeJ: 9.0 x 10 5 PMNs after HMGB1 vs. 2.0 x 10 5 after vehicle, P<0.02; BALB/c: 3.5 x 10 5 vs. 0.8 x 10 5 , P<0.01), and generation of PGP (189.7 vs. 80.7 pg/mL in C3HeJ, P<0.05, n=3) . In summary, PGP and HMGB1 are elevated in BALF of βENaC mice that present with CF-like lung pathology; in CF airway secretions, HMGB1 is present in vivo at concentrations that induce neutrophil chemotaxis in vitro and PGP production in vivo, and is inhibited by blocking antibody. The role of PGP and HMGB1 in CF deserve further evaluation as they may be potential therapeutic targets of dysregulated inflammation.

Esther, C.R. 1 ; Jasin, H.M. 2 ; Collins, L.P. 3 ; Boysen, G. 3 ; Boucher, R.C. 2 1. Pediatric Pulmonology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 2. CF Research Center, Univeristy of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 3. Center for Environmental Health and Susecptibility, Univeristy of North Carolina at Chapel Hill, Chapel Hill, NC, USA Biomarkers of airway inflammation are needed in cystic fibrosis (CF) to aid clinical management and assess the efficacy of novel therapies. We describe a simple and non-invasive method to measure airway purines, signaling molecules previously shown to be biomarkers of neutrophilic airway inflammation.

Airway secretions were obtained using exhaled breath condensate (EBC) collection, a technique that was easily performed in the outpatient setting on children as young as three years. The purines adenosine and AMP in EBC were measured using ultra-sensitive liquid chromatography/tandem mass spectrometry (LC/MS/MS). In addition, we used LC/MS/MS to simultaneously measure urea as a dilution marker to control for the known dilutional variability of airway secretions within EBC.

Detection of adenonosine, AMP, and urea was optimized using positive mode MS with selected reaction monitoring, and stable isotope dilution was utilized to improve quantification. Detection was linear with concentration over four orders of magnitude, with coefficients of variation <15%. The method was sensitive, with limits of detection for adenosine and AMP (~0.1 nM) and urea (~0.5 µM), below expected EBC concentrations. Applying the methods to EBC demonstrated that detection of purines and urea was reliable, with intraclass correlation coefficients greater than 0.9 between duplicate measures (n=7). We also measured purines and urea in EBC samples collected prospectively from children with CF (n=17) and healthy controls (n=6) during regular clinic visits. All samples were lyophilized and reconstituted to increase concentration 10-fold. Adenosine, AMP, and urea were detected in all samples. Neither adenosine nor AMP levels differed between groups, but urea was significantly lower in CF (6.6 ± 2.3 µM) versus healthy control (15.1 ± 4.9 µM, p=0.02), suggesting that airway secretions were more dilute in CF samples. When dilutional variability was controlled using purine to urea ratios, the AMP to urea ratio was found to be elevated in subjects with CF (0.50 ± 0.14) compared to healthy controls (0.09 ± 0.03, p=0.015), consistent with previous findings. The AMP to adenosine ratio was also elevated in CF (4.3 ± 1.9) compared to control (0.87 ± 0.30), although the difference did not reach statistical significance in this small sample set (p=0.19). Experiments are ongoing to increase sample size and assess EBC purine levels before and after treatment of a CF exacerbation.

These results demonstrate that LC/MS/MS analysis of EBC provides a non-invasive method to measure purine biomarkers of CF airway inflammation. Given the flexibility of MS, this methodology could also prove applicable for study of other small molecule biomarkers. Importantly, we also show that dilution of airway secretions in EBC may be altered in CF, and LC/MS/MS can be used to control for this variability through measurement of urea as a dilution marker.

Introduction: Mucociliary clearance and antimicrobial peptide activity may be affected by airway surface liquid (ASL) pH, which is regulated by epithelial ion transport. ASL pH is abnormal in disease states in which IL-17A is elevated, including cystic fibrosis (CF), and the IL-17A receptor is expressed at the basolateral surface of bronchial epithelium. Therefore, we investigated the effects of IL-17A on vectorial ion transport in well-differentiated HBE cells. Methods: HBE cells were grown with an apical air interface and incubated with IL-17A at 0 or 50 ng/ml in the basolateral medium for 48 hours prior to being studied with standard short-circuit current (Isc) techniques. Results: IL-17A treated cells had a minimal increase in resting and amiloride-sensitive Isc compared to control cells, and had a dosedependent, statistically significant increase in forskolin-stimulated Isc. At a dose of 50 ng/ml, IL-17A doubled forskolin-stimulated Isc (10.5 ± 1.72 µA/cm2 for control vs. 20.3 µA/cm2 for IL-17A). The increased Isc was not bumetanide-sensitive, but was sensitive to acetazolamide and DNDS, and was not present in HCO 3 --free solutions, suggesting it was due to HCO 3 secretion. To investigate whether this HCO 3 secretion was Cl --dependent, we studied HBE cells in Cl --free solutions, and there was no difference between untreated and IL-17A treated HBE cells. To test the hypothesis that IL-17A promoted Cl -/HCO 3 exchange, we performed experiments in HBE cells mounted in Cl --free solutions, and after addition of amiloride and forskolin, 30 mM NaCl was added to the mucosal bath and 30 mM NaGluconate was added to the serosal bath. Under these conditions, one would predict a decrease in Isc due to mucosal to serosal Clmovement down its electrochemical gradient. In untreated HBE cells, we saw a decrease in Isc followed by a recovery. In IL-17A treated cells there was a sustained increase in Isc, suggesting that movement of Cldown its electrochemical gradient resulted in exchange of Clfor an anion at the apical membrane. In the absence of HCO 3 -, both untreated and IL-17A treated cells responded to Claddition to the mucosal bath with decreases in Isc. To assess the CFTR dependence of the IL-17A-induced Isc, we stimulated primary CF HBE cells treated with IL-17A. In CF cells, there was a minimal increase in amiloride-sensitive Isc (30.6 ± 3.35 µA/cm2 for control vs. 37 ± 8.24 µA/cm2 for IL-17A ) and in forskolin-stimulated Isc (1.2 ± 0.56 µA/cm2 for control vs. 3.4 ± 1.16 µA/cm2 for IL-17A ). Conclusion: The proinflammatory cytokine IL-17A induces Cl -/HCO 3 exchange in HBE cells. Notably, the observed Cl -/HCO 3 exchange appears to be CFTR-independent. However, CFTR activation is required for maximum Cl -/HCO 3 exchange as demonstrated by the smaller Isc generated in CF cells. These data suggest that the cytokine milieu of the airway epithelium can alter ion transport and potentially ASL physiology. Furthermore, they suggest that CFTR may interact with or regulate novel proteins in the presence of inflammation.

Supported by CFF and NIH Periciliary fluid balance is maintained by the coordination of sodium and chloride (Cl-) channels in the apical membranes of the respiratory epithelia. In the absence of the cystic fibrosis transmembrane conductance regulator (CFTR), Cl-secretion is diminished and sodium reabsorption becomes exaggerated. Activation of non-CFTR-dependent Cl-channels can provide an alternate pathway for Cl-secretion in the airways and gastrointestinal tract. The pH and voltage-dependent type-2 Cl-channel (ClC-2) is expressed in airway epithelial cell luminal membranes. We hypothesize that topically applied ClC-2 agonists may restore Cl-secretion in cystic fibrosis (CF) murine airways. Using in vivo nasal potential difference measurements, we quantified ClC-2-mediated Cl-transport in both CF and wild-type mice during nasal perfusion with lubiprostone (a prostone compound and specific ClC-2 agonist; Sucampo, Pharmaceuticals, Inc., Bethesda MD) or vehicle control. Wild-type (C57Bl6, n = 9) and CF knock-out (CFKO; n=10(JAX stock# 002364)(Jackson Laboratories, Bar Harbor, ME) mice were sedated and intubated to protect the lower airways from aspiration of perfusate. Nasal and subcutaneous bridges were connected to fluid-filled silver chloride electrodes. Baseline (in Ringer's), amiloride-inhibited, low Cl-(Cl-free, gluconate-substituted Ringer's with amiloride), and low Cl-plus lubiprostone (with increasing concentrations of lubiprostone or vehicle) mice were perfused and potential differences were measured with a high impedance voltmeter (World Precision Instruments, Sarasota FL) and continuously recorded. A clear dose-response relationship was detected in both wild-type and CF mice. At 20 µM lubiprostone, wild-type mice showed hyperpolarization of -7.8 ± 3.4 mV and CF mice responded with -4.7 ± 1.5 mV hyperpolarization. A paired t-test of low Cl-perfusion and lubiprostone perfusions revealed significant (P<0.001) differences in both genotypes. Five ClC-2 CFKO mice were similarly tested and showed no response to lubiprostone (+1.1 ± 1/5 mV). CFTR inhibitor-172(50 µmol) (Calbiochem, San Diego CA) added to the low Cl-perfusion in 6 additional wild-type mice eliminated the low Cl-response but did not abolish the lubiprostone response, confirming that ClC-2 is present and independent of CFTR regulation. We conclude that direct application of a ClC-2 agonist in the CF murine upper airways restores near normal levels of Cl-secretion. CFTR is an apical membrane chloride channel whose activity is required to maintain airway surface liquid (ASL) volume and efficient mucociliary clearance. CFTR is activated by cAMP dependent stimulation of protein kinase A. However, studies have suggested that the actin cytoskeleton may also be required for regulation of CFTR, by acting as a structural element in second messenger compartmentalization (1) . To determine whether an intact cytoskeleton is required for ASL volume homeostasis, adenosine (ADO), ATP, or isoproterenol (Iso) (all at 200 µM) were added apically to Human Bronchial Epithelial Cultures (HBEC's) before or after the cytoskeleton had been disrupted by cytochalasin D exposure (5µM for 1h). In control cultures ADO, ATP and Iso all resulted in a significant increase in ASL height (∆ASL) within 10 min (∆ASL; ADO, 6.2±1.2µm; ATP, 7.2±2.3µm; Iso, 8.3±1.3µm). However, in cytochalasin D exposed cultures ADO was without effect while ATP and Iso were still able to evoke an increase in height (∆ASL; ADO, -0.7±1.2µm; ATP, 7.4±3.1µm; Iso, 10.2±3.1µm). However, we showed that A2B receptor function was not compromised by cytoskeleton disruption as cAMP levels were seen to increase both after addition of adenosine and adenosine in the presence of cytochalasin D. Thus, the effect of cytochalasin D on ASL height appeared specific for adenosine-mediated Cl-secretion. To further test the effect of cytochalasin D on the relationship between adenosine and CFTR, we designed a fluorescent resonance energy transfer (FRET) pair of cfp-CFTR (labeled at the n-terminus) and yfp-A2B receptor (labeled at the c-terminus). Addition of adenosine (100µM) increased FRET by 2.3 times (% efficiency; control, 15.2±2.3%, ADO, 35.3±3.5%) compared to the vehicle control (n=13). Cytochalasin D exposure (5 µM for 1 h) completely inhibited the increase in FRET that was observed after adenosine addition (p<0.01). In contrast, cytochalasin D had no effect on FRET between CFTR and the β2adrenoceptor. To determine whether cytochalasin D treatment (5 µM for 1 h) had any direct effect on CFTR protein levels and localization we used a cell line stably expressed with an exotope-tagged CFTR (HB8 cells) (2) . CFTR protein levels were found to be significantly reduced as compared to nontreated controls by both Western blot analysis and chemiluminescence detection (n=3). Cells were also stained with an antibody against the external CFTR exotope and imaged by confocal microscopy and a significant decrease in staining at the plasma membrane was observed after cytochalasin D treatment (n=3). We conclude that cytochalasin D exposure decreases CFTR protein levels and specifically inhibits adenosine-mediated regulation of CFTR activity in the airways, leading to disruption of ASL homeostasis and inhibition of mucociliary clearance. Numerous studies have reported evidence that CFTR is expressed in the apical membrane of serous cells in the airway submucosal glands. Thus, it has been reasoned that (1) CFTR plays some role in normal anion and fluid secretion by airway glands and (2) loss of this channel's function in cystic fibrosis (CF) airways plays some role in the etiology of CF lung disease. A recent study by Joo et al. (JBC 277:50710-50715, 2002) indicates that CF airway glands lose the ability to secrete fluid when treated with forskolin or VIP but maintain the ability to secrete when treated with muscarinic agonists. This result suggests that VIP and muscarinic agonists mediate secretion by CFTR-dependent and CFTR-independent pathways, respectively. The present study was undertaken to explore the dynamics of fluid secretion by individual submucosal glands to determine if further differences in their responses to these two agonists could be distinguished. Pig tracheas were obtained from a local slaughterhouse. After removal of the cartilage, tissues were mounted in a warm (37°C) observation chamber that permitted exposure of the submucosa to Krebs solution. The mucosal surfaces of the airways were dried with a stream of warm dry air and then layered with mineral oil. Liquid secreted from the gland duct openings formed spherical aqueous droplets within the oil layer, and their volumes were estimated from spatial dimensions taken from sequential digital images. Liquid secretion rates were estimated for individual glands present within a 28.9 mm 2 region of interest. Rates were determined for two experimental protocols. For one group of tissues, secretion rates were determined for a control period, followed by a period of exposure to 1 µM VIP, and then a period of exposure to both VIP and 10 µM acetylcholine (ACh). In the second group, the order of VIP and ACh treatment was reversed. These agonist concentrations were expected to produce near maximum rates of secretion. When the agonists were applied first, secretion rates were significantly (P<0.05) increased from the control period by both ACh (2.01±0.44 nL/min to 4.52±0.38 nL/min) and VIP (2.26±0.63 nL/min to 4.64±0.78 nL/min). The secretion rates for individual glands were highly variable with both agonists (ACh: 0.11-19.20 nL/min; VIP: 0.03-32.21 nL/min). When ACh application followed VIP treatment, the mean secretion rate was significantly increased further to 6.13±0.87 nL/min; however, when VIP application followed ACh treatment, the secretion rate was only marginally and insignificantly increased to 4.97±0.48 nL/min. We conclude that both ACh and VIP are efficacious secretagogues for porcine airway glands and that the rate of secretion from individual glands to either agonist is very heterogenous. Because VIP has no effect following stimulation by ACh, we speculate that ACh induces secretion by activating all transporters relevant to VIP-induced secretion including CFTR. However, ACh must also activate channels and/or transporters not induced by VIP since its effects are additive to VIP. This most likely includes activation of a non-CFTR anion channel. ( Serous cells are thought to serve as the principal anion and fluidsecreting cells of airway submucosal glands. Mucous cells primarily secrete gel-forming mucins. These respective physiological roles for these two cell types imply that the mucous cells, which secrete the thick mucus gel, should lie downstream of the serous cells so that their watery secretions can flush the mucins out of the ducts and onto the airway surface. Early studies indicated that such arrangements existed within glands but reports of cell distributions have been descriptive rather than quantitative. Because of resurgent interest in serous cell function, we revisited this issue to formally document the distribution of these cell types within the secretory tubules of the submucosal glands as well as at the surfaces of the glands, where serous cells are most accessible for study. A piece of porcine lung, containing a 5mm diameter bronchus, was treated with formalin fixative, and 250 consecutive 5µm sections were taken. The apices of virtually all gland epithelial cells, except the cells lining the collecting and ciliated ducts, stain positive for PAS indicating that these cells are either serous or mucous cells. However, PAS staining does not allow distinctions to be made between these two cell types. Consequently, all slide sections were stained with hematoxylin and eosin. With these stains, mucous cells were identified as cells containing lucent granules in their apices whereas serous cells were considered to be cells where the apical cytoplasm stained a uniform dark reddish color. First, the distribution of serous and mucous cells in the secretory tubules were determined. The acini of individual tubules were located in the slide sections, and cells were identified and counted in consecutive sections that approached the collecting ducts until the tubules were no longer distinguishable as discrete tubes. Secretory tubules from 6 individual glands were studied. In the acini, serous cells accounted for 95.0±3.2% of total cells whereas mucous cells accounted for 3.8±2.2% of total cells (remaining cells were not distinguishable as either cell type). There was a significant (P<.05) negative correlation of serous cell numbers with distance from the acini so that at 35µm from the acinus serous cells accounted for only 73.4±13.5% of total cells. There was a coincident significant positive correlation with mucous cell numbers so that at 35µm from the acini mucous cells accounted for 20.1±10.5% of total cells. Next, we examined the cell distribution at the adventitial and mucosal surfaces and at the lateral margins of the glands. Serous cells accounted for 82.7%, 85.4%, and 96.9% of the total cells at these respective locales. Mucous cells represented 15.7%, 10.1%, and 2.4% of total cells in these same respective regions. We conclude that serous cells are by far the dominant cell type in secretory tubules of porcine submucosal glands. Mucous cells are rare in acini but increase in frequency with distance from the acini. The outer surfaces of the glands are dominated by serous cells as well. At the lateral margin of the glands, serous cells outnumber mucous cells by 20-to-1. (Supported by NIH HL63302) . Many of the membrane transporters that participate in this process have been identified; however, the identity of the class of Kchannels that maintains resting membrane potential and/or cell membrane polarization during the secretion process remains poorly characterized. Liquid secretion by porcine submucosal glands is insensitive to numerous K + channel blockers including Ba 2+ , tetraethylammonium, apamin, charybdotoxin, iberiotoxin, clotrimazole, penitrem A, 4-aminopyridine, and quinidine. Recently, a new class of K + channels, the "tandem-pore" or K 2P channels, have been described. Potassium channels of this class are insensitive to many traditional K + channel inhibitors but are blocked by local anesthetics, such as lidocaine and bupivacaine. Additionally, many K 2P channels are sensitive to changes in extracellular pH. In the present study, we considered whether K 2P channels might participate in the liquid secretion response to acetylcholine (ACh). Intrapulmonary bronchi, excised intact from the lungs of pigs, were cleared of accessible luminal liquid, cannulated, and treated with 10 µM ACh to induce secretion. Paired airways were pretreated for 1 h with 3 mM bupivacaine. Bupivacaine blocked 90.4±1.8% of the liquid secretion response to ACh. The IC 50 for the bupivacaine inhibition was approximately 220 µM. The inhibitory effect of bupivacaine was not due to nonspecific toxicity since tissues exposed to bupivacaine for 1 h and then washed with fresh inhibitor-free buffer recovered to 80% the control AChinduced secretory rate. The bupivacaine effect was not due to inhibition of voltage-gated Na + channels in intrinsic neurons since 1 µM tetrodotoxin did not inhibit ACh-induced secretion. No significant effect was seen in the secretion rate when the extracellular solution pH was lowered from 7.43±0.01 (normal HCO 3 --buffered Krebs gassed with 5% CO 2 ) to 6.60±0.01 by gassing with 40% CO 2 gas (balance O 2 ). While we recognize the relative nonspecificity of this agent, the inhibitory effects of bupivacaine on liquid secretion by porcine bronchi are consistent with K 2P channel participation in this secretory response. The failure to demonstrate inhibition with acidification of extracellular solution could signify a K 2P channel subtype that is insensitive to low extracellular pH. ( 4 1. Novartis, Horsham, United Kingdom; 2. Rosalind Franklin University, Chicago, IL, USA; 3. Mount Sinai Medical Center, Miami, FL, USA; 4 . Genomics Institute of the Novartis Foundation, La Jolla, CA, USA ENaC activity in the human airway epithelium has been reported to be partially-sensitive to the broad spectrum trypsin-like protease inhibitors aprotinin and placental bikunin (Bridges et al., 2001 Am J Physiol 281:L16-23; Donaldson et al., 2002 JBC 277:8338-45) . A low molecular weight inhibitor of the airway Channel Activating Protease (CAP) would represent a potential therapeutic approach to attenuating ENaC function in cystic fibrosis (CF) lung disease.

The aims of the current study were to: (1) further characterise the in vitro efficacy of CAP inhibitors in human bronchial epithelial cells (HBEs), (2) to evaluate the relevance of a CAP mechanism to the regulation of ENaC function in vivo, and (3) to establish whether the inhibition of the airway CAP can modulate mucociliary clearance in vivo.

Primary cultures of HBEs (normal and CF), cultured under air-liquid interface conditions, demonstrated an amiloride-sensitive short circuit current (ISC), that was sensitive to aprotinin but that was insensitive to inhibition by SBTI, α1-anti-trypsin and α1-PDX. The amiloride sensitive ISC was also attenuated by the low molecular weight CAP inhibitor NVP-QAU145 (HBC276) with an IC50 value of approximately 30nM. NVP-QAU145 caused a time dependent inhibition of the ISC with a T 1 ⁄ 2 of approximately 20 min and this effect could be reversed by the addition of excess trypsin.

In vivo, aprotinin attenuated the guinea pig tracheal potential difference (TPD) from -10.0±0.8mV to -4.7±0.6mV (n=4-11) with an ED50 value of 40pmoles/kg, measured at 2 hours following intra-tracheal instillation. The TPD was unaffected by either SBTI or α1-anti-trypsin at doses up to 1500pmoles/kg. The combination dosing of an ENaC blocker with aprotinin did not attenuate the TPD beyond the effect observed with either agent alone, consistent with aprotinin attenuating ENaC function in this model system. Intra-tracheal instillation of NVP-QAU145 attenuated the TPD from -10.3±1.0mV to -4.7mV with an ED50 value of 3µg/kg (n=6-8). In the sheep, the administration of NVP-QAU145 into the airways by aerosolisation of a 1mg/mL solution resulted in a 3-4 fold enhancement of the rate of clearance of 99mTc sulfur colloid from the lungs compared with the vehicle control.

These studies indicate that the CAP mechanism of ENaC regulation can translate through in vitro human assays, into in vivo models. CAP inhibitors can attenuate TPD in the guinea pig in vivo and enhance the rate of mucociliary clearance in the sheep, thereby representing an approach to the therapeutic regulation of ENaC function in CF lung disease.

Cotton, C. Pediatrics, Physiology and Biophysics, Case Western Reserve University, Cleveland, OH, USA Cystic fibrosis is caused by mutations in the gene that encodes CFTR, a cAMP-activated apical membrane anion channel. Loss of CFTR-mediated anion conductance is a primary defect in CF but secondary defects such a sodium hyperabsorption are also implicated in CF pathophysiology. It is generally accepted that mucociliary clearance is compromised in CF airways due to reduced fluid secretion and/or increased fluid absorption. The recent generation of a transgenic mouse model with ENaC overexpression that exhibits CF-like lung disease highlights the importance of sodium hyperabsorption. Therapies designed to reduce mucus production and viscosity, increase lumenal water content, stimulate fluid secretion, and inhibit fluid absorption are under development for treatment of CF airways. Although controversial, several recent studies have demonstrated that SERCA pump inhibitors partially correct the trafficking defect associated with delta F508 mutant CFTR. The goal of this work is to determine if SERCA pump inhibitors affect ENaC-mediated sodium absorption. A renal collecting duct epithelial cell line and primary cultures of well-differentiated human tracheal epithelial cells (air/liquid interface culture) were used for these studies. Amiloride-sensitive short circuit current was determined as a measure of ENaC-mediated sodium absorption. Quantitative RT-PCR was used to evaluate expression levels for each of the 3 subunits of ENaC (alpha, beta, and gamma). Epithelial monolayers were treated for 18-24hours with SERCA pump inhibitors (thapsigargin, 100-500 nM; DBHQ,10-50 uM; and curcumin, 10-50 uM). Treatment with SERCA pump inhibitors decreased amiloride-sensitive short circuit current by 50-80%. In contrast, cAMPstimulated short circuit current was not reduced by treatment with SERCA pump inhibitors. The steady-state levels of alpha, beta, and gamma ENaC were reduced by 60-80% in renal epithelial monolayers treated with thapsigargin whereas beta and gamma ENaC were reduced by 80-90% with no significant change in alpha ENaC in airway epithelial cells. The results of these studies demonstrate that sustained depletion of endoplasmic reticulum calcium stores and/or elevation of intracellular calcium by inhibition of SERCA pump activity reduces ENaC mRNA expression and ENaC-mediated sodium absorption. Thus, inhibition of SERCA pump activity in airway epithelial cells of CF patients that carry the delta F508 mutation may provide dual benefit by promoting delivery of mutant CFTR to the membrane 1 1. Pediatrics, National Jewish Medical and Research Center, Denver, CO, USA; 2. Integrated Department of Immunology, National Jewish Medical and Research Center, Denver, CO, USA; 3. Department of Medicine, Cystic Fibrosis/Pulmonary Research and Treatment center, The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Increased cytosolic calcium ([Ca 2+ ] i ) initiated by the release of stored Ca 2+ from the endoplasmic reticulum (ER) is a key signal to elicit a wide range of essential cellular responses, including secretory functions of the respiratory epithelium that are modified in CF. SERCA pumps are responsible for (re)filling the ER Ca 2+ stores, and SERCA blockers such as thapsigargin are very potent and commonly used pharmacological triggers of Ca 2+ -signals. Modulation of the activity of SERCAs can profoundly affect Ca 2+ homeostasis. Although defective calcium homeostasis is a characteristic of several pulmonary diseases including CF, the role of SERCA is unknown. Lung tissue samples (bronchus and distal lungs) from normal (n=2) and CF subjects (n=2) were evaluated by immunohistochemistry. SERCA2 expression was decreased in the bronchial and bronchiolar epithelia of CF. Non-CF and CF bronchial epithelial cell line pairs including Calu-3 and JME/CF15, C-38 and IB3-1, 16HBE14o-(16HBE), CFBE41o-(CF41o-) and CFBE45o-(CF45o-) were also probed. Given certain limitations of such cells, several consistent findings still emerged. A 65% and 75% decrease in SERCA2 expression was observed in CF41o-and CF45ocells as compared to 16HBE cells. Immunocytochemical studies in these cells confirmed that the SERCA2 was localized in the ER and that the decreased SERCA expression was not associated with decreased ER content. Reduced SERCA2 expression and activity (3.51±0.27 vs 1.76±0.23 and 2.59±0.29 pmol/min/mg protein in 16 HBE vs. CF41o-and CF45ocells respectively) was observed in the purified ER membranes from CF cell lines. Northern blot analysis revealed a parallel reduced mRNA expression as well. Decreased SERCA2 was accompanied by increase in the low affinity isoform SERCA3 (0.95±0.20 vs. 1.39±0.10 and 1.68±0.32 arbitrary SERCA2 intensity/b-actin in respectively) . We have also evaluated a limited number of primary airway epithelial cells isolated from lung samples of normal and CF subjects for SERCA2 expression. SERCA2 expression in polarized tracheobronchial epithelial cell lysates from CF subjects was decreased by about 60% as compared to those from normal subjects (1.86±0.56 vs. 5.68±1.24 arbitrary SERCA2 intensity/b-actin units in CF and normal respectively, n=3). Expression of SERCA2 could also be suppressed by inhibiting CFTR with CFTR inh 172 in normal human bronchial epithelial cells ( Various studies indicate that the airway surface fluid possess an antibacterial system based on the combined action of lactoperoxidase, H 2 O 2 , and SCN -(thiocyanate). The enzyme lactoperoxidase (secreted by submucosal glands) utilizes H 2 O 2 and SCNto generate OSCN -(hypothiocyanite) a molecule with antimicrobial activity. H 2 O 2 is produced by dual oxidases expressed on the apical membrane of airway epithelial cells whereas SCNis transported across the epithelium through anion transporters and channels. In particular, SCNtransport seems to occur through CFTR and other Ca 2+dependent Clchannels. Therefore, reduced SCNtransport in the airways of cystic fibrosis patients may contribute to impaired antimicrobial activity.

In a recent study, we have found that cytokines, in particular IL-4, cause a strong increase in the ability of cultured bronchial epithelia to transport SCNfrom the basolateral to the apical membrane. This effect is mediated by upregulation of Ca 2+ -dependent Clchannels and of the SLC26A4 (pendrin) anion transporter. These findings suggest that under proinflammatory conditions the activity of the SCN -/H 2 O 2 /lactoperoxidase system is potentiated.

We evaluated the antimicrobial activity of airway surface by seeding bacteria on the apical membrane of human bronchial epithelia grown with an air liquid interface on a porous membrane. An inoculum corresponding to 1,000 cfu of S. aureus was added to the cells in 25 µl of saline solution with and without lactoperoxidase (6.5 µg/ml). Experiments were performed in the presence or absence of SCN -(100 µM) in the basolateral solution. Bacteria were recovered after four hours and plated on agar plates for colony counting. Our preliminary results show that simple addition of lactoperoxidase to the apical surface decreases bacterial survival. In addition, bacterial killing is strongly enhanced by prestimulation of cells for 24 hours with IL-4 (10 ng/ml). The effect of lactoperoxidase is dependent on H 2 O 2 as it is prevented by addition of catalase and is absent in cell-free experiments when H 2 O 2 is omitted from the reaction mixture. Surprisingly, the presence of SCNin the basolateral compartment did not increase bacterial killing and, in some cases, appeared to generate a protective effect.

Our results suggest that lactoperoxidase, in the presence of H 2 O 2 , is an effective antimicrobial molecule. The contribution of SCNand the mechanism of the potentiation caused by cytokines is less clear and requires further investigation. A possibility is that, in the absence of SCN -, lactoperoxidase generates an oxidant molecule that is more toxic to bacteria than OSCN -. Elucidation of this mechanism and comparison between CF and non-CF epithelia is in progress to assess the role of CFTR and of other anion channels.

Supported by CFFT and Telethon-Italy. CF patients become infected with Pseudomonas aeruginosa, which release flagellin into the airway surface liquid to activate toll-like receptor 5 and proinflammatory signaling. Flagellin has been shown to inhibit Na absorption by airway epithelia. We tested flagellin on Cl secretion and proinflammatory signaling (NF-κB activation) by Calu3 cells, a CFTRexpressing, serous-like airway gland cell line. Calu3 cells were grown on filters and either mounted in Ussing chambers (clamped to zero mV) in the presence of a serosa-to-mucosa gradient of [Cl] for measurements of transepithelial Cl secretion (I Cl ) or treated with an adenovirus expressing NF-κB-controlled luciferase to assay NF-κB activation. Flagellin (10 -7 g/ml) on either the apical or basolateral surface of cells increased apparent anion secretion that began in 3-10 mins and increased over 20-30 mins by 20-50 µA/cm 2 . This I Cl was blocked by glibenclamide and GlyH101, indicating that it resulted at least in part from activation of CFTR. Flagellin also stimulated fluid secretion by intact human tracheal glands. Flagellin activated NF-κB (luciferase assays) in both Calu-3 cells and in the CF cell line CF15, while I Cl was stimulated in Calu3 but not in CF15 cells. Flagellin-stimulated I Cl in Calu-3 cells was blocked 50% by SB202190 (a p38 MAPK blocker) and by a similar amount by wortmannin (PI3 kinase blocker). Interleukin 1β and the TLR2-agonist Pam3Cys also activated I Cl . The effect of flagellin was not due to increases in cytosolic [Ca 2+ ] (Ca i ) because flagellin did not alter Ca i . In contrast to the slow effects of flagellin, Pam3Cys and IL1β, ATP rapidly increased I Cl (within secs, up to 80 µA/cm 2 ) followed by slower decrease to steady I Cl = 20-40 µA/cm 2 that had a similar time signature as the increases in Ca i . Forskolin (to increase cytosolic [cyclic AMP]) increased I Cl within 2 mins to a steady value = 50-80 µA/cm 2 . Flagellin had small or no effects on I Cl following maximal stimulation with either ATP or forskolin. ATP and forskolin on their own had no effect on NF-κB. Flagellin increased phosphorylation of p38 MAPK and of AKT (down-stream kinase phosphorylated by PI3K) with a time course similar to the increase in I Cl . Flagellin-activated NF-κB was reduced by roughly 50% by either wortmannin or SB202190. These results indicated that TLR agonists and inflammatory cytokines stimulate both NF-κB and proinflammatory processes and also CFTR-dependent Cl and fluid secretion by airway gland cells. These responses are mediated in part by activation of both p38 MAPK and PI3 kinase. Flagellin-TLR5-activated increases in CFTR-mediated Cl secretion and reduction in ENaC-mediated Na absorption will increase fluid secretion into the airways, which may facilitate bacterial removal by the mucociliary escalator and thereby reduce the proinflammatory stimulus. In CF it is expected that bacteria will activate TLR signaling to trigger innate immune responses, but Cl and water secretion and the resulting "bacterial flush," will be missing, leading to sustained inflammation. (1, 2) . Treatment with HS improved several measures of lung function including mucociliary clearance. In one study amiloride was seen to have a negative impact on the benefit of HS (2) . The long duration of action of HS and the paradoxical effect of amiloride are surprising and require further investigation.

Aims: Our goal was to study the effect of exposure to a hypertonic challenge (HC) on amiloride sensitive Na+ transport (INa) in primary cultures of human bronchial epithelial (HBE) cells from non-CF and CF patients.

Methods: HBE cells were grown at an air liquid interface and studied under short circuit current conditions. Cells were challenged with HS by exchange of the apical or basolateral baths. In some experiments the timedependent effect of a small (30 µl) isosmotic or HS volume at the airway surface was tested on INa and apical osmolality.

Results: Application of hypertonic NaCl and mannitol solutions from either the mucosal or serosal sides of the epithelium inhibited INa. The degree of inhibition was a saturable function of the imposed hypertonicity with an IC50 of 379 (Mannitol) and 398 (NaCl) mOsmol/kg H2O and a maximal inhibition of 97%. The inclusion of amiloride (10-100 µM) did not affect the HC-induced reduction in INa. The inhibition in INa by HC was rapid in onset and accompanied by a fall in the transepithelial resistance and the total transepithelial capacitance as expected for cell shrinkage. The inhibition in INa was only partially reversible (~20%) after returning to isotonic solutions for 45 min. Exposure of the airway surface to hypertonicity (680 mOsmol/kg H2O) inhibited INa by 72-79% at 30 min of exposure. As osmotically-driven water moved from the serosal to the mucosal side, the apical volume increased concomitant with a decrease in its osmolality and a recovery of INa. The osmolality declined exponentially and reached the isosmotic value after ~4 hrs. The recovery of INa lagged behind the recovery of apical osmolality by 80, 20 and 10% at 0.5, 2 and 3 hrs of incubation, respectively. After 4 hrs of airway HC exposure, the osmolality and INa recovered completely. Our results demonstrate that HC causes a rapid, dramatic and prolonged decrease in Na+ absorption and that continuous presence of amiloride had no effect on the inhibition of INa or the recovery of apical osmolality. HC also causes the epithelium to shrink. No appreciable differences were observed between CF and non-CF HBE cells.

Conclusions: We propose cell shrinkage together with the continued influx of Na+ increases intracellular Na+ ([Na]i) and lead to an inhibition of Na+ transport. Elevated [Na] i is a known inhibitor of epithelial Na+ channels (3). The sustained nature of the inhibition in Na+ transport to a HC may help explain the longer than expected duration of action of HS in the clinical trials (1, 2) . Explanation of the paradoxical effects of amiloride observed in a clinical trial (2) requires further investigation. References: 1. N Engl J Med 354:229-240,2006; 2. N Engl J Med 354:241-250, 2006; 3. Physiol Rev 77:359-396, 1997 Reduced airway surface liquid (ASL) volume resulting from ENaC mediated Na+ hyperabsorption and CFTR-mediated hyposecretion plays a critical role in CF lung disease pathogenesis. However, the mechanisms involved in ENaC/CFTR regulation are poorly understood. Using human bronchial epithelial cultures (HBECs) under thick film conditions, i.e. in the Ussing chamber with the ASL washed away, adenosine (ADO) directly stimulates Cl-secretion. However, in a Cl-free environment, ADO has no effect on the amiloride-sensitive current, which is mediated by ENaC. Mean change in I sc following 100µM ADO to the apical chamber was 0.05 µA/cm 2 from baseline compared to vehicle alone (-0.17 µA/cm 2 ;ns;n=3) in HBECs. Mean delta amiloride (100µM) was -5.32 and -5.38 µA/cm 2 respectively (ns; n=3 each).

In contrast, under thin film conditions we have previously shown that ADO stimulates sustained ASL secretion via activation of CFTR in the presence of the protease-inhibitor aprotinin, whereas trypsin pre-treatment abolished this secretion, likely by activating ENaC and abolishing the electrical driving force for Cl-secretion 1 . Further, ADO-mediated ASL secretion is significantly increased if HBECs (thin film conditions) are left with intact ASL for 24 h, an action that is abolished by acutely washing the apical surface 1 .

Thus, we hypothesised that a soluble ENaC inhibitor was secreted into the ASL under thin film conditions, which could accumulate sufficiently to inhibit ENaC and provide the necessary driving force for CFTR-mediated Cl-secretion after ADO-addition. To search for such an inhibitor, we incubated ASL with trypsin-coated beads and identified bound proteins by mass spectrometry (albumin-coated beads were used as a control to exclude non-specific binding). The major identified protein was PLUNC, a protein that is secreted into the ASL both in vivo and in vitro which has no current known function. Secretion of PLUNC into HBEC ASL was confirmed by Western blot. To test whether PLUNC could alter ASL volume homeostasis, we made an anti-PLUNC shRNA retroviral construct which we used to infect HBECs. Unlike the control anti-luciferase shRNA infected HBECs, anti-PLUNC shRNA-infected HBECs exhibited >90% knockdown of PLUNC and a failure to regulate ASL volume (ctrl, 9.0±1.8µM; 5±0.9µM; n=5) . When co-expressed with ENaC in oocytes, PLUNC inhibited ENaC currents by 78% (n=12). CFTR function in contrast was unaffected by PLUNC expression (n=10).

Thus,it is likely that ADO stimulates CFTR mediated Cl-secretion but is not involved in ENaC regulation. However, experiments were performed in a very simplified environment i.e. static cultures with little mucosal ATP or mucus and these findings will need to be expanded and further evaluated in native tissues. We conclude that PLUNC, rather than ADO may be the soluble mediator in the ASL which regulates ENaC function by reducing its activity. 1. Tarran, R. et al., J. Gen. Physiol., 2006. 127 (5) Airway surface liquid (ASL) absorption is mediated by epithelial Na + channels (ENaC), which establish an osmotic gradient that drives fluid absorption. We and others have recently reported that a protease / anti-protease balance regulates ENaC in normal human bronchial epithelial cells (HBE) and provides a mechanism for auto-regulation of ASL volume. In CF, this balance is disturbed, leading to constitutive proteolytic activation of ENaC and the pathological Na + hyper-absorption characteristic of the disease. To determine if channel activating protease expression is regulated by changes in ASL volume and is altered in CF, we examined prostasin expression in control and CF HBE under basal and ASL volume expansion conditions using Western blotting.

Prostasin migrates as 37 kDa and 40 kDa bands in apically biotinylated proteins, apical secretions, and whole cell lysates of primary HBE. Following apical aprotinin exposure, only the 40 kDa prostasin molecular species was present in the biotinylated proteins, suggesting that the proteolytic conversion of prostasin zymogen to active enzyme occurs on the cell surface. Following ASL volume expansion, cell surface prostasin expression increased by > 55% (p=0.007, n=12 tissue donors, >2 cultures from each).

As our recent studies indicated that increased proteolytic activation of ENaC occurs in CF airway epithelia, we next compared prostasin expression in CF and control HBE cells. Prostasin expression in the apical biotinylate of CF HBE was 1.5 fold greater than in control HBE (p=0.025, n=9 tissue donors). Furthermore, the ratio of the 37 kDa (active enzyme) to 40 kDa (zymogen) prostasin molecular species was 2.5 fold greater in CF (p=0.009), indicating that increased activation of prostasin occurs in CF airway epithelium. We next determined whether increased prostasin activation in CF may reflect the deficiency of a protease inhibitor. Serpin E2, also known as protease nexin-1 (PN-1), forms an ~82 kDa complex with prostasin and permanently inactivates its protease activity. While no significant difference in PN-1 expression was observed between control and CF HBE, PN-1 was found to contribute to prostasin regulation by (i) forming an inactive prostasin complex, (ii) inhibiting the amiloride-sensitive short-circuit current across HBE, and (iii) preventing conversion of prostasin zymogen to active enzyme.

These findings demonstrate that cellular mechanisms coordinate prostasin expression and activity with ASL volume. Accordingly, at times when the ASL volume is high, prostasin expression increases, and this presumably augments Na + and ASL absorption to absorb the excess luminal fluid. These regulatory mechanisms govern the apical membrane expression and processing of zymogen to active enzyme. Because prostasin is incapable of autocatalysis, these findings support the existence of a proteolytic cascade that controls ENaC activity during ASL volume homeostasis, perhaps reflecting the recently reported matriptase-prostasin cascade. Furthermore, the increased expression of prostasin in CF suggests that abnormal regulation of prostasin contributes to Na + hyper-absorption in CF airways.

Supported by the NIH, CFF, and ALA.

We report that annexin 2 (anx 2)-S100A10 forms a functional cAMP/PKA/calcineurin (CaN)-dependent complex with CFTR. Cell stimulation with forskolin/IBMX significantly increases the amount of anx 2-S100A10 that reciprocally co-immunoprecipitates with cell surface CFTR and CaN A. Pre-inhibition of the cells with PKA or CaN inhibitors attenuates the interaction. Furthermore, we find that the acetylated peptide (STVHEILCKLSLEG, Ac1-14), but not the non-acetylated equivalent N1-14, corresponding to the S100A10 binding site on anx 2, disrupts the anx 2-S100A10/CFTR complex. Analysis of DIDS and CFTR inh172 -sensitive currents, taken as indication of the outwardly rectifying Cl-channels (ORCC) and CFTR-mediated currents, respectively, showed that Ac1-14, but not N1-14, inhibits both the cAMP/PKA-dependent ORCC and CFTR activities. CaN inhibitors (cypermethrin, cyclosporin A) discriminated between ORCC/CFTR by inhibiting the CFTR inh172 , but not the DIDS, sensitive currents, by more than 70%. Furthermore, peptide Ac1-14 inhibited acetylcholine-induced short-circuit current measured across a sheet of intact intestinal biopsy. Our data suggests that the anx 2-S100A10/CFTR complex is important for CFTR function across epithelia. The content and water (H2O) mobility in airway epithelium is tighly coupled to airway function. In cystic fibrosis (CF), H2O epithelial permeability is reduced in airways. The demand of noninvasively imaging techniques with high spatial resolution potential is rising because such imaging tools would expedite anatomical and functional phenotyping in the genetically altered mice. Magnetic resonance microscopy (MRM) is a noninvasive, inherently three-dimensional (3D) imaging technique capable of visualizing anatomical structures in the mouse and allows for interpretation of complex spatial relationships between substructures and H2O. In this study, we explore different MR contrast parameters and signal-to-noise ratios at a 30 µm pixel size to characterize microstructure and H2O mobility in ex vivo trachea of CF transmembrane conductance regulator (CFTR)-deficient (CFTR knockout, Cftr tm1UNC ) mice and their aged-matched WT littermates. This study is performed using a Bruker MRM system at 11.7 tesla. We demonstrate for the first time the ability of 3D-MRM to map the H2O content and mobility in trachea epithelium. From the 3D-MRM videoimages, differential H2O content was visualized in different levels of trachea in WT and CF mice. T2 MRM images depicting the H2O rotational mobility which is related to environmental viscosity of trachea epithelium will be also shown. Finally, this 3D-MRM imaging method is a valuable method for measuring H2O content and permeability in airways and can serve for assessing the effects of drugs on H2O mobility in CF airways.

Supported by grants from Inserm, UPMC-Paris6, CNRS, and the French Cystic Fibrosis Association (VLM). Cystic fibrosis (CF) is a lethal inherited disorder caused by mutations in a single gene encoding the cystic fibrosis transmembrane conductance regulator (CFTR) protein resulting in progressive lung oxidative damage. In this study, we evaluated the role of CFTR in the control of ubiquitin-proteasome activity (UPS system) and the NF-κB / IκB-αsignaling after lung oxidative stress. We exposed CFTR deficient (cftr-/-) and wild type mice for 64 h to hyperoxia-mediated oxidative stress. CFTR deficient mice exhibited significantly higher lung proteasomal activity than CFTR+/+ animals after oxidative stress. This was accompagnied by a strong reduction of lung caspase-3 activity and an absence of degradation of NF-κB inhibitor IκB-α. In vitro, human CFTR-deficient lung cells also exhibited higher proteasomal activity and a lack of increased NF-κBdependent transcriptional activity compared to CFTR-sufficient lung cells after oxidative stress. Furthermore, inhibition of the proteasomal activity by MG132 in CFTR-deficient lung cells restored the NF-κB/IκB-α signaling to that of CFTR-sufficient lung cells. Inhibition of caspase-3 by Z-DQMD in CFTR-sufficient lung cells mimicked the response profile of increased proteasomal degradation and lowered NF-κB transcriptional activity of CFTR-deficient lung cells when exposed to oxidative stress. All together, these results suggest that CFTR is a crucial molecule in regulating proteasomal degradation and NF-κB activity in lung epithelium under oxidative stress. Staphylococcus aureus, one of the major pathogen involved in airway infections, releases in the airway lumen virulence factors that may impair the airway epithelial functionality. To date, the effect of S. aureus virulence factors on the loss of electrolyte homeostasis of the airway epithelium has not been investigated. We have previously shown that the combination of a corticosteroid and a long-acting beta 2 agonist attenuated the airway epithelial cell inflammatory response induced by S. aureus virulence factors. The aim of the present work was to investigate the effect of S. aureus virulence factors on the airway epithelial tightness and on the chloride efflux of airway epithelial cells incubated or not with Salmeterol hydroxynaphthoate 2×10 -7 M and Fluticasone propionate 1×10 -8 M (SM/FP). The airway epithelial tightness was assessed by immunocytochemistry, western blotting and transepithelial resistance measurement. The chloride efflux was evaluated by dynamic imaging using the 6-methoxy-N-(3-sulfopropyl) quinolinium (SPQ) fluorescent probe. S. aureus (strain 8325-4) virulence factors were obtained by growing bacteria at 5×10 8 CFU/ml for 18h at 37°C. The bacteria supernatant containing S. aureus soluble virulence factors was then centrifuged, filtered and diluted at 2% in cell culture medium (a concentration that did not alter cell viability). Human airway epithelial cells (MM39) were incubated with 2% S. aureus virulence factors for 1h and then co-incubated or not with SM/FP for 4h. We observed that the incubation with 2% S. aureus virulence factors alone did not significantly alter the epithelial integrity assessed by the expression of tight junction proteins and transepithelial resistance measurement. However, when the cells were incubated with SM/FP alone or with 2% S. aureus virulence factors plus SM/FP, the expression of tight junction proteins was significantly increased (p<0.05) as compared to cells incubated with S. aureus virulence factors. Interestingly, the chloride efflux in airway epithelial cells was significantly decreased in a time-dependent way by S. aureus virulence factors (3 fold decrease after a 4h incubation). The co-incubation of airway epithelial cells with S. aureus virulence factors and SM/FP prevented the S. aureus virulence factorsdependent chloride efflux decrease. Our results demonstrate that bacterial virulence factors induce the loss of the electrolyte homeostasis and suggest that the treatment by the combination of a corticosteroid and a long-acting beta 2 agonist may preserve the airway epithelial functionality.

Supported by Association Vaincre La Mucoviscidose and Glaxo-SmithKline

Nilsson, H.E. 1 ; Dragomir, A. 1 ; Ahlander, A. 1 ; Johannesson, M. 2 ; Roomans, G.M. 1 1. Medical Cell Biology, Uppsala University, Uppsala, Sweden; 2. Women's and Children's Health, Uppsala University, Uppsala, Sweden Inhalation of hyperosmotic solutions (salt, mannitol) has been used in the treatment of patients with cystic fibrosis or asthma, but the mechanism behind the effect of hyperosmotic solutions on mucociliary clearance (MCC) is unclear. One explanation has been suggested, namely that hypertonic solutions open tight junctions, which may lead to increased water transport followed by an increased airway surface liquid (ASL) volume. Furthermore, the role in CFTR-mediated HCO3-conductance in regulating ASL pH has led us to investigate if pH changes may have an effect on the tightness of tight junctions.

The effect of osmolarity was investigated on the 16HBE14o-cell line by the addition of NaCl, NaBr, LiCl, mannitol or xylitol (295-700 mOsm). The effect of pH was investigated on the 16HBE14o-, Calu-3 and T84 cell lines as well as the cystic fibrosis cell line CFBE41o-. Transepithelial resistance was measured as indicator of the tightness of the cultures. Cell-cell contacts and morphology were investigated by immuno-fluorescence and by transmission electron microscopy, with lanthanum nitrate added to the luminal side of the epithelium to investigate tight junction permeability. The electrolyte solutions caused a significant decrease in transepithelial resistance from 450 mOsm on, when the hyperosmolar exposure was gradually increased from 295 to 700 mOsm, whereas the non-electrolyte solutions caused a decrease in transepithelial resistance from 700 mOsm on. Immunofluorescence micrographs showed weaker staining for the proteins ZO-1, claudin-4 in treated samples compared to the control. The ultrastructure revealed an increased number of open tight junctions as well as a disturbed morphology with increasing osmolarity, and with electrolyte solutions opening a larger proportion of tight junctions than non-electrolyte solutions. It was noted that during the exposure of the cultured cells to the hyperosmolar solutions, the pH of the medium increased from 7.4 to 8.0. 16HBE14o-cells exposed to both a rise in pH and hyperosmotic stress showed an overall lower TEER and a significantly reduced ability to recover from stress compared to cultures at a pH hold constant at 7.4. Without exposure to hyperosmolar solutions, a rise in pH caused a significant decrease in transepithelial resistance in 16HBE14o-cells, Calu-3 and T84 cell lines but not in the CFBE14o-cell lines, where the reaction was significantly less and delayed. In conclusion, hyperosmolar solutions caused a reversible opening of the TJ in 16HBE14o-cell cultures with electrolytes having stronger effects than non-electrolytes, where one of the effects on MCC may be due to increased water transport across the leaky paracellular space. An increase in pH caused a significant decrease in TEER in the healthy cell lines compared to the CFBE41o-cell line. We speculated that an impaired alkalinisation of the apical fluid due to a defective CFTR also will cause the tight junctions to react different to other external stimuli, such as osmolarity, compared to healthy cells, which is also indicated by preliminary experiments on the effect of NaCl on TEER in CFBE41o-cultures. Extracellular nucleotides regulate surfactant secretion in alveoli and mucociliary clearance in airway epithelia, but the mechanism(s) of their release and their regulatory pathways remain incompletely understood. Previously, we showed that hypotonic swelling of A549 epithelial cells induces Ca 2+ -dependent secretion of several adenosine and uridine nucleotides, implicating regulated exocytosis. In this study, we examined sources of intracellular Ca 2+ ([Ca 2+ ] i ) elevation evoked by acute 50% hypotonic stress and the role of autocrine purinergic signaling in Ca 2+ -dependent ATP release. We found that ATP release does not directly involve Ca 2+ influx from extracellular spaces, but depends entirely on Ca 2+ mobilization from intracellular stores. The [Ca 2+ ] i response consisted of slowly-rising elevation representing mobilization from thapsigargin (TG)-insensitive stores and a superimposed rapid spike due to Ca 2+ release from TG-sensitive endoplasmic reticulum (ER) stores. The latter could be abolished by hydrolysis of extracellular tri-and di-nucleotides with apyrase; blocking P2Y 2 /P2Y 6 receptors of A549 cells with suramin; blocking UDP receptors (P2Y 6 ) by PPADS; emptying TG-sensitive stores downstream with 1 µM TG or 10 mM caffeine in Ca 2+ -free extracellular solution; or blocking the Ca 2+ -release inositol 1,4,5-triphosphate (IP 3 ) receptor channel of the ER with 75 µM 2aminoethyl diphenylborinate. These results demonstrate that the rapid [Ca 2+ ] i spike results from the autocrine stimulation of IP 3 )/Ca 2+ -coupled P2Y 2 /P2Y 6 receptors, which accounts for ~70% of total Ca 2+ -dependent ATP release evoked by hypotonic shock. Our study reveals a novel paradigm in which ATP release is amplified by the synergistic autocrine/paracrine action of co-released uridine and adenosine nucleotides. We suggest that a similar mechanism of purinergic signal propagation operates in other cell types.

(This study was supported in part by the Canadian Institutes of Health Research and the Canadian Cystic Fibrosis Foundation (CCFF). S.T. was the recipient of a CCFF studentship).

The maintenance of a thin liquid layer on the airways and alveoli surfaces is essential for normal lung physiology.Yet, the mechanism sensing the height of the layer remains obscure. In this study, we examined ATP secretion from human lung A549 epithelial cell monolayers mounted in a closed, flowthrough chamber (0.5 mm height) and found that passage of an air bubble over the monolayer caused transient (<1.5 min) but significant ATP release (600 to 2,000 pmoles/10 6 cells). Air bubble-induced ATP release was reduced by~70% from cells loaded with the intracellular Ca 2+ chelator BAPTA-AM, and was completely abolished in N-methylmalemide-treated (1 mM) cells, suggesting the involvement of Ca 2+ -dependent exocytosis. Fura-2 intracellular Ca 2+ ([Ca 2+ ] i ) imaging experiments revealed transient [Ca 2+ ] i elevation during the passage of an air bubble over the cell monolayer, but the [Ca 2+ ] i response did not involve non-specific Ca 2+ influx from the extracellular space, e.g. due to cell damage, since similar responses were observed in Ca 2+ -free extracellular solution. Ethidium bromide staining did not disclose any cell damage in these experiments. Confocal fluorescence microscopy study showed reversible cell deformation (flattening) of 10% to 40% in height in the monolayer part in contact with the air bubble and confirmed that cell integrity was entirely preserved. These experiments demonstrate that in close proximity to the air-liquid interface (i.e. between the air bubble and the wet cell surface), surface tension forces are transmitted directly on cells, causing their mechanical deformation, elevation of [Ca 2+ ] i and subsequent ATP release. We propose that a similar mechanism may operate in vivo in the airways, where surface tension forces acting directly on exposed epithelial cell surfaces may provide a fail-proof mechanism to protect proper airway surface liquid volume via mechanosensitive, Ca 2+ -dependent ATP release and purinergic modulation of fluid secretion. This mechanism may be defective in cystic fibrosis due to excessive mucus layer covering the airways.

Introduction : Cystic fibrosis is a genetic disease that reflects the consequences of mutations in the cystic fibrosis transmembrane conductance reg-ulator (CFTR) gene, affecting anionic transport in epithelia. SLC26 family members are anionic transporters involved in Cl-and HCO3-absorption or secretion in epithelia. In addition, the activation of some SLC26 family members by CFTR has been demonstrated (Nature cell biology, 2004, 6: 343-350) . SLC26A9 is a poorly characterized member of this family, being the solely expressed in lung. Putative interaction domains with CFTR are also present in the SLC26A9 protein. In this study, we have investigated the transport properties of SLC26A9 (human) to determine the functional and pharmacological characteristics of this transporter.

Methods : To this aim, electrophysiological studies (two-electrode voltage clamp or current clamp methods and intracellular pH measurements using pH sensitive microelectrode) were performed in Xenopus laevis oocytes expressing SLC26A9 proteins. Complementary (36Cl) transport studies were also undertaken.

Results : The protein expression results in the appearance of an anionic current showing a linear current/voltage relationship. 36Cl influxes experiments confirmed the induced Cl-permeability following SLC26A9 protein expression. The sequences of conductivity, Cl->I-> NO3-≥ gluconate > SO4 2-and selectivity (Px/PCl), I-> Cl-= NO3-> gluconate > SO42-are found. The HCO3-conductance mediated by the SLC26A9 protein expression is low. Using CO2/HCO3-containing Ringer solutions, no intracellular pHi changes were detectable in conditions (low chloride in external medium) favoring the Cl-/HCO3-exchange whereas pHi changes (alkalinization) were observed with the expression of the AE1 exchanger. However, pHi changes could be detected in conditions largely favoring the driving force for HCO3-entry. DIDS and NS 3623 inhibited SLC26A9 associated currents. The specific CFTR inhibitor (CFTRinh172) or glybenclamide had little effect. Elevation of intracellular cAMP (a CFTR activator) was also ineffective while maneuvers increasing intracellular calcium blocked the SLC26A9 associated currents.

Conclusions : Our study demonstrates that SLC26A9 presents an electrogenic anion conductance (characteristics of an anionic channel) when expressed in xenopus oocytes (however a transporter functioning as a nCl-/HCO3-exchanger can not be excluded). This function (channel) has also been described for another member of this family, SLC26A7 (J Biol. Chem, 2005, 280: 6463-6470) . The physiological role of SLC26A9, present in the bronchial cells of airway epithelia and its potential interaction with CFTR has to be precised in situ or in oocytes and in mammal expression systems. Medicine, Women's & Children's Hospital, Nth Adelaide, SA, Australia; 2. CF Research & Treatment Center, University of Nth Carolina at Chapel Hill, Chapel Hill, NC, USA; [3] [4] [5] [6] [7] [8] JASRI, Hyogo, Japan; 4. Physics and Synchrotron Science, Monash University, Clayton, VIC, Australia; 5. Paediatrics, University of Adelaide, Adelaide, SA, Australia Non-invasive imaging of lung (e.g. HRCT, MRI, PET) is a valuable modality for detection and monitoring of the effects of CF in human airways, but resolution is limited; e.g. airway HRCT detects airways no smaller than ~ 1.5mm dia. Detection of wall or lumen changes in smaller airways, or the detection and monitoring of induced airway disease in live rodent models, demands significantly greater resolution coupled with rapid image capture. We report substantially increased airway resolution is achievable using synchrotron phase-contrast X-ray and can produce three-dimensional reconstructions encompassing the smallest mouse lung airways.

Nembutal-anaesthetised C57Bl/6 mice (~ 20 gm) were imaged at the SPring-8 synchrotron in Hyogo, Japan. 2-D phase-contrast airway images were obtained on CCD detectors (1.1 µm pixels) with a 90 cm phase-contrast propagation distance at 25 keV; nose and trachea were imaged at 15 or 30 sec intervals (100-300 ms exposures) over 45 mins. Mice killed with nembutal overdose were imaged in axially aligned segments to obtain 3dimensional CT data voume of approximately 20x20x15 mm3 with 12x12x12 um3 voxels of nose, trachea, and lung (Hammamatsu CCD detector). Volume renderings were produced using Volview or OsiriX software.

Live, 2-D nasal or tracheal imaging revealed airway-surface activity in some live mice consistent with dynamic mucociliary clearance activity. Individual CT slices revealed the fine detail in the mouse lung, with dynamic ("fly through") sequences of lung CT slices permitting visual identification and tracking of lung tree branching from trachea (~ 1mm dia) into small airways (approx 100 um dia). Adjustment of X-ray contrast, opacity, and range in these volume reconstructions permitted selective display of the mouse lung conducting-airway tree and airway surfaces. Selected regions of the lung could be examined statically, or in rotation through different orientations in high-resolution 3-D.

Synchrotron phase contrast X-ray provides a new option for non-invasive imaging of intact mouse lung, at a resolution that permits examination of small conducting airways in mice. Combined with volume-reconstruction software these ultraCT images can provide a powerful option for understanding the structure-function relationships produced by airway disease. Ultra-CT 3-D studies are underway to compare lung airways at high resoltuion in normal and transgenic mice having altered airway pathophysiology. The epithelial sodium channel, ENaC, has a vital role in the function of the pulmonary epithelia and significantly contributes to the pathophysiology of the CF airway. Thus, strategies to repair mutant CFTR dysfunction must also consider the influence of such repair on ENaC functional expression. The ∆F508 trafficking repair agent 4-Phenylbutyrate modulates the expression of the 70 kDa molecular chaperones Hsc70 and Hsp70 in CF epithelial cells. We therefore assessed the role of these chaperones in the regulation of ENaC trafficking. In Xenopus oocytes, we previously observed that overexpression of Hsc70 inhibits murine ENaC (mENaC) functional and surface expression. In contrast, Hsp70 can either enhance or inhibit mENaC functional and surface expression depending on the extent of overexpression [Goldfarb et al. (2006), Proc. Natl. Acad. Sci. 103:5817-22 ]. Here, we tested the hypothesis that these differential effects of Hsc70 and Hsp70 on mENaC expression also occur in epithelial cells. MDCK cell lines with stable expression of α-HA, β-V5, γ-Myc-mENaC and tetracyclineinducible expression of either Hsc70, ATPase-deficient Hsc70, Hsp70 or ATPase-deficient Hsp70 containing myc/His epitope-tags were selected. All epitopes were fused to the respective C-termini. These cells were grown as high-resistance (>500 Ω/cm 2 ) monolayers on permeable supports. Doxycycline-induced overexpression of Hsc70 decreased amiloride-sensitive I sc and the whole cell content of α-HAand β-V5-mENaC. In contrast, lower amounts of tetracycline-induced Hsp70 overexpression increased amiloridesensitive I sc and whole cell α-HAand β-V5-mENaC expression; these effects were absent at higher levels of Hsp70 overexpression. These data are consistent with our previous data in Xenopus oocytes. Interestingly, the ATPase-deficient chaperones had the opposite effect on mENaC functional expression. Modest doxycycline-induced expression of ATPase-deficient Hsc70 increased amiloride-sensitive I sc in these cells, while modest overexpression of ATPase-deficient Hsp70 decreased amiloride-sensitive I sc . These effects may result from "dominant negative" interference with Hsc70 and Hsp70 function, respectively. These data are consistent with Hsc70 and Hsp70 having different effects on mENaC functional expression in epithelial cells, and that these effects are dependent upon the ATPase activity of the respective chaperones.

Supported by grants from NIDDK. We have studied survival, airway epithelia bioelectrics and lung pathology of mice over-expressing various combinations of the 3 epithelial Na + channel (ENaC) subunits (α, β and γ, genes Scnn1a, Scnn1b and Scnn1g). We generated double-transgenic mice and crossed them with single-transgenic mice to obtain litters comprising all 8 possible genotypes. Survival analysis revealed that overexpression of βENaC in combination with either αENaC or γENaC significantly reduced survival in comparison to wild-type (WT) littermates. Strikingly, at 3 days of age, all genotypes were represented according to the expected Mendelian proportion, except for the triple transgenic αβγENaC, which was significantly under represented. In utero studies are ongoing to understand if over-expression of αβγ affects the fetuses or the newborn pups during the first hours post-partum. Due to the high mortality of mice over-expressing βENaC subunit combinations, we studied tracheas from pups at 3 days of age. As previously reported, mice overexpressing β, but not the α or γ ENaC subunits, exhibit airways hyperabsorption of Na + and lung pathology (Mall et all 2004) . The βENaC tracheas exhibited Na + absorption [as measured by the change in short circuit current (Isc) in response to amiloride in the Ussing chamber] that was ~5.4 fold greater (∆Isc 24.6±3.9 µA×cm -2 , n=9, means±SEM) than WT (4.6±0.8 µA×cm -2 , n=7). Over-expressing αβ or αγ ENaC subunits resulted in rates of Na + absorption that did not differ significantly from those of βENaC or γENaC subunit alone, respectively. However over-expressing the β and γ ENaC subunits together resulted in an amiloride sensitive Isc that was ~13.7 fold greater (∆Isc 63.1±8.2 µA×cm -2 , n=9) than WT. We also studied the tracheal bioelectrics of the few 3 day-old αβγENaC pups available. Tracheas from the αbγ pups exhibited Na + absorption that was ~16.7 fold greater (∆Isc 77±8.2 µA×cm -2 , n=3) than WT. Due to the small sample size, we cannot determine if this response is significantly different from the response of the βγENaC pups. Analysis of lung pathology in 3 day-old pups revealed that all the combinations that exhibited increased airway Na + absorption and decreased survival in comparison to WT littermates, e.g. αβ, βγ and αβγ, also presented with alveolar space enlargement, maybe due to postobstructive air trapping, and airway epithelia necrotic degeneration associated with AB-PAS negative bronchial obstruction. From these data, it appears that the rate of airway Na + absorption negatively correlates with the survival of the pups. Thus, it is likely that the mucus and airway surface liquid is more dehydrated as a function of the rates of Na + absorption (aβγ > βγ > αβ, β > WT) which produces gradual failure to survive due to asphyxia secondary to airway obstruction. Supported by NIH SCOR P50 HL060280

Choi, J. 1,2 ; Joo, N. 1 ; Wu, J. 1 ; Krouse, M. 1 ; Wine, J.J. 1 1. Cystic Fibrosis Research Laboratory, , Stanford, CA, USA; 2. otorhinolaryngology, Yonsei university, Seoul, South Korea Submucosal glands, which produce most airway mucus, are mainly controlled by parasympathetic pathways that stimulate glands via airway intrinsic neurons distributed along the airway walls. The intrinsic neurons can be stimulated via axon reflexes from receptors in the mucosa. In previous experiments, it was shown that capsaicin applied to the mucosa stimulated gland secretion in the upper tracheas of WT but not cftr -/-mice (Ianowski et al., J Physiol, 2007, 580, 301) . It seemed important to assess a possible role of capsaicin-sensitive pathways in pigs and humans. We obtained pig tracheas following acute experiments carried out for other reasons, and human airways following lung transplants from donor tracheas and from the excised lungs of the transplant recipients. Secretion from individual submucosal glands was quantified optically by time-lapse digital imaging of the growth of spherical bubbles of mucus in an oil layer In humans, commercial chili oil (5 µM dispersed in 20 µM mineral oil) stimulated mucus secretion from submucosal glands from both donors (n = 46 glands from 9 subjects, 0.30 ± 0.24 nl/min/gland) and disease control subjects (n = 45 glands from 7 subjects, 0.25 ± 0.16 nl/min/gland). However, there were no responses to chili oil in submucosal glands from CF subjects (n = 36 glands from 5 subjects, 0.01 ± 0.01 nl/min/gland). In pigs, we determined concentrationresponse relations for gland secretion in response to purified capsaicin (Sigma). The threshold was ~1 µM and the EC50 was 19.01 ± 4.9 µM. The response to 100 µM capsaicin was partially blocked by high dose TTX (>1 µM). As in humans, the capsaicin response required CFTR, because it was blocked ~70% by CFTRinh-172, (n = 29 glands, from 4 pigs, p < 0.05). CFTRinh-172 is presently the most specific inhibitor of CFTR that can be used with glands, where we do not have access to the apical membrane. In ferrets, intrinsic airway neurons express acetylcholine, VIP and SP, often colocalized within the same neurons (Dey et al., Am J Respir Cell Mol Biol, 1996, 14, 207) . In pigs, all three of these transmitter systems appear to be involved in gland secretion to capsaicin, because the response was partially blocked by the NK-1 receptor blocker L703606 (oxalate salt, 1µM), the VIP receptor inhibitor L8K (10µM), and the muscarinic receptor blocker atropine (1µM); each of these blocked 36 to 59 % of the response (n = 21 to 39 glands from 3 to 5 pigs). Indeed, most of the local pathways are probably damaged, which may account for the relatively small responses to strong stimulation. Relatively few experiments have been carried out in intact airways with uninterrupted innervation and circulation, but in those experiments the glands were highly responsive to modest mucosal stimuli (reviewed in Wine, Auto Neurosci, 2007, 133, 35) . Hence, we hypothesize that (1) reflex stimulation of glands plays an important role in lung innate defenses; (2) defects in these responses is an important reason that CF airways are prone to infection, and (3) local reflexes may assume greater importance for maintaining the mucosal defenses of transplanted lungs.

Supported by NIDDK RO1-51817 (JJW), CFF and CFRI.

Most airway mucus arises from submucosal glands, which express CFTR in their ciliated ducts and in serous cells. The airway glands secrete to agonists that either increase intracellular Ca2+ (e.g. Acetylcholine) or cAMP (e.g. VIP).Substance P(SP) also stimulates airway gland liquid secretion in pig (Trout L et al., Am J Physiol Lung Cell Mol Physiol. 2001, 281,:L639) , suggesting that sensory afferents or local neurons that express SP play a role in stimulating submucosal glands. We used pharmacological methods to dissect the secretory mechanism of SP-induced gland secretion in comparison with responses to carbachol. Tracheas were obtained from pigs after acute surgeries carried out for other reasons. The ventral mucosa with underlying glands was dissected from the cartilage, pinned mucosal side up at the gas/bath interface of a physiological chamber, and covered with oil. Secretions from individual glands could be visualized as spherical bubbles in the oil, and secretion rates determined by optical monitoring of bubble diameters. Concentration-response relations for gland secretion were determined for SP (19-42 bubbles at each concentration from 9 pigs). The threshold was ~100 nM for SP and the EC50 was 1.76 ± 0.57 µM. The maximum secretion rate to SP was 1.17 ± 0.21 nl/min.gland, which is only ~ 1/3 of the maximal secretion rate to carbachol (3.43± 0.29 nl/min.gland). The inhibition profile for responses to SP (10 µM) was quite different from those to carbachol (200 nM, a value chose to mimic the secretion rate to 10 µM SP). Gland secretory responses to 10 µM SP (0.75 ± 0.17 nl/min.gland, 41 glands from 4 pigs) were strongly inhibited by 20 µM CFTRinh-172 (0.34 ± 0.10 nl/min.gland, 36 glands from 4 pigs, p<0.05), and by 25 µM Clotrimazole (0.28 ± 0.06 nl/min.gland, 31 glands from 4 pigs), whereas 100 µM Niflumic acid did not inhibit (0.81 ± 0.25 nl/min.gland, 32 glands from 3 pigs). In contrast, secretory response to 200 nM carbachol (0.61 ± 0.22 nl/min.gland, 36 glands from 3 pigs) were only weakly inhibited by CFTRinh-172 (0.49 ± 0.13 nl/min.gland, 29 glands from 3 pigs, p<0.05), and by Clotrimazole (0.53 ± 0.16 nl/min.gland, 33 glands from 3 pigs, p>0.05), but niflumic acid, which was ineffective with SP, inhibited the carbachol response (0.44 ± 0.13 nl/min.gland, 41 glands from 3 pigs, p<0.05). Gland secretion to SP depends at least partially on intracellular Ca2+ release, because the response was partially inhibited by 50 µM BAPTA-AM (0.33 ± 0.09 nl/min.gland, 21 glands from 2 pigs, p<0.05). There was no additional effect of SP on top of 1 µM carbachol, but a subthreshold level of SP (50 nM) showed synergic response (0.44 ± 0.22 nl/min.gland, 21 glands from 3 pigs) with subthreshold level of VIP (10 nM). We are presently testing how SP activates CFTR-dependent mucus secretion. In initial experiments to determine if PKC was being recruited, however the PKC inhibitor (GF 109203, Sigma) did not suppress SP-induced mucus secretion. In contrast, the NKCC inhibitor bumetanide (100 µM) markedly suppressed secretion to SP (0.22 ± 0.06 nl/min.gland, 23 glands from 2 pigs, p<0.05).

Supported by NIDDK RO1-51817 (JJW), CFF and CFRI. Mucus obstruction of airways is considered the most vicious agent of morbidity and mortality in cystic fibrosis (CF). To provide optimal defense of the small airways the volume/thickness of airway surface fluid must be controlled according to physiological demand. To date we have no clear understandings of how these fluids are maintained in a steady state between two pathological extremes (too little or too much) in native small airways. Earlier studies have attempted to measure the ion transport properties of small airways of sheep (Al-Bazzaz et. al. 2001 ) and pigs (Ballard et. al. 1992 ), but the complicated branching structure of small airways may have compromised electrical signals. To avoid dissection trauma Wang et. al. (2005) examined the electrophysiological properties of undissected intact small airways from pigs in vitro by microperfusing bronchioles (diam.1-2 mm) embedded in the lung parenchyma; however, these studies were limited to luminal manipulations and transbronchiolar electrical potentials only. To more rigorously define small airway properties, we designed a special micro Ussing chamber ( Area ≈1 mm 2 ). We excised and opened small airways (diam.1-2mm) of pig lung that we mounted as a flat sheet over a PVC supporting filter (10-20 µm holes). Both sides of the tissue could then be bathed with different solutions at 37°C. The luminal side was isolated by pressing a pipette (≈1 mm diam., tip fire polished) on to the apical surface of the tissue until it sealed electrically. Constant current pulses (0.5-3.0 µA) were passed across the tissue. Transepithelial potential (TEP) and resistance in bilateral 150 mM NaCl Ringers were, -1.3 ± 0.1 mV and 83.1 ± 8.1 Ω.cm 2 (Mean ± SE) respectively. When 150 mM NaGlu Ringers replaced the luminal solution, TEP increased significantly to -36.7 ± 1.5 mV (n=30, p< 0.0005) and resistance increased to 173.8 ± 21.0 Ω.cm 2 (n=30, p< 0.0005. Adding forkskolin (Fsk, 10 µM) plus IBMX (0.1 mM) hyperpolarized the TEP to -40.22 ± 1.99 mV (n=16, p<0.002), but decreased the resistance to 142.9 ± 28.2 Ω.cm 2 (n=16, p< 0.0001). Luminal amiloride (10 µM) depolarized the TEP to -32.6 ± 1.6 mV (n=25, p<0.0001) but significantly increased resistance to 215.7 ± 26.6 Ω.cm 2 (n=25, p<0.0005). These results show that this system supports measurements of ion transport properties of airways smaller than 1 mm diameter where CF lung pathology is thought to originate. These airway epithelia express a very high constitutively active Clconductance and are apparently capable of secretory as well as absorptive functions. Amiloride sensitive Na + conductance (≈ 43 mS/cm 2 ) was not significantly affected by Fsk added to activate secretion. The observation that stimulation with Fsk did not significantly suppress Na + conductance (≈ 41 mS/cm 2 ) (p= 0.4) may suggest that separate groups of reabsorptive and secretory cells comprise small airway epithelia. Mucins are class of proteins uniquely characterized by large glycosylated domains. These proteins likely play a key role in cystic fibrosis (CF) lung disease as major constituents of mucus. In airways of both humans and mice, the gel-forming secreted mucins are represented by MUC5AC and MUC5B, and the transmembrane mucins are represented primarily by MUC1, MUC4, and MUC16. We have been interested in the role of the transmembrane mucins in airways and have previously demonstrated that lack of Muc1 in a knock-out mouse model (provided by S. Gendler) shows increased susceptibility to adenoviral-mediated gene transfer (Stonebraker, J. Virology, 2004) . To expand upon these findings, mice deficient for Muc4 were developed using embryonic stem cell technology by deleting exon 1 of the mouse Muc4 gene, which contains the ATG start codon and signal peptide. RNA analysis reveals loss of exon 1 in homozygous mutant mice and immunohistochemistry using a VNTR antibody indicated a loss of full length Muc4 protein. Muc4 homozygous mice are viable and do not develop any spontaneous, readily detectable disease phenotype. Although the Muc4 deficient mice do have reduced fertility consisting of both reduced litter size and number, all major tissue systems where the Muc4 protein is easily detected, including the eye, respiratory tract, intestine, and reproductive tracts, appear normal in year old mice after histological analyses. In contrast to the results we reported in the Muc1 mice, the Muc4 deficient mice do not show an increase in adenoviral mediated gene transfer, suggesting that Muc4 is not involved in the barrier aspect of the airway glycocalxy, at least for adenoviral infection. Interestingly, quantitative real-time PCR analysis revealed up-regulation of Muc16 protein in the trachea of the Muc4 deficient mice, suggesting that Muc16 may be able to compensate for loss of Muc4. The expression of MUC4 in human ciliated cells as revealed by recent antibody studies prompted us to evaluate the potential role of Muc4 in ciliary function. Ciliary beat frequency (CBF) was monitored in a shear chamber using microscopy image analysis. CBF under baseline conditions for this study (low shear stress) from Muc4 deficient trachea (C57BL/6J congenic N5) was significantly lower than that for littermate controls (8.86 vs. 11.53, SEM ±1.31, P = 0.046). Application of shear resulted in increased CBF in both groups, although the WT response to shear was smaller in magnitude. Thus, Muc4 may have a role in supporting basal CBF and may make CBF less vulnerable to stimulation by shear, perhaps by reducing the inherent friction of ciliary motion. This finding underscores the usefulness of this model for dissecting out the normal roles that the transmembrane mucins play in airway biology. Supported by NIH (HL066973) and CFF (R026-CR02).

ENaC (Joo et al. JBC 279, 2004 and JBC 281, 2006) . In a search for a model to test the hypothesis we found that Ussing chamber short-circuit current(I sc ) measurements with freshly obtained airway mucosa provided useful information about ENaC activity in response to various drug treatments. In these studies we observed an as yet unidentified pathway that regulates surface ENaC activity. Airway mucosa from sheep tracheas displays a large, amiloride/benzamil-sensitive I sc prior to stimulation. When stimulated with basolateral carbachol, the ENaC I sc was almost abolished, as indicated by a sustained decrease (77.1 ± 3.4%, n=14) in I sc after an initial increase and left a small effect to subsequent amiloride or benzamil. In contrast, when carbachol was given after benzamil, it again produced a peak response followed by a sustained increase (~30%, n=4) in the I sc . ENaC inhibition by carbachol was dose dependent with a minimal sensitivity between 0.1~1 µM and was slowly reversible. Apical 100 µM ATP also inhibited ENaC I sc by 48.4 ± 8.1% (n=3) and carbachol on top of ATP further reduced the ENaC I sc by 93.6 ± 3.9% (n=3). However, 10 U/ml of apyrase pretreatment, in attempt to breakdown ATP, failed to inhibit carbachol-induced ENaC inhibition, indicating that ATP release is not on obligatory component of carbachol inhibition of ENaC. Carbachol inhibition of ENaC could result from a direct action of carbachol on the surface epithelia, by activation of airway intrinsic neurons, or by activation of the glands, which were intact in these preparations and are strongly stimulated by carbachol. Rabbit tracheas, which lack airway submucosal glands, displayed large ENaC currents and these were not inhibited by cholinergic stimulation (n=3). Carbachol-induced ENaC inhibition, like gland stimulation but unlike stimulation of neurons, was mediated by muscarinic receptors since it was eliminated by atropine but not by the nicotinic receptor antagonist, hexomethonium bromide (n=3). However, inhibition of gland secretion with HEPES/bumetanide did not eliminate the ability of carbachol to inhibit ENaC. The effects of carbachol on surface Na + transport in our study are consistent with previous observations in microperfused sheep bronchiole preparations (Al-Bazzaz, AJP-Lung 11, 1994). Although we do not yet know the mechanism, it would be physiologically efficient if parasympathetic, cholinergic activation of airway gland secretion was accompanied by ENaC inhibition, which would minimize the fluid absorption by surface epithelia and increase clearance. It will be important to determine if cholinergic inhibition of ENaC is present in human tissues, and if so, if it is defective in CF.

Supported by CFF and NIH. Introduction: Lung disease accounts for more than 95% of the current morbidity and mortality associated with CF. The majority of research aiming to elucidate the reason(s) of severe lung disease has focused on airway surface liquid (ASL) layer homeostasis or composition. The CFTR gene encodes a cAMP-regulated Clchannel located on the apical side of the epithelium and is instrumental in maintaining a hydrated ASL layer and promoting effective mucociliary clearance (MCC). In patients with class I or class II CFTR genetic mutations, normal CFTR Clchannel activity in the epithelium is impaired resulting in Na + hyperabsorption and consequently in the generation of a thick and static overlaying ASL layer creating an environment that favors progressively worsening bacterial infections.

In this study we investigated the potential contributory role of altered respiratory cilia function to inefficient MCC in CF airways allowing for bacterial infestations. Our ex vivo model system was the nasal airways of the CFTR gene knockout (cftr tm1Unc ) mouse in which key aspects of the phenotype of human CF lung disease, such as dehydrated ASL layer and goblet hyperplasia, are observed.

Methods and Results: Ciliary beat frequency (CBF) was measured using a dual temperature controlled perfusion chamber, differential interference contrast microscopy and high speed digital video analysis system. We evaluated the CBF of nasal septa explants from homozygote CF (affected) and non-affected littermate wild type (wt) and heterozygote mice congenic on the C57Bl/6 background. The basal CBF of homozygote CF mice (12±3 Hz, n=7) was significantly lower than that of wt mice (28±5 Hz, n=7) and also heterozygote mice (19±3 Hz, n=17) (P<0.05, ANOVA, SNK test). To determine whether the dehydrated ASL layer was still present on the surface of the CF mouse nasal airway epithelium and as such impeding normal cilial beating, we fixed nasal septa (n=3) of CF and wt mice immediately following isolation using the PFC/OsO 4 fixation method to preserve the fragile ASL layer and compared the height of the ASL layer to nasal septa (n=3) from CF and wt mice fixed with PFC/OsO 4 immediately following CBF analysis. Electron microscopy analysis revealed that although the ASL layer was present at the time of isolation, there was no ASL layer following the perfusion of solutions necessary for the CBF measurements. This finding demonstrated that the activity of cilia was not impaired in the CF nasal airway epithelium due to the presence of a dehydrated ASL layer. CBF frequency was also significantly reduced in three different codes of mouse ciliated nasal airway epithelial cultures derived from affected CF mice compared to those derived from non-affected wt mice.

Conclusion: We show that the absence of the CFTR gene product in the CF knockout mouse nasal airway epithelium is associated with significantly decreased basal CBF compared to the non-affected wt and heterozygote mice. This observation potentially provides an additional mechanism as to why lung disease persists in human CF lungs.

ML and MBA contributed equally. Clinical studies have linked increased sputum and peripheral blood neutrophil MPO activity with increased airflow obstruction in cystic fibrosis (CF) patients of the same age, gender, airway bacterial flora, and CFTR genotype. Variations in the TGF-β1 gene associated with increased TGF-β1 production have been linked to worse airflow obstruction in CF patients of similar age, gender and CFTR genotype. We hypothesized that in the presence of MPO, TGF-β1 production would increase in airway epithelial cells. We obtained normal human airway epithelial cells (hAE) and cultured them in trans-wells. TGF-β1 mRNA was measured by PCR. TGF-β1 protein was detected by immunofluorescence staining and immunoprecipitation. We found increased TGF-β1 mRNA after hAE were exposed to MPO (at activities found in CF sputum) normalized to mRNA of constitutive beta actin. Under identical conditions, TGF-β1 protein expression was increased in hAE. We conclude that neutrophil MPO present on the apical surface of cultured hAE induces transcription and protein synthesis of TGF-β1. These results suggest that MPO induction of airway epithelial TGF-β1 is one molecular system linking the chronic neutrophilic endobronchitis seen in CF and subsequent airway wall fibrosis. are more oxidatively stressed than normal. As oxidative stress (OS) is a potent activator of NF-κB, these data supported our hypothesis that CFTR dysfunction causes alterations in the expression and/or modification of redox related proteins thereby enhancing activation of NF-κB. Delineation of this interaction will produce good therapeutic targets. In this report we examine the role of OS in CF inflammation. Methods: To address our hypothesis, we used both in vitro and in vivo models of CF. Our airway epithelial cell models are the 9HTEo-pCEP and pCEP-R cell pair; and the 16HBEo-sense (S) and antisense (AS) cell pair. Additionally, we examined whole lungs from 8-12 week old B6.129S6-Cftr tm2Mrc mice (R117H mutants back-crossed into the C57BL6j background). Proteins were prepared from homogenates of whole samples, run on 2-D gels, and their densities compared. The gel spots of differentially expressed proteins were excised, subjected to in-gel trypsin digestion, and identified by LC MS/MS. For biochemical analysis, we tested for the levels of H 2 O 2 , glutathione, and lipid peroxidation. We correlated these data with measurements of NF-κB activity (ELISA and cytokine production), and the activity of Nrf-2, an antioxidant (AO) response pathway transcription factor. We studied cultured cells in the presence and absence of inflammatory stimulation (TNF-α/IL-1β), and compared non-stimulated CF mouse lungs to lungs from normal litter mates. Results: Our proteomic analysis revealed that in the absence of inflammatory stimulation, a paradoxical decrease in AO proteins exists in CF cells by 2 or more fold compared to matched pairs. This is despite a significant increase in intracellular OS, which we confirmed by 3 different biochemical measures. Activity of Nrf-2 was decreased by ~30% in CF cells vs. normal. The data predict an increase in H 2 O 2 , which is a potent activator of NF-κB, and we confirmed both a significant elevation in H 2 O 2 (up 3-5 fold in CF) and a related significant increase in NF-κB activity, as assessed by ELISA for nuclear p50, and IL-6 and IL-8 production. When we analyzed whole lungs from CF mice for antioxidant protein levels, we similarly observed decreases vs. normal littermates. When cells were stimulated, differences in protein expression and oxidative stress between CF and normal were further enhanced, corresponding to an excessive activation of NF-κB and increased production of IL-6 and IL-8. Treatment with the antioxidants NAC or selenium decreased the activation of NF-κB to normal levels. Members of the ClC family of proteins form either voltage-gated chloride channels or Cl -/H + exchangers; nine subtypes are found in mammals. Among these, ClC-2 is one of the most ubiquitous variants, which is expressed in several of the types of epithelial cells that also express the cystic fibrosis transmembrane conductance regulator (CFTR). It has been postulated that ClC-2 may be a suitable alternative chloride pathway in these cells; however the contribution of ClC-2 to chloride transport in these cells is poorly understood. This is partly due to the lack of appropriate pharmacological tools that are capable of specifically inhibiting ClC-2. We describe here the isolation, from venom of the scorpion Leiurus quinquestriatus hebraeus, of the first peptide inhibitor of any ClC channel. This toxin, which we have named GaTx2, specifically inhibits the ClC-2 channel. GaTx2 bears primary sequence identical to that of a toxin thought to serve as a K + channel inhibitor, although no molecular target for this toxin was previously known. A homology model of GaTx2 reveals that it adopts a similar fold to other scorpion toxins. GaTx2 was prepared via solid-phase chemical synthesis, and the synthetic toxin inhibits ClC-2 with higher affinity than any other inhibitor of ClC-2 or any other chloride channel. The K D for GaTx2mediated inhibition of ClC-2 in multichannel patches is only 11 pM at V M = -100 mV, and 80 pM at -160 mV. Single channel experiments show that GaTx2 does not alter single channel amplitude, but may alter channel gating. These experiments provide the basis for developing GaTx2 into a useful tool that can be used to define the role of ClC-2 in airway and intestinal epithelial cells. Furthermore, the high affinity and specificity of inhibition suggest that GaTx2 interacts intimately with the channel, and thus may The exploitation of alternative chloride channels to bypass the defect in CFTR-mediated chloride secretion is an established therapeutic strategy in CF. In search of such potential therapeutic targets, we assessed cAMP/PKAdependent chloride secretion across excised nasal epithelium of Cftr null mice and wildtype controls, by measuring short-circuit currents (Isc) in an Ussing chamber set-up. All experiments were performed in the presence of amiloride to block the contribution of epithelial sodium channels to the Isc. In Cftr null mice of mixed background (C57BL/6;129), the cAMP agonist forskolin activated a bumetanide-sensitive rise in Isc that was comparable in magnitude to the CFTR-mediated Isc response of littermate controls (118±30 vs. 90±21 µA/cm 2 , respectively). The forskolin-induced Isc in these null mice was inhibited by the general chloride channel inhibitor DPC, but appeared insensitive to the CFTR inhibitors CFTRinh-172 and glibenclamide, the ORCC and calcium-activated chloride channel inhibitor DIDS, and the CLC-2 inhibitors cadmium and pre-activated omeprazol. Moreover, the Isc response to forskolin greatly exceeded the response to the calciumlinked agonists ionomycin and carbachol. Intriguingly, in nasal epithelium of Cftr null mice with a different genetic background (FVB), no such forskolin-inducible secretory pathway was evident (Cftr-/-: 7±4 vs. Cftr+/+: 98±23 µA/cm 2 ). When, in nasal epithelium of Cftr null mice (C57BL/6;129), the basolateral membrane electrical resistance was negated by nystatin treatment, and in the presence of a transepithelial chloride gradient, forskolin failed to elicit a rise in Isc, suggesting that cAMP/PKA signaling targets a basolaterally located ion transport system. Indeed, in intact epithelium, addition of chromanol 293B, a selective inhibitor of cAMP/PKA-activated potassium channels, to the serosal bath, blocked the forskolin-induced rise in Isc. We tentatively conclude that the hyperpolarization of nasal epithelial cells, which ensues from cAMP/PKA-induced potassium efflux across the basolateral membrane, drives chloride exit across the apical membrane via a constitutively active anion conductive pathway that is absent from FVB Cftr null mice. RT-PCR and immunostaining demonstrated that, among the candidate anion channels that fit the above pharmacological profile, CLC-Ka, but not CLC-Kb, was expressed in nasal epithelium of both Cftr null mice and controls (C57BL/6;129), together with the accessory protein barttin. Importantly, immunostaining suggested a strong increase in ClC-K protein abundance in the Cftr null mice, as compared to wildtype animals. On the basis of these findings, we postulate that CLC-Ka channels may serve as CFTR bypass channels in nasal epithelium of Cftr null mice that, albeit indirectly, are responsive to cAMP/PKA signaling. Future studies aim to characterize CLC-Ka and barttin in the airways from CF patients and controls, and to identify the post-transcriptional mechanism resulting in the up-regulation of CLC-Ka protein in Cftr null mice. In the disease cystic fibrosis (CF), the most common mutation F508del results in endoplasmic reticulum retention of misfolded CF gene proteins (CFTR). The endoplasmic reticulum (ER) F508del-CFTR protein retention is dependent upon chaperone proteins, many of which require Ca 2+ for optimal activity. An increase in cytosolic free Ca 2+ concentration is dependent on the activity of ion channels in the membrane of ER and on extracellular Ca 2+ influx. This influx, named capacitative Ca2+ entry (CCE), induced by calcium store depletion, represents the major Ca 2+ influx mechanism in cells (Putney et al., 1986) . During the CCE, Ca 2+ influx can be mediated by one or several of the 7 isoforms of transient receptor potential canonical channels (TRPC) (Putney et al., 1986) .

Aims & Methods:

In the present work, we studied the CCE in human CF tracheal gland serous CFKM4 cells compared to non CF human tracheal MM39 cells. For this, we characterized the molecular identity of TRPC channels responsible of CCE by RT PCR technique and we compared global Ca 2+ influx in CF and non CF cells by using Fluo4-AM probes. Moreover, we measured the single channel activity of plasma membrane TRPC channels by cellattached patch-clamp configuration in CF and non CF cells.

We also studied the CCE in CF cells after correction of the abnormal F508del-CFTR trafficking by miglustat (2h at 100 µM of N-butyldeoxynojirimycin, NB-DNJ) (Norez et al., 2006) and low temperature (24h at 27°C) (Denning et al., 1992) .

Results:

We detected by RT-PCR technique the presence of TRPC1 and TRPC6 in both cell lines and we observed that CCE was increased 2-fold in CF-KM4 cells compare to non-CF MM39 cells. Following stimulation by histamine, we recorded elementary Ca 2+ currents with similar amplitude (0.5pA at -60mV) and conductance (8pS) whatever the cells tested (CF-KM4 or MM39 cells). However, we observed a 2-fold increase of the NPo (number of channels in each patch × open probability of single channel) in CF cells compare to non-CF cells.

In F508del-CFTR corrected-CF cells, the Ca 2+ influx stimulated by the Ca 2+ stores depletion was reduced to a similar level to non CF MM39 cells. The permanent contact of the airway epithelium with the external milieu induces frequent injuries caused by inhaled pathogens and particles. Regeneration of the wounded airway epithelium needs to respect its prior structure in order to restore its defence functions. The underlying molecular mechanisms associated with the regeneration of the human airway epithelium are poorly described. The aim of our study was to determine, using a global expression approach, the transcriptional profile of human airway cells during the epithelial regeneration process. Human airway epithelial cells collected from nasal polyps were seeded on type IV collagen-coated porous membranes, cultured up to confluence in liquid-liquid(LL) conditions, then put at the air-liquid interface (ALI) allowing epithelial mucociliary differentiation. RNA were extracted from epithelial cells at 70% confluence in LL conditions (epithelial cell proliferation and migration stage, step I), after 5 days in ALI conditions (beginning of epithelial cell differentiation, step II), when first ciliated cells appeared (step III) and when cultures were completely differentiated (step IV). Our results demonstrated that between step I and step II, 127 genes were significantly modulated (ratio R<-3 or R>3, P<0.001), among them 74 were activated and 53 were repressed. Between step II and step III, 127 transcripts were up-regulated (tubulin, dynein, STATH, microtubule associated proteins…) and only one gene (fatty acid desaturase-1) was repressed. Finally, between step III and step IV, except STATH which was~12-fold down-regulated, we observed only few genes differentially and slightly modulated. All along the regeneration process, between step I and step IV, 305 genes were significantly modulated with 84% of activated and 16% of repressed genes, among them a family of genes encoding proteins involved in the ciliary differentiation (FOXJ1, tektin, dynein, tubulin, …), extracellular matrix proteins (collagen, laminin, …) and inflammatory-related genes like cytokines, growth factors and their receptors (CCL20, VEGFC, CSF3, IL32, IL13RA2…). By elucidating the specific transcriptome involved in each step of the human airway epithelium regeneration, our study could allow to better understand the different cellular and molecular mechanisms, as well as the chronological events involved in these processes, and will further be applied to the study of the regeneration process in cystic fibrosis airway epithelium.

Supported by french Association Vaincre La Mucoviscidose. Electrolytes may be transported either through cells, the transcellular pathway, or between cells, the paracellular pathway. Most research has focused on the channels and transporters of the transcellular pathway, while much less is known about paracellular transport. Paracellular transport is regulated by tight junctions. Claudins, the main component of tight junctions, are necessary and sufficient for tight junction formation and are important in establishing ion selectivity. The goal of this study was to further investigate the role of tight junctions in human airway epithelia. Real-time PCR identified claudins 1, 3, 4, 7, 10, and 16 in primary human airway epithelia. Immunocytochemistry localized claudins 1, 3, 4, and 7 to tight junctions. To measure the relative ion selectivity of tight junctions, we studied well-differentiated human airway epithelia in Ussing chambers. To minimize transcellular transport, we used CF epithelia, which lack CFTR and blocked ENaC with amiloride. We found the PNa/PCl is 1.55 ± .34, suggesting tight juctions are more permeable to cations than anions. The relative cation selectivity was Na+ > K+ > Cs+ > TMA+ > Mg2+ and the relative anion selectivity was Cl-> Br-≥ I-> NO3-> Gluconate-. We compared these in vitro results to in vivo studies in people with CF and measured nasal voltage in the presence of amiloride. We found the PNa/PCl is 1.2 ± 0.1. Tight junctions were slightly more selective for cations than anions in vitro and in vivo. To determine the effect of specific claudins on the selectivity of the paracellular pathway, primary human airway epithelia were infected with adenovirus expressing claudin 4 or GFP. Epithelia overexpressing claudin 4 had a higher transepithelial resistance (TER) and the PNa/PCl was 1.11. However, infection efficiency was only 22%. This lead us to investigate if transformed human airway epithelial cells lines may be a better model to study permeability. CuFi cells also express claudins 1, 3, 4, 7, 10, and 16. Furthermore, like the primary human airway, CuFi cells are more selective for cations; PNa/PCl = 1.38 ± .14. Future studies will investigate the effect of claudin expression on the paracellular permeability of these cells. The results will help identify the properties of a pathway critical to epithelial transport and barrier functions. Altered adenosine (ADO) signaling is associated with chronic lung disease and increased ASL ADO levels are thought to drive inflammation by stimulation of A2B receptors (A2B-R) in airway epithelia, which can activate NFκB. When measured by lavage, ASL ADO is ≤200 nM. Due to its relatively low affinity to ADO, A2B-R are thought to be inactive under normal conditions (EC50 is ~1.5 µM), and only stimulated during chronic lung disease. However, we have previously demonstrated that the ADO-receptor (ADO-R) antagonist 8-SPT inhibits ASL volume regulation, and that A2B-R is in close proximity to CFTR. Thus, we hypothesized that the ASL ADO concentration close to the apical membrane is higher than measured by ASL lavage in bulk solution. We tested this hypothesis by looking for ADO-R expression and function in well-differentiated human bronchial epithelia cultures (HBECs) and in vivo. By PCR analysis, we found that A2B-R was the only ADO-R expressed in HBECs. We then measured ASL height over 8 h ± ADO-R antagonists by XZ confocal microcopy when ASL height was preset to ~5 µm at t=0. At 30 µM, caffeine is a non-specific ADO-R antagonist, which inhibited ASL volume regulation (8 h ASL height was 4.1 µm vs. 8.2 µm for controls, n=6). Similar inhibition was observed with alloxazine, which is relatively specific for A2B-R (n=6). In contrast, DPCPX, ZM241385 & MRS241385, which inhibit A1, A2A & A3-R respectively had no effect on ASL volume regulation. ASL height regulation was also ablated by the addition of adenosine deaminase (ADA) to the ASL (n = 6). We next attempted to recapitulate the changes in ASL height seen under control conditions, i.e. increase from 5 to 8 µm over the initial 2 h followed by a plateau (the remainder of the 8 h) by adding the non-hydrolyzable ADO analogue NECA to the ASL in the presence of ADA. 100 nM NECA was insufficient to regulate ASL volume and height remained at 5 µm over 8 h.

In contrast, addition of 1 µM NECA to the ASL resulted ASL volume regulation that was indistinguishable from control conditions (n=6), suggesting that near-membrane ASL is much higher than has been measured by lavage. To confirm that A2B-R stimulates CFTR chloride secretion, we measured transepithelial potential differences of HBECs mounted in Ussing chambers. A dose-dependent increase in Cl-secretion was observed by ADO addition with an EC50 of 1.6 +/-0.09 µM, similar to the concentration of NECA required to regulate ASL height. To demonstrate which ADO-R are expressed in human airways in vivo, we used laser capture to obtain ciliated cells from frozen tissue sections of excised donor bronchi. Relative expression of A2B-R by qPCR was found to be 4.42 +/-0.30 fold higher than the A1, A2A-R, or A3 receptors (n=3). To confirm that ADO-R activate CFTR in vivo, we measured nasal potential differences. We found that ADO activated chloride secretion at similar levels to isoproterenol. The effect of ADO was inhibited by pre-exposure to 30 µm caffeine (ADO, 20.2 +/-2.5 mV; ADO + caffeine, 5.6 +/-4.9mV). We conclude that A2B-R has an important role in ASL homeostasis through regulation of CFTR and it is active in human airway epithelia under normal physiological conditions. Through osmotic forces, hypertonic saline (HS) may increase the volume of CF airway surface liquid, restore mucus clearance, and improve lung function. The Na + channel antagonist amiloride is predicted to increase the duration of action of HS. However, patients taking amiloride with HS showed less clinical benefits than patients taking placebo with HS. In vitro data suggested that amiloride inhibits transepithelial water flux (pF) in cystic fibrosis airway epithelia. We have extended this study by looking at the effects of amiloride on aquaporins in a heterologous expression system (MDCK cells) and have used immunofluorescence to determine which aquaporins are expressed in human airways in vivo and in vitro. Inhibition of ENaC by aprotinin pretreatment had no effect on HS-induced pF (n=6), suggesting that amiloride did not exert its actions on pF by inhibiting ENaC. However, since HS-induced pF was also inhibited with HgCl 2 , we speculated that amiloride directly blocked an aquaporin. To test this hypothesis, we measured the rate of cell shrinkage of wild-type MDCK cells, and those stably-expressing AQP1 or 5, or expressing AQP3 (after 24 h exposure to a mildly hypertonic media; 400 mOsm). Following mucosal exposure to a 450 mOsm hypertonic Ringer solution, MDCK cultures grown to confluency on permeable filters shrunk by 8% after 30 s (n = 6), as measured by XZ confocal microscopy following staining with calcein-AM. In contrast MDCK/AQP1 or MDCK/AQP3 cultures shrunk by 20% over the same period and MDCK/AQP5 cultures shrunk by 35% (all n=6). A 10 min pretreatment with 200 µM amiloride abolished hypertonic Ringer-induced cell shrinkage in MDCK/AQP5 cultures (93% at 30 s; n=6) but had no affect on cell shrinkage in AQP1 or AQP3 expressing cultures. Amiloride is fluorescent and can be imaged in epithelia by XZ confocal scanning under UV excitation. Amiloride was not internalized immediately (<10 s) before hypertonic Ringer addition and did not block AQP5-mediated cell shrinkage (69% at 30 s; n = 6). However, a 10 min pretreatment with amiloride caused it be internalized and at this time AQP5-induced cell shrinkage was blocked, suggesting that amiloride blocks AQP5 intracellularly. To confirm that AQP5 is present in airway epithelia, we performed immunostaining with an AQP5 antibody and found that AQP5 is highly expressed at the apical membrane of ciliated cells in human bronchial tissue and well-differentiated bronchial cultures. We conclude that amiloride blocks AQP5, which is abundantly expressed in the superficial airway epithelia.

Supported We have previously observed that CFTR, annexin 1 (A1) and cytosolic phospholipase A2 (cPLA2) are partially recruited in detergent insoluble microdomains (DIM) upon proinflammatory stimulation. This is prevented by specific inhibition of CFTR. A1 participates in the regulation of inflammation by inhibiting cPLA2 activity. We have also shown that A1 expression is decreased in cftr-/-mouse tissues. We postulate that CFTR, A1 and cPLA2 participate in the regulation of inflammation by their dynamic interaction within DIM. In this study we aimed to (i) demonstrate and characterize this interaction, and (ii) identify other potential partners of this complex in DIM. By coimmunoprecipitation, we examined the interaction between CFTR, A1 and cPLA2 in human pulmonary epithelial cells CALU3. To search for additional partners we developed a new proteomic approach based on double SDS-Page (dSDS-Page), which is compatible with membrane protein analysis. CALU3 cells and human epithelial bronchial CFBE cells (expressing either WT or F508delCFTR) were subjected or not to proinflammatory conditions (TNFα). DIM were isolated by density gradient and analyzed by dSDS-Page. CFTR immunoprecipitated with A1 and cPLA2 in CALU3 cells. dSDS-Page analyses showed significant differential expression of cytokeratin 18, actin and protein disulfide isomerase (PDI, involved in protein folding) in DIM between control and proinflammatory conditions in CALU3 cells, and between WT and F508del CFBE cells, as determined by mass spectrometry. In conclusion, CFTR, A1 and cPLA2 interact in basal conditions. Proinflammatory treatment causes differential expression of cytoskeletal proteins and PDI depending on the expression of normal and mutated CFTR.

This work suggests that CFTR could participate in the regulation of inflammation within a complex with A1 and cPLA2. Cytoskeletal proteins and PDI could be involved in the dynamics of this complex. This could be related to the abnormal inflammatory response characteristic of CF.

We Acute and chronic inflammation, common features of CF airways disease, have been linked to higher ATP release and changes in ATP metabolic patterns in vivo, e.g., in induced sputum and exhaled breath condensate from CF patients and bronchoalveolar lavage fluid from inflamed lobes (Esther, Ped Pulm sppl 29, 2006) . We, therefore, investigated the contribution of airway epithelia per se to the raised ATP concentrations on CF airway surface during inflammation.

When exposed to supernatant of mucopurulent material (SMM) from CF airways for 48 h, or infected with respiratory syncytial virus (RSV) 72 h prior to the study, basal ATP release from well-differentiated primary human bronchial epithelial (HBE) cultures did not differ from that of control cultures. However, basal ATP release was increased 6 weeks after RSV infection when inflammation had waned (as indexed by IL-8 secretion). The increase in basal ATP release correlated with increased basal UDP-glucose and mucin release, and goblet cell metaplasia. These data indicate that the increased basal ATP release following RSV infection reflected increased goblet cell secretion.

A different pattern was observed for stimulated ATP release. HBEs exposed to SMM for 48 h, or infected with RSV 72 h prior, released 2-3 times more ATP following hypotonic challenge than control cultures (Okada, Ped Pulm sppl, 2006) . The increase in peak ATP concentrations correlated with the degree of inflammation as indexed by IL-8 secretion and the increase in intracellular calcium (Ca 2+ i ) mobilization in response to hypotonic challenge, but not with the goblet cell number. Ca 2+ i mobilization in response to hypotonic challenge in control HBEs was almost completely due to extracellular Ca 2+ i influx which was inhibited by antagonists of a transient receptor potential V4 (TRPV4) Ca 2+ channel. In contrast, an additional component of ATP-induced Ca 2+ i release from endoplasmic reticulum (ER) was observed in SMM-treated HBEs. The increase in hypotonicity-induced ATP release in SMM-treated HBEs compared to control HBEs was almost completely sensitive to chelation of Ca 2+ i by BAPTA (10 µM; 1 h), or inhibition of exocytosis by brefeldin A (5 µM; 2.5 h), monensin (10 µM; 18 h) or N-ethylmaleimide (1 mM; 15 min), whereas these reagents were ineffective against hypotonicity-induced ATP release from control HBEs. These data suggest a positive feedback loop between the increased ATP release linked to Ca 2+ i -dependent exocytosis and the increase in ATP-induced Ca 2+ i mobolization linked to increased ER Ca 2+ i stores (Ribeiro, JBC 280, 2005) in inflamed HBEs.

In conclusion, we propose that basal and hypotonicity-induced ATP release from HBEs are regulated in distinct mechanisms during inflammation. Basal ATP release reflects continuous exocytosis processes independent of Ca 2+ i mobilization and differs as a function of cell type (ciliated vs goblet cell), whereas acute inflammation upregulates hypotonicity-induced ATP release via increased Ca 2+ i mobilization and Ca 2+ i -dependent exocytosis. We conclude that airway epithelia exhibit multiple mechanisms of increasing ATP release into the lumen in response to inflammation.

Supported by CFF grant OKADA06I0 to SFO and grants from NIH.

The main biological role of Vitamin D is in control of bone and mineral metabolism, but it has also been discovered to have many immune system functions. Vitamin D, a fat soluble vitamin, is poorly absorbed in CF leading to low serum levels. We sought to investigate the involvement of Vitamin D in the control of inflammation in a CF model.

Methods CF tracheal epithelial cells (CFTE) were pretreated with vitamin D (1, 25 [OH]2D3) or control and exposed to P aeruginosa lipopolysaccharide (LPS) for 24 hrs before collecting the cell supernatants. Cytokine levels in the supernatants were assessed by cytokine array and subsequent ELISAs specific for IL-6, IL-8 and MCP-1. Vitamin D was measured in the plasma of 37 children with cystic fibrosis aged 3 months to 18 years. Neutrophil numbers, Neutrophil elastase (NE) activity and IL-8 concentrations were measured in the bronchoalveolar lavage (BAL) of these children. Children with cystic fibrosis were deemed to be non-colonised, intermittently colonised or chronically colonised with P aeruginosa on the basis of the Leeds criteria.

Pretreatment with vitamin D significantly reduced secretion of IL-6, IL-8 and MCP-1 from CFTEs after challenge with P aeruginosa LPS in a dose dependant manner. Plasma Vitamin D levels were highest in the noncolonised group and lowest in the chronically colonised group with intermediate levels in the intermittently colonised group. BAL Neutrophil counts, NE activity and IL-8 concentration showed a reciprocal pattern.

Conclusions Vitamin D acts on CF lung epithelial cells to protect against excessive cytokine production in response to LPS. Levels of Vitamin D are inversely related to both P aeruginosa colonisation status and BAL markers of inflammation. Low levels of Vitamin D may contribute to advancing lung disease in children with Cystic Fibrosis. More aggressive correction of vitamin D deficiency may have a role in protecting against excessive inflammatory response to infection.

Sleeping in a mist tent has been used as a treatment for cystic fibrosis (CF) patients, in order to hydrate the viscous mucus and make it easier to be removed by coughing (Matthews et al., 1968) . However, the efficiency of the method has been questioned, and its use was largely discontinued. With a new method to measure the ion content of human nasal fluid the effect of sleeping in a moisture tent on the ion content of nasal airway surface liquid (ASL) in CF-patients and healthy controls was determined. The CF-patients and healthy controls spent a night (8h) in a mist tent, and samples of the nasal ASL were taken before the experiment, after the period in the tent, and then at each hour during 4h after the persons had left the tent. Samples of nasal fluid were collected on Sephadex G-25 beads that had been mounted on strips of filter paper, which were then put into the nasal cavity of the CF patients or controls for 10-15 min. The strips were removed, and the beads were isolated, dried, and analyzed by X-ray microanalysis (Vanthanouvong et al., 2006) .

The concentration of Na, Cl, and K in the nasal fluid of CF patients was 132, 147, and 50 mM, respectively, before the patient entered the tent, significantly higher than the levels in the nasal fluid of the controls (Na 118 mM, Cl 107 mM, and K 19 mM). During the period in the tent, the ion content decreased, to levels of Na 44 mM, Cl 88 mM, and K 19 mM (CFpatients) and Na 32 mM, Cl 45 mM, and K 9 mM (controls) immediately after leaving the tent. After leaving the mist tent the ion levels in the nasal fluid increased, reaching after 4 h values of Na 131 mM, Cl 176 mM, and K 59 mM (CF-patients) and Na 133 mM, Cl 157 mM and K 37 mM (controls), which was for both groups higher than before entering the tent. No major changes in the ion content of the ASL occurred after 4h.

The salt content of the ASL may be relevant in CF, since the ASL is known to contain anti-bacterial proteins, defensins, which are sensitive to the salt concentration (Bals et al., 1998) . Hence, the higher salt concentra-tions in the ASL of CF-patients may have negative consequences for the anti-bacterial defense system in the lungs, and conversely, the decrease in ion concentrations, caused by spending time in a mist tent, may have positive effects. (We have previously shown that none of our patients was chronically colonized with Pseudomonas aeruginosa while regularly sleeping in mist tents.) However, with currently used procedures, the effect of sleeping in a mist tent on the ion content of the nasal ASL is short-lived.

Bals Background: Antibody microarrays for clinical applications have been anticipated for more than a decade but the technology has only recently become sufficiently mature for reproducible, robust detection of low abundance proteins. The advantage of this new technology is the high throughput, parallel detection of identified, known, low abundance proteins. Antibody microarray results are usually given in terms of ratios between two samples (e.g., "experiment" and "control"). However, given the printing and calibration issues inherent with antibody microarrays, this ratio'ing approach precludes statistically valid inter-array comparisons of the individual protein concentrations. We have therefore developed and implemented a different calibration strategy using a benchmark mixture that is applied to every array as an internal standard, thus controlling for any differences in antibody activities as well as printing imperfections. The amount of protein bound to each spot is determined relative to the respective protein in the internal standard. Since all samples in the study are compared to the same internal standard, this semi-quantitative approach permits the comparison of multiple samples. All data for the same antibody on multiple arrays can then be used to calculate the average and standard deviation for a given population (such as patients or control groups) in a parallel fashion.

Methods: CF lung epithelial cultured cells IB3-1 and the CFTRrepaired IB3-1/S9 were biosynthetically labeled with a 1 hour pulse of 35 [S]methionine, and then chased with cold methionine for 2, 4 or 6 hours. We then isolated total protein from each sample, and labeled the proteins with Cy3. Labeled proteins were applied to Clontech® antibody arrays (507 antibodies printed in duplicates) according to the manufacturer's protocols. The internal standard, consisting of a mixture of cell cultures and tissue extracts known to contain the cognate protein for each antibody on the array, was labeled with Cy5. The data were then used to compare the CF and the corrected cells in terms of amount, biosynthesis rate and degradation half-life (t 1/2 ) for each one of the 500 features on the array.

Data Analysis: For each spot on the array the ratio of Cy3/Cy5 (after background subtraction) was calculated, normalized and stored for further analysis. For "pulse-chase" experiments the ratio of 35 [S] counts to Cy3 at each time point gives a measure of the fraction of newly synthesized (radiolabeled) to total protein (dye labeled) bound to each given antibody on the array. The addition of the Cy5 labeled internal standard enables the calibration of these totals, allowing the comparison all the spots for any given antibody on multiple arrays. Furthermore, the internal standard calibration makes it possible to compare not only t 1/2 values but also to calculate the relative degradation rates in different cell lines even when the amounts of that protein are different in the "control" and "experiment" cells.

The new approach quantitatively identifies and subjects to analysis the effect of the CFTR mutation on protein biosynthesis and degradation of the cellular signaling proteome.

Supported by NIH NO1-HV-28287 (HBP) and RO1-53051 (HBP) In CF patients, defective apical chloride secretion, due to lack of functional CFTR, leads to dehydration of airway surface liquid (ASL) and mucus plugging. This defect is likely ameliorated, in part, by an alternative route of chloride conductance, the calcium activated chloride channel (CaCC) . Female gender appears to be associated with more rapid decline of lung function and earlier death in CF. This suggests that gender differences in hormonal environments might influence ASL height regulation, thus modifying the pulmonary phenotype. We hypothesized that changes in estrogen concentrations might alter ASL hydration by affecting CaCC. Consistent with this, we have shown that estrogen attenuates increases in intracellular calcium and ASL height that follow purinergic receptor activation in primary cultures of human bronchial epithelial cells (HBECs, see accompanying abstract by Hengrui Sun et al). Here, we report correlations between estrogen levels and nasal ion transport in vivo and further probe hormonal effects on ion transport and ASL height regulation in vitro.

In vivo studies: In 10 healthy, spontaneously menstruating females, nasal potential difference (NPD) was measured (both nostrils) on two occasions during a single monthly cycle. Based on a history of the 3 most recent menstrual cycles, we measured NPD on a day when estrogen levels were predicted to be low (generally 8-9 days following onset of menses) and subsequently on the day when the pre-ovulatory estrogen surge was predicted. The recording electrode was positioned at the location where maximal stable basal PD was detected and changes in PD recorded continuously during sequential perfusion with 1] Ringer's solution, 2] Ringer's solution containing amiloride (to inhibit sodium absorption via ENaC), 3] modified (low chloride) Ringer's containing amiloride and finally 4] a similar solution also containing UTP (to activate CaCC). As predicted, blood estrogen levels were higher on the "pre-ovulatory" day (0.44±0.01 vs 0.10±0.09 nM, p<0.001). UTP-stimulated change in PD was significantly lower on the study day when estrogen levels were elevated (13.5±1.77 vs 18.9±0.7 mVs, p<0.05). Further, the amiloride-sensitive component of basal PD was higher on this day (14.1±1.7 vs 9.4±2.8 mVs, p<0.05). The change in PD in response to perfusion with a low chloride solution was not different.

In vitro studies: Expression of estrogen receptor (ER) α (but not β) was confirmed in non-CF and CF HBECs by real time PCR. ERα was detected in HBECs from males and female donors. In CF and non-CF HBECs mounted in modified Ussing chambers, exposure to 10 nM E 2 was associated with a significant (~25%) reduction in UTP-stimulated short circuit current. E 2 inhibited the ATP-stimulated increase in ASL height in male and female HBECs and this ASL regulatory phenotype was not reproduced by testosterone or progesterone.

In conclusion, estrogen inhibits UTP-activated chloride secretion in vivo and in vitro suggesting that elevated pre-ovulatory estrogen levels may further impair mucociliary clearance in females with CF.

All data mean±SEM. Supported by CFF Leroy Matthews Award [1] . Data demonstrating an upregulation of CaCCs when CFTR is absent or defective [2, 3] have triggered further studies of these channels and prompted further elucidation of the functional (and possibly also physical) interactions between these two channels. Although searched for long, the molecular identity of CaCCs is still under debate. The role of bestrophin family proteins as putative candidates for CaCCs is discussed controversially. Bestrophin 1 (Best-1) has been proposed to form Ca 2+ activated Clchannels in epithelial cells. Moreover, CaCCs have been shown to support cell proliferation, namely in the development of cancer [4] .

Here, our goal is two-fold: 1) to investigate the correlation between expression levels of Best-1 and epithelial cell proliferation; and 2) to pursue a search for interacting protein partners of Best-1 to better understand the role and function of bestrophin(s) in epithelial cells.

To study the correlation between Best-1 expression and cell proliferation, we analyzed two populations of the T84 colonic carcinoma cell line with very different proliferation rates: the original (T84-slow) and the spontaneously transformed T84 cells (T84-fast). We observed that T84-slow cells have a small proliferation rate and express low amounts of Best-1, while T84-fast cells express high levels of this protein. Best-1 is spontaneously active in T84-fast cells. Best-1-RNAi inhibited Ca 2+ activated Clconductance and cell proliferation, therefore establishing a novel role of bestrophins in cell proliferation, which may be of relevance for the regeneration of the epithelia in CF and also for alternative therapies for CF.

For the second objective, two putative cytoplasmic domains of Best1, polyhistidine -tagged (pHis) N-term (aa 1-30, Best1-N) and C-term (aa 291-584, Best1-C), were cloned into the pET-SUMO bacterial expression vector. These domains were then immobilized onto metal-affinity resin to capture interacting proteins from human T84 whole cell and sub-cellular lysates. Protein-containing fractions recovered from Best1-N/C-coated and blank resins were subjected to 2D-gel and protein identification. It is expected that functional characterization of the interaction between the captured proteins and Best-1 will give new insights into the biological role, plausibly of relevance to CF.

Work supported by DFG-SFB699-A7 grant (Germany) and pluriannual funding of CIGMH (FEDER-EU and FCT, Portugal Extracellular nucleotides acting on epithelial cell surface purinergic receptors regulate the mucociliary clearance process that removes noxious materials from the lung. However, how epithelial cells release nucleotides into the airways remains largely unknown. CFTR has been suggested to mediate ATP release in airway epithelia, but conclusive demonstration remains elusive. CalU-3 is a human airway derived cell line often described as a model of airway gland serous cells. At least 70% of cells in a monolayer expressed high levels of CFTR protein and activity at the apical membrane. ATP, adenosine, and UDP-glucose, the naturally occurring agonists for the P2Y2, A2b, and P2Y14 receptors, respectively, were quantified in airway surface liquid with nanomolar sensitivity using radiolabeling and fluorescence HPLC-based techniques, and real-time ATP measurements. ATP (and UDP-glucose) was found to be released constitutively and selectively onto the mucosal surface of Calu-3 cells. However, the specific CFTR inhibitor 172 (10 µM, 10-30 min) failed to decrease ATP concentration in the mucosal bath of resting CalU-3 cells. Forskolin (10 µM, 10 min) stimulated CFTR channel activity, but did not affect basal levels of ATP (or UDPglucose) release. In contrast, elevation of intracellular calcium using ionomycin (1-10 µM, 1-10 min) significantly increased ATP (and UDP-glucose) release into the mucosal (but not serosal) bath of CalU-3 cells. Calciummediated nucleotide release enhancement was not abolished by CFTR inhibitor 172. Maneuvers like pre-incubation in nominally free-calcium solution, BAPTA (10 µM, 30 min), or cytochalasin D (5µM, 30 min) to disrupt actin cytoskeleton, which inhibited regulated exocytosis in CalU-3 cells, also reduced calcium-mediated increase in nucleotide release. CalU-3 monolayers presented a fraction of cells (<30%) exhibiting negligible CFTR immunostaining signal, but displaying airway goblet cell mucin granules. Remarkably, stimulation of ATP and UDP-sugar release with ionomycin resulted in concomitant enhanced exocytotic secretion of MUC5AC. These results suggest that CFTR is not involved in regulation of ATP release in airway epithelial cells. More likely, nucleotides and nucleotide-sugars residing in the biosynthetic pathway involved in modification of apically targeted glycoconjugates (e.g. mucins) are released to the extracellular environment via calcium-triggered exocytosis.

Supported Although the regulation of Na+ transport is relatively well described in normal and CF airway epithelial cells, the impact of inflammation and infection, two major components of CF lung disease, on Na+ transport has not been studied extensively. Significant amounts of evidence suggest that bacterial by-products may influence Na+ transport in airway epithelium. Furthermore, the secretion of exoenzyme S by Pseudomonas aeruginosa is enhanced during an exacerbation of lung disease in cystic fibrosis. In the past few years, our laboratory has shown that exposition of lung epithelial cells to P. aeruginosa LPS leads to a rapid decrease in activity and expression of ENaC channel. The aim of our study was to evaluate the mechanism involved in the modulation of ENaC expression and activity by LPS. Primary culture of lung epithelial cells grown on filters were treated with LPS (15µg/ml) for different time (30min, 1, 2, 4h). At each time point, the total and amiloride-sensitive short circuit current (Isc) was determined in Ussing chamber. In parallel, the cell proteins were collected to investigate by western blot the signalling pathways involved in this response to LPS. A 4 hours LPS treatment leads to a significant decrease in alpha and gamma ENaC mRNA to 44% and 53% of control respectively. Similarly, the amiloridesensitive current was decreased by 53% of the current in untreated cells after a 30 min exposure to LPS. By western blot, we demonstrated the strong implication of the PI3K/Akt pathway in response to LPS in the regulation ENaC mRNA and the potential implication of tyrosine/kinase receptors signalling pathway through mTOR in the regulation of ENaC channel activity.

In conclusion, we demonstrate that inflammatory molecules can modulate the expression and function of the Na+ transport mechanism in lung epithelial cells and could contribute to the modulation of airway surface fluid in the chronic infectious process associated with CF. Serous acinar cells are thought to be crucial for secretion of airway surface liquid by submucosal glands in large airways, but the molecular mechanisms of serous cell ion and fluid transport are not well understood. To address this, we isolated murine nasal serous acinar cells and studied them using simultaneous DIC and quantitative fluorescence microscopy to track changes in cell volume and intracellular concentrations of ions involved in fluid secretion ([Cl -] i ; SPQ) and its regulation ([Ca 2+ ] i ; fura-2). Previous results indicated that serous acinar cell volume changes were indicative of solute efflux (shrinking) and solute influx (swelling), reflecting the secretory state of the cell. Acinar cells stimulated with 100 µM carbachol (CCh) exhibited a rapid increase in [Ca 2+ ] i and subsequent ~ 20 % cell shrinkage due to efflux of ~ 67 % of cell Clcontent and parallel loss of K + and osmotically obliged H 2 O. Because murine serous cells are sites of apical CFTR expression in the airway, we tested the requirement for CFTR in CCh/Ca 2+activated shrinkage/Clefflux. Acinar cells isolated from either cftr tm1Unc-/mice or Wt cells pre-treated with CFTR inh 172 exhibited identical rates and magnitudes of Ca 2+ -induced cell shrinkage and Clloss compared to Wt control cells. Resting [Cl -] i was also identical in Wt and cftr tm1Unc-/serous acinar cells (65 ± 4 and 66 ± 2 mM, respectively), suggesting similar levels of resting Clconductance. Thus, murine serous acinar cells can secrete KCl/H 2 O by Ca 2+ -dependent mechanisms in the absence of functional CFTR, in agreement with studies of intact glands suggesting CCh-induced glandular fluid secretion is CFTR-independent. Wt serous acinar cells were stimulated with cAMP-agonists, including 15 µM forskolin, 100 µM isoproterenol, 200 µM adenosine, and 3 µM VIP. However, no cAMP-induced shrinkage was observed, even in the presence of a cocktail of inhibitors to block solute uptake. To determine if CFTR activation could enhance CCh/Ca 2+ -induced cell shrinkage at a sub-optimal [CCh], serous acinar cells were pre-treated with forskolin followed by stimulation with 1 µM CCh in the continued presence of forskolin. However, similar rates and magnitudes of cell shrinkage were observed in forskolin-treated and control cells. These results suggest that the magnitude of the CFTR Clconductance in murine serous acinar cells is much smaller than the Ca 2+ -activated Clconductance, below the resolution of these optical assays. More sensitive electrophysiological approaches are being used to determine the molecular identity of the non-CFTR Clconductance(s) and the role of CFTR in murine serous acinar cell ion transport. Females with CF exhibit more rapid decline in FEV1 than males, which reduces their life span. Normal airway epithelia regulate airway surface liquid (ASL) volume and mucus transport by secretion through both CFTR and the Ca2+-activated Cl-channel (CaCC), and inhibition of either of these pathways in isolation does not abolish mucus transport. However, a loss of both pathways is predicted to result in mucus plugging which contributes to lung disease. Despite a loss of CFTR, CaCC is functional in CF airways due to mechanical stimulation of ATP release and activation of the P2Y2 pathway which serves to raise intracellular Ca2+. We hypothesized that E2 may inhibit CaCC, leaving females with CF more prone to lung disease than males due to a greater depletion of ASL volume. Using confocal microscopy, we found that acute pre-exposure with E2 inhibited ATP induced ASL secretion in normal and CF human bronchial epithelial cultures (HBECs) from both male and female donors. The IC50 for E2-inhibition of ATP-induced ASL secretion was 0.75 nM. To better understand how estrogen receptors (ERs) interact with the P2Y2 pathway, we transiently transfected ER-α or ER-β linked to an orange fluorescent protein in an ER null BHK cell line. With the co-transfection of HA tagged P2Y2-R, we observed that E2 addition did not induce internalization of P2Y2-R, neither did it alter the internalization induced by ATP. We then measured total cell inositol phosphates (IP) in CF airway cells +/-E2 pretreatment. Cell IP levels were not different between groups (n=9), suggesting that the effect of E2 is downstream of IP production. The investigation of the changes of the intracellular Ca2+ (Ca2+i) by recording Fura-2 emission ratio over time in ER-α or ER-β expressed cells provided clues. After 10 nM E2 exposure, the Fura-2 emission ratio in untransfected cells (n=7) or ER-β transfected cells (n=7) increased by 0.6 following 10 uM ATP addition. This increase was significantly inhibited by~40% in ER-α transfected cells (n=7), providing good evidence that ER-α is respon-sible for inhibiting the P2Y2 pathway. We next measured Ca2+i in normal HBECs. While acute 10 nM E2 addition had no affect on basal Ca2+i levels, which were around 370 nM, this maneuver reduced the Ca2+i response to 70 nM following ATP addition (both n=9), suggesting that E2 addition altered Ca2+ signaling in response to ATP. Thapsigargin induces Ca2+i release from the endoplasmic reticulum, but the magnitude of Ca2+i release after thapsigargin addition was not altered by E2 pretreatment, indicating that E2 works further up stream to affect the P2Y2 signaling pathway. Together, these results suggest that E2 directly interacts with the P2Y2 signaling pathway through ER-α by altering Ca2+i. We propose that this interaction leaves females more vulnerable to infections due to a reduced ASL volume during periods of high E2, such as occur prior to ovulation. The nasal potential difference (or NPD) is a sensitive, real time bioelectric assay of CFTR-dependent and independent ion transport that is a commonly used study endpoint in early clinical stages of drug development. Technical limitations of the assay must be controlled to facilitate its use in multicenter trials and limited intra-site variability.

Two forms of equipment are commonly employed to evaluate nasal potential difference in humans: Silver-silver chloride electrodes (AgCl) with saline bridges or calomel electrodes with agar bridges; however, these techniques have never been directly compared. We repeatedly evaluated (within 7 days) the basal potential difference of five normal subjects using the two available techniques, then compared variance and repeatability of the measures. Basal PD was measured during perfusion of Ringer's solution, and measured at 0.5, 1.0, 1.5, 2.0, and 3.0 cm within the inferior meatus, as indicated in the CF-TDN standard operating procedures. Mean basal PD were within 0.8 mV using the two techniques, with very similar coefficients of variation (CVar; 0.36 vs. 0.32 in AgCl vs. calomel, respectively). Maximum basal PDs were also similar between the two methodologies (mean difference 1.5 mV; CVar 0.32 and 0.28, respectively). The correlation between first and second measures of mean PD was highly reproducible, and similar between the two techniques (r=0.94 vs. 0.91 in AgCl vs. calomel, respectively); however, when each individual basal PD measure was compared, correlation between first and second PD using the AgCl device appeared superior (r=0.90 vs. 0.60 respectively; n=25 measures per device). Comparison of values obtained in CF subjects are currently in progress and will also be reported. Next, we examined diagnostic NPD tracings in 12 CF subjects for whether the effect of amiloride might influence values obtained in the contralateral nare. An effect of cross-contamination by amiloride or other agent (gene therapy vectors or other nasal administered compounds, for example) could influence accurate measure of NPD values. When basal PDs were performed in both nostrils prior to amiloride exposure in either nostril, mean basal PD between nares remained stable (<1 mV, p=NS for change). However, if the basal PD in one nostril was measured after amiloride exposure in the contralateral side, mean PD depolarized by 12 mV (from -50 to -38 mV; p<0.01), indicating the presence of a 'cross-nostril' amiloride effect that might potentially affect interpretation of this endpoint if not performed in a sequence designed to avoid amiloride exposure. In summary, we show that calomel and AgCl electrodes perform similarly in repeated measures from the same normal subjects, and that improper NPD sequence can adversely affect basal PD measures. Other potential impediments to the PD measurement (e.g. solution temperature, tonicity, ionic content, and sources of electrical offset) will also be discussed. These findings are relevant to the optimizing the use of the NPD assay in clinical trials.

Supported by NIH, CFF, and CFFT. The gene defective in patients with cystic fibrosis (CF) is the cystic fibrosis transmembrane conductance regulator (CFTR), a cAMP-activated chloride channel in the apical membrane of epithelia, but alteration of expression and function of many genes undoubtedly contributes to the pleiotropic manifestations of CF. A hallmark of CF in airway epithelia is the hyperabsorption of sodium through the epithelial sodium channel (ENaC) in the apical membrane followed by extrusion through the Na,K-ATPase in the basolateral membrane. This results in airway surface liquid (ASL) dehydration, mucus buildup and bacterial infection characteristically observed in the lungs of CF patients. Previous studies have shown that the Na,K-ATPase is increased in CF airway epithelia, consistent with increased sodium transport along the ENaC/Na,K-ATPase axis. We have previously demonstrated that FXYD5, a member of a small family of proteins known to regulate the Na,K-ATPase, is increased in S489X CF mouse lung and nasal epithelia and in primary human tracheal epithelial cells (HTE) after treatment with an inhibitor of CFTR (172). Recent evidence has indicated that ENaC activation occurs through modification of the serine protease-protease inhibitor balance in ASL. Thus, we hypothesized that ENaC-mediated increases in sodium absorption may be partially responsible for the increase in FXYD5 expression observed in CF airway epithelia. We now report that increasing ASL volume on HTE cells upregulated FXYD5 expression (P<0.01) after 24 hours, an effect that was further increased after trypsin addition and abrogated by the serine-protease inhibitor aprotinin. Treatment of HTE cells with nystatin also increased FXYD5 expression measured by quantitative RT-PCR (P<0.01). Based on these results and the observation that mice overexpressing ENaC beta-subunit exhibit cystic fibrosis-like lung pathophysiology, we examined FXYD5 expression in the lungs of mice overexpressing the Scnn1b-transgene. Contrary to our expectations, FXYD5 lung expression was decreased as assessed by quantitative RT-PCR and immunoblot analysis (P<0.005), indicating FXYD5 may be alternatively regulated in epithelia under conditions of acute versus chronic sodium hyperabsorption. Therefore, we measured FXYD5 expression in an epithelial cell culture model using an inducible, tri-cistronic vector capable of expressing ENaC alpha, beta and gamma subunits. We determined that FXYD5 expression is decreased after chronic ENaC induction. We conclude that FXYD5 expression is coordinately regulated based on acute versus chronic ENaC-mediated sodium absorption and suggest that FXYD5 is increased in cystic fibrosis airway epithelia as a result of an imbalance between increased inflammatory signaling and altered ion transport.

Picher, M.; van Heusden, C. Cystic Fibrosis Center, University of North Carolina, Chapel Hill, NC, USA Airway infection by the respiratory syncytial virus (RSV) is considered an important cause of pulmonary exacerbation and hospitalization in cystic fibrosis (CF) children. This disease is characterized by functional mutations in the CF transmembrane regulator (CFTR). The inability of CFTR -/-mice to clear RSV is associated with exaggerated inflammatory responses. A relationship was recently established between airway inflammation and chronically-elevated airway adenosine. This signaling molecule regulates key aspects of lung defenses, including bacterial clearance and inflammatory cells. High adenosine levels measured in bronchoalveolar lavage fluid of asthmatic or CF patients cause airway inflammation, remodeling and fibrosis in animal models. The objective of this study was to investigate the longterm effects of RSV infection on adenosine regulation and inflammatory responses in human airway epithelia. Methods. Polarized primary cultures of bronchial epithelial cells from healthy subjects and CF patients were exposed to vehicle, UV-inactivated or active green fluorescent protein (GFP)-RSV. The cultures were monitored 21 days for infection (fluorescence microscopy), adenosine levels (fluorescence chromatography), the release of cytokines regulating adenosine TNFα) and enzyme activities (ecto 5'-nucleotidase [ecto 5'-NT: AMP Ç adenosine] and adenosine deaminase 1 [ADA1: adenosine Ç inosine]). Results. (A) Both normal and CF cultures exhibited detectable RSV infection over 21 days, with peak density on day 3-4. (B) We demonstrate that RSV enhances the release of all five cytokines from normal cultures over 21 days. In contrast, CF cultures responded to RSV by a transient decrease in IL-1β, IL-4 and TNFα (peak on day 3), but a sustained decrease in IL-12, and a sustained increase in IL-13, secretion. (C) On normal epithelia, RSV transiently stimulated ecto 5'-NT (peak on day 3) but caused a sustained increase in ADA1 activity, resulting in chronically-low adenosine levels. In CF, RSV raised adenosine levels by the combined effects of a sustained increase in ecto 5'-NT activity and the lack ADA1 response. (D) Chronically-elevated adenosine generated by ADA1 inhibition [erythro9-(2-hydroxy-3-nonyl)adenine] stimulated the secretion of all cytokines, except IL-12. Conclusion. These data support a complex relationship between adenosine and cytokine regulation in human airway epithelia. The IL-13-adenosine amplification pathways identified in RSV-infected CF epithelia are in agreement with animal models and their key role in chronic airway inflammation/remodeling. On the other hand, the data also suggest that ADA1 activity is influenced by IL-12, known to stimulate the expression of a major protein anchoring soluble ADA1 to epithelial surfaces: CD26. Altogether, this study suggests that an RSV infection in CF children may accelerate the progression of the disease by setting the stage for airway adenosine-cytokine amplification pathways. Supported by the Cystic Fibrosis Foundation (Picher 05G0). Cystic fibrosis (CF) airway submucosal glands are defective in mucus secretion rate (Jayaraman, PNAS, 2001, 98, 8119; Joo, J Biol Chem, 2002, 277, 50710; Joo, Pediatr. Pulmonol., 2003, Suppl. 70, 281; Wine, Proc ATS, 2004, 1, 47; Thiagarajah, Faseb J, 2004, 18, 875; Song, AJP Cell, 2005; Salinas, Faseb J, 2005, 19, 431) . One hypothesis is that loss of CFTR results in reduced secretion of electrolytes leading to underhydration of mucins, thereby altering the mucus properties. It has been generally believed that the CF mucus in response to cholinergic agonists is thicker and more viscous. However, comparisons of pure gland mucus are difficult because of variability in gland secretion rates and a decline in mucus solids with repeated stimulations (Wu, Pediatr. Pulmonol. 2006, Suppl. 29, 252) . In the present study, we directly compared the solid content of airway submucosal gland mucus between control and CF individuals. Human donor tracheas (HN) or disease control bronchi (DC) were dissected and mounted as previously described but without mineral oil. Tissue was kept warmed and humidified in a chamber with two layers of saturated 37 degree C 95/5% CO 2 /O 2 water vapor. The mucosal surface was blotted dry and coated with a thin layer of dimethylpolysiloxane to prevent re-absorption by the surface epithelium. The tissue surface was covered with a cover slip, leaving minimal air space between it and the surface to further reduce evaporation. A 12 µL droplet of 2% NaCl was placed at the side of every tissue as a standard to assess evaporation and measurement error. Any basal secretions were removed prior to stimulation with 10 µM carbachol. Mucus secretion from individual glands was pooled using forceps and then drawn into a constant bore microcapillary (Drummond) that had been alcohol-cleaned, dried, and weighed. Stimulation time was allowed to vary so that at least 2 µL of mucus was collected. After mucus collection, both the standard and the bath solution were refreshed. Mucus filled capillaries were weighed with a Mettler microbalance before and after overnight 80 degree C oven drying to obtain nonvolatile solid content. The average solids measured in six repeated measures of 2% salt solutions was 2 ± 0.2 % with no trend over time. Results for HN (N=5) and DC (N=3) did not differ and were pooled as 'Control'. The initial Control secretion yielded 6.5 ± 0.8 % solids (mean ± SEM, n=8 subjects) and required 24 ± 3 min to produce adequate mucus. The initial CF secretion gave 9.1 ± 0.5 % solids (n=4) in 31 ± 2 min. The second Control secretion yielded 5.6 ± 0.5 % solids (n=8) and required 40 ± 5 min, whereas the second CF secretion gave 7.4 ± 0.9 % solids (n=4) and required 56 ± 7 min. In summary, CF gland mucus solids content was 40% higher than Control following the first stimulation (P = 0.05). After the second stimulation, CF solids content was 32% higher than the control (P = 0.08). Solids contents of controls observed in our experiment are much higher than was found for mucus collected from airway lumens after 2 hr stimulation (c. 2%, Trout, Am J Physiol, 1998, 274, L258) . Supported by NIDDK RO1-51817 (JJW) and CFF. In patients with cystic fibrosis (CF) as well as in some animal models of CF lung disease, higher levels of prostaglandin E2 (PGE2) have been detected compared to control subjects. It is not clear whether this excess of PGE2 contributes to the airway pathology of CF, or whether it may serve a protective function against airway constriction. Inhaled PGE2 has been tried in asthma and may be of benefit as a bronchodilator. Interestingly, PGE2 has shown benefit in aspirin-sensitive asthmatics, a population that shares some features with the CF airway phenotype including nasal polyposis. The mechanism of airway relaxation to PGE2 is mediated primarily by EP2 receptors signaling through cAMP. In CF, beta-adrenergic bronchodilators are commonly used, but are often only partially effective at reversing bronchospasm. Beta-agonists and PGE2 both signal through G-proteins coupled to cAMP and PKA signaling, thus PGE2 may offer another treatment when beta-agonists are ineffective at reversing bronchospasm in CF. In mouse tracheas, the beta-agonist isoproterenol (ISO) will partially relax a pre-contracted airway, but 47 ± 2% of contractile force remains (n=13). In contrast to the limited relaxation with ISO, constricted mouse airways demonstrate significantly greater relaxation to PGE2, with only 18 ± 3% of force remaining (n=9, p<0.001 compared to ISO). This difference in relaxation responses suggests different intracellular signals are activated by each agonist despite both signaling through PKA. To investigate phosphorylation events in response to ISO or PGE2, we developed a new method for perfusing isolated mouse lungs. Using this technique, tissues were perfused with buffered solution containing 32P for radiolabeling. Lungs were then stimulated with either ISO or PGE2, and phosphoproteins were compared from each condition. Radiolabeling was effective under all three conditions. Targets were chosen from each group (unstimulated, ISO, or PGE2 treated) that showed phosphorylation or dephosphorylation, and proteins were identified by mass spectrometry. Over 20 differentially phosphorylated proteins were identified, and two of these (Rho GDI 1 and Inositol-requiring protein 2) were selected as candidates to be tested in the context of relaxation to PGE2 compared to ISO. By identifying signaling targets involved in airway relaxation to PGE2, it may be possible to activate beneficial airway relaxation responses without the limitations seen in ISO relaxation and without activating PGE2 responses other than relaxation that may link PGE2 to the airway pathology seen in CF. The parental ∆F508/∆F508 CF bronchial epithelial cell line (CFBE41o-) was compared to clones of CFBE41o-complemented with 6.2kB wild type (wt) or 4.7kB ∆F508 CFTR cDNA in terms of their ion transport characteristics. The transepithelial resistance (RT) (~310 (cm2) observed in the parental CFBE41o-was maintained in the complemented cells. Two clones complemented with wt CFTR cDNA (c7-6.2wt and c10-6.2wt) demonstrated RT ~315 Ωcm2, while another clone (c4-4.7∆F) complemented with the ∆F508 CFTR cDNA showed an RT of ~340 Ωcm2. Two stable clones expressing wt CFTR were selected based on electrophysiological characteristics and stability of CFTR expression. Forskolin-stimulated Cl secretion in clone c7-6.2wt was on average 13.4±2.5 µA/cm2 and 41.3±25.3 µA/cm2 in clone c10-6.2wt. Furthermore small cAMP-stimulated transepithelial current of 4.7±0.7 µA/cm2 was observed in clone c4-4.7∆F indicating that ∆F508 CFTR can traffic to the plasma membrane if present at high enough levels at physiological temperatures. The role that CFTR plays in Ca-activated Cl conductance was investigated following treatment of the cells with ATP in parental CFBE41o-and wtCFTR6.2-CFBE41o-monolayers. In symmetrical Cl-containing solutions, mucosal application of ATP (100 µM) elicited a small, transient Cl secretory response in the parental CFBE41ocells (8.5±2.1 µA/cm2), while the magnitude of the ATP-stimulated Cl current was markedly enhanced in CFTR-corrected CFBE41o-cells with peak currents at 29.8±6.4 µA/cm2. Recordings done in presence of a serosal-tomucosal Cl gradient to increase the driving force for Cl secretion showed a transient ATP-dependent activation of Cl currents in parental CFBE41ocells with peak Cl currents of 33.3±3.8 µA/cm2 (n=30). CFTR corrected CFBE41o-cells showed a sustained activation of Cl currents of similar magnitude (42.7±10.5 µA/cm2; n=16). ATP-stimulated Cl currents were effectively blocked by the CFTR inhibitor GlyH101 (20µM). Histochemical analysis showed prominent epithelial characteristics in all clones. These observations indicate that i) the apical driving force is limiting for ATP-stimulated Cl secretion and that ii) CFTR participates in the Cl secretory response to ATP. Furthermore, the findings suggest that these novel clonal isolates of the CFBE41o-bronchial epithelial cell line can be useful for the investigation of CF therapies. This work was supported by grants from the CFF, CFRI, NIH, Elizabeth Nash Foundation, and PACFI. The lactoperoxidase (LPO) system functions in normal airways to prevent infection and may be compromised in cystic fibrosis (CF) due to defective transport of its substrate thiocyanate. A computational model of normal airway surface liquid suggests that hydrogen peroxide levels regulate LPO system activity and that the lower thiocyanate in CF airways may result in elevated levels of hydrogen peroxide that is likely produced by Duox the major NADPH oxidase in airway epithelial cells. To investigate the regulation of Duox by bacterial products, passage 1 normal human airway epithelial cells were grown and redifferentiated at the air liquid interface and treated for 12, 24 or 48h with either Pseudomonas aeruginosa flagellin or LPS, applied apically. Treated and mock treated cultures were analyzed for levels of Duox and LPO mRNA and protein, and for hydrogen peroxide production. Quantitative PCR was performed using Taqman kits, protein was measured by Western blots, and hydrogen peroxide was quantified using amplex red-HRP coupled fluorescence. The data showed the Duox 2 and LPO mRNA but not Duox 1 mRNA, were significantly upregulated at all measured time points following apical treatment of the cultures with purified flagellin compared to mock treated. At 24 h Duox 2 was increased 7.7 +/-3.8 fold and LPO was increased 2.3 +/-0.7 fold (means +/-S.E.M., n = 5 lungs, 1-3 cultures of each lung).

MUC5AC mRNA was also upregulated by flagellin treatment (4.9 +/-1.6 fold, n = 5 lungs, 1-3 cultures of each lung). Apical treatment with LPS had no effect on any tested mRNA and induced minimal IL-8 secretion in these cultures. Western blots using anti-Duox antibodies (courtesy of F. Miot, Brussels, Belgium) showed corresponding increases in Duox protein following flagellin treatments. Hydrogen peroxide production by flagellin treated cultures was increased in proportion to increases in Duox 2 mRNA. Application of flagellin alone to the apical surface of previously untreated cultures did not increase hydrogen peroxide production. Thus, Pseudomonas flagellin up-regulates Duox 2 and results in higher activity in normal human airway epithelial cells perhaps serving to increase innate host defense activity of the LPO system. Chloride ion in phagolysosomes is an essential substrate for neutrophils to produce hypochlorous acid (HOCl), an important bactericidal agent. We have previously shown that CFTR was present in the phagolysosomal membrane of neutrophils and the CFTR defect led to deficient chlorination of ingested Pseudomonas aeruginosa (PA) (Painter et al, Biochemistry, 45:10260-10269, 2006) . These results suggested that chloride accessibility to the phagolysosomal lumen might be impaired in CF neutrophils thereby affecting neutrophil-mediated bacterial killing. To understand the role of chloride ion in the process, we first assessed the effect of extracellular chloride concentration on PA killing by normal neutrophils under iso-osmotic conditions. Surprisingly, bacterial killing was strongly dependent on extracellular chloride concentration. The initial rate of killing of PA was~2 fold less in a chloride-free medium than a physiological chloride medium. Killing efficiency increased as the chloride concentration was raised in a dosedependent fashion and plateaued at~60 mM chloride concentration. To determine what percent of the chloride-dependent killing was oxidantdependent, we applied the CFTR inhibitor GlyH-101, the myeloperoxidase (MPO) inhibitor aminobenzoic acid hydrazide (ABAH) and the NADPH-oxidase inhibitor diphenylene iodium chloride (DPI), respectively to bacterial killing assays. The results showed that the killing of PA could be divided into three discernible components: (1) a component accounting for about 20-25 % of the total killing rate that was inhibited by GlyH-101, ABAH or DPI; (2) a second component amounting to~25-30% that was chloride-dependent and partially sensitive to GlyH-101 and; (3) a third accounting for~50% of the total killing activity that was not sensitive to oxidant inhibitors, or extracellular chloride concentration. DPI and ABAH had no effect on killing of PA in the absence of extracellular chloride suggesting that most of the oxidant-mediated killing activity toward PA was due to HOCl. None of the inhibitors had any significant effect on phagocytosis. Finally, the rate of PA killing by CF neutrophils was found to be significantly lower (30-40 %) than that seen for normal neutrophils and was comparable to the effect of GlyH-101 on killing of PA by normal neutrophils. The data suggest that CFTR plays a significant role in killing of PA by normal neutrophils and, when defective as in CF, may compromise the ability of neutrophils to efficiently kill PA. This research was supported by a CFF grant to GW. 1%. The CF group demonstrated a significantly higher frequency of Pseudomonas respiratory infections than the non-CF group. Interestingly, no significant differences were detected in any infections from other systems including blood, sinuses, skin, wounds, oral cavity, bowel, eyes, peritoneal cavity, and urinary tract. Moreover, the CF lung-transplant patients had significantly less time free from Pseudomonas infections in the transplanted lungs.

CONCLUSION Normal lungs transplanted into CF patients had significantly higher susceptibility to Pseudomonas infections than those transplanted into non-CF patients, which strongly suggests that defective host defense mechanism(s) beyond the respiratory system contribute to CF lung pathogenesis. Pyocyanin (N-methyl-1-hydroxyphenazine, PCN) is produced by Pseudomonas aeruginosa and is found in pulmonary secretions of infected CF patients at concentrations up to 100 µM. PCN is a redox-active compound that generates superoxide and hydrogen peroxide in the presence of cellular reductants. Since CFTR was reported to be modulated by H 2 O 2 , we hypothesized that PCN, through its ability to redox cycle and to produce H 2 O 2 , would affect CFTR-mediated Cl transport. We therefore tested PCN and H 2 O 2 for effects on Cltransport (transepithelial short-circuit currents in Ussing chambers measured with serosal-to-mucosal Clgradient) and cytosolic redox potential (imaging microscopy on cells transfected with a redox sensitive GFP, roGFP1) using CFTR-corrected CF bronchial epithelial cells (wtCFTR6. . Under resting conditions, acute exposure of the apical epithelial surface to 100 µM PCN elicited a small Clsecretory response (7.3±0.9 µA/cm 2 , n=10) that was completely blocked by the CFTR inhibitor GlyH101 (20 µM). In contrast, in presence of the cAMP-elevating agonist forskolin, PCN caused a time-dependent decline of forskolin-stimulated CFTR Clcurrents by 86%. Exposure to H 2 O 2 (100 µM) elicited a more pronounced Clsecretory response than PCN in resting cells and H 2 O 2 -stimulated Clcurrents peaked at 47.4±5.8 µA/cm 2 (n=10). H 2 O 2 was less effective than PCN at inhibiting forskolin-stimulated CFTR Clcurrents and currents declined by 33%. Both PCN and H 2 O 2 oxidized the cytosolic redox potential from a resting value of -270 mV by similar amounts (48±8 mV for PCN vs. 44±10 mV for H 2 O 2 ), but rates of oxidation were faster for H 2 O 2 (2.5±0.1 mV/min) than for PCN (1.1±0.2 mV/min). Inhibition of cAMP-stimulated CFTR Clcurrents by oxidation was a linear function of the cytosolic redox potential, but PCN (2 µA×cm -2 ×mV -1 ) was more potent than H 2 O 2 (0.69 µA×cm -2 ×mV -1 ). CF bronchial epithelial cells homozygous for ∆F508-CFTR did not show a Clsecretory response to PCN or H 2 O 2 . These data suggest that CFTR-corrected CF bronchial epithelial cells respond acutely to oxidative stress by turning on CFTR activity probably as part of the innate host defense response and this mechanism is absent in CF airways. Furthermore, prolonged exposure to PCN and H 2 O 2 is detrimental for the proper function of the cAMP-regulated CFTR Clconductance. We conclude that oxidative stress in P. aeruginosa-infected airways is a key factor that needs to be considered for successfully correcting CFTR function in CF airways.

Supported by NIH (NCCAM P01 AT002620), CFF (Fischer07G0) The lung is continually exposed to environmental agents and pathogens. The lung epithelial lining fluid (ELF) is first to encounter these agents and GSH is a major antioxidant found in this apical fluid. The cystic fibrosis transmembrane conductance regulator protein (CFTR) is the only known GSH transporter that maintains lung ELF GSH levels. Cystic fibrosis (CF) patient have low levels of GSH in their ELF and have copious and viscous mucus. Hypertonic saline is used in CF patient to help clear mucous and improve lung function however the mechanism(s) by which it does so are poorly understood. The purpose of these studies was to examine whether hypertonic (3.0%) saline can modulate apical GSH levels and protect lung epithelial cells against the oxidant, t-butyl hydroperoxide (t-BOOH). We used human lung epithelial cell lines sufficient (C38) and deficient (IB3) in CFTR. The CFTR deficient IB3 cells have 40% lower basal levels of apical GSH as compare to CFTR sufficient C38 cells. Cells were exposed separately and in combination to 3% saline, and t-BOOH (100 uM) for 48 hours. The CFTR deficient IB3 cells were much more sensitive to t-BOOH-mediated oxidative injury as measured by lactate dehydrogenase (LDH) release. Hypertonic saline exposure was associated with an increase in apical GSH levels in both CFTR sufficient and deficient cells and decreased t-BOOHmediated oxidative injury in both cells lines. Hypertonic saline increased GSH in the apical compartment, which appeared to be largely CFTR mediated. We examined this issue in mice given a clinically used dosing regimen of 7% hypertonic saline nebulized twice daily and examined changes in ELF GSH levels 12 hours after last treatment. Mice receiving the hypertonic saline treatment had significantly higher GSH levels in their ELF than untreated controls. This data suggests that CFTR and GSH adaptive responses play an important role in lung's reaction to oxidants. We propose that factors, which interfere with the lung's capability to mount and maintain an adequate adaptive apical GSH response, may compromise and sensitize the lung to oxidative injury. (Supported in part from NIH grant HL075523).

The absence of functional CFTR in airway epithelia of CF patients leads to abnormal airway surface liquid, which favors chronic bacterial infection. This chronic infection in CF lungs is associated with an exaggerated inflammatory response characterized by abnormal cytokines secretion and massive lung neutrophils infiltration.

We have previous recapitulated the abnormal inflammatory response in CFTR-/-(CF) mice by repeated administration of Pseudomonas Aeruginosa lipopolysaccharide (LPS) (Ped. Pulmonology, Supp. 29, 2006) . We demonstrated that the bronchoalveolar lavage (BAL) fluid of CF and heterozygous (HET, CFTR+/-) littermate control mice when compared to the BAL from WT mice shows 1) higher numbers of neutrophils and 2) higher concentrations of 6 cytokines (IL-1alpha, IL-6, KC, MCP-1, GM-CSF, M-CSF) as assessed using multiplex analysis of 22 selected cytokines.

Of note, airway epithelial cells from CF patients are known to have elevated secretion of some of these cytokines. However to date, the contribution of immune cells to these abnormal cytokine levels has not been well characterized. This study aims to investigate the contribution of CF macrophages in this altered cytokine profile/response.

We examined cytokine secretion from two different macrophage populations: alveolar macrophages (AM) obtained from the BAL fluid and bone marrow (BM)-derived macrophages differentiated in presence of M-CSF. Both macrophage populations were cultured in the presence or absence of LPS, and the supernatants were tested for the concentration of the 6 cytokines that we had shown to be abnormal in vivo. These macrophage populations were examined from CF, HET and WT mice.

In absence of LPS exposure, BM-derived macrophages from CF mice have lower levels of IL-6 and KC compared to WT and HET mice. In contrast, AM of CF mice have higher levels of IL-6 and KC than AM of WT mice. This difference may be due to AM's prior exposure to antigens in vivo whereas the BM-derived macrophages are a truly naïve population. Interestingly, after LPS stimulation, IL-1a, IL-6, MCP-1, KC, G-CSF and GM-CSF are higher in CF macrophages compare to WT and HET mice suggesting that CF affected macrophages display an altered inflammatory response when compared to WT and HET. The abnormal production of these cytokines was also detected at the transcriptional level using quantitative RT-PCR analysis.

In conclusion, our data support the hypothesis that CFTR-null macrophages may have an impaired cytokine profile at baseline as well as after stimulation. Furthermore, CFTR deficient macrophages may contribute directly to the abnormal inflammatory response in CF.

(Supported by CFF, NIH, NIDDK)

Cystic fibrosis lung disease is associated with an excess of free neutrophil elastase (NE) in the airway arising from persistent neutrophil inflammation. DX-890 is a potent small protein inhibitor of neutrophil elastase that could be a useful therapy for diseases involving neutrophilic inflammation such as cystic fibrosis and COPD.

DX-890 effectively inhibits NE activity in CF sputum sol (IC50 = 766 pM). However the physical barriers posed by CF whole sputum may prevent DX-890 accessing and inhibiting NE ensconced within the mucus. The mucolytic DNase acts by cleaving extracellular DNA in mucus thereby reducing viscosity. However DNA binds and inactivates NE and therefore DNase treatment of mucus results in an acute increase in active NE. The lipid portion of pulmonary surfactant acts as a lubricant and our group has previously shown that bovine lung extract surfactant (BLES) increases the fluidity of whole sputum. ( Aim

We aimed to show that, by reducing the surface tension of sputum, BLES would enhance the anti-elastase effect of DX-890 in whole sputum.

Methods Whole sputum (0.2g per well in a 12-well plate) was pre-incubated with 20 µl saline as control, 20 µl DNase (100µg/ml) or 20 µl bovine lung extract surfactant (BLES, 100 µg/ml) for 30 minutes at 37 °C followed by addition of 20 µl DX-890 (50 µM or 100 µM) for 1 h at 37 °C.

The contents of each well were aspirated, added to 1 ml NaCl (0.9%) and centrifuged at 10,000 g for 30 min. The resulting sputum sol was assayed for NE activity using the colorimetric substrate N-MeOSuc-AAPV-pNA.

Results NE activity in whole sputum was significantly inhibited by 50 µM DX-890 (p=0.0071) and 100 µM DX-890 (p=0.0009). 50 µM DX-80 inhibited significantly more NE when sputum was pre-treated with BLES (p=0.0173) Treatment of sputum with DNase caused a significant increase in NE activity compared to untreated sputum (p=0.0398) whereas treatment of sputum with BLES did not have this effect.

Conclusion DX-890 effectively inhibits elastase in CF sol. Pre-treatment of CF whole sputum with BLES enhanced the anti-elastase effect of DX-890 and treatment of sputum with BLES alone did not cause an increase in NE activity in contrast to treatment with DNase. DX-890 inhibitory capacity in whole sputum could be improved by co-treatment with surfactant which increases fluidity of sputum and could allow greater access of drugs to sputum components.

Bovine lung extract surfactant, BLES™ was a kind gift from BLES Biochemicals Inc., ON, Canada.

DX-890 was a kind gift from Dyax Corp, MA who also funded part of this research. Table 1 : Results are summarised as neutrophil elastase activity µg/ml and show the mean and SEM of n=27 experiments.

Cystic fibrosis lung disease is associated with an excess of free neutrophil elastase (NE) in the airway arising from persistent neutrophil inflammation. DX-890 is a potent small protein inhibitor of neutrophil elastase that could be a useful therapy for diseases involving neutrophilic inflammation such as cystic fibrosis and COPD.

DX-890 effectively inhibits NE activity in CF sputum sol (IC50 = 766 pM). However the physical barriers posed by CF whole sputum may prevent DX-890 accessing and inhibiting NE ensconced within the mucus. The mucolytic DNase acts by cleaving extracellular DNA in mucus thereby reducing viscosity. However DNA binds and inactivates NE and therefore DNase treatment of mucus results in an acute increase in active NE. Pulmonary surfactant acts as a lubricant and our group has previously shown that bovine lung extract surfactant (BLES™) increases the fluidity of whole sputum. ( 

We aimed to show that, by reducing the surface tension of sputum, BLES™ would enhance the anti-elastase effect of DX-890 in whole sputum.

Methods Whole sputum (0.2 g per well in a 12-well plate) was pre-incubated with 20 µl saline as control, 20 µl DNase (100 µg/ml) or 20 µl bovine lung extract surfactant (BLES™, 100 µg/ml) for 30 minutes at 37 °C followed by addition of 20 µl DX-890 (50 µM or 100 µM) for 1 h at 37 °C.

The contents of each well were aspirated, added to 1 ml NaCl (0.9%) and centrifuged at 10,000 g for 30 min. The resulting sputum sol was assayed for NE activity using the colorimetric substrate N-MeOSuc-AAPV-pNA.

Results NE activity in whole sputum was significantly inhibited by 50 µM DX-890 (p=0.0071) and 100 µM DX-890 (p=0.0009). 50 µM DX-80 inhibited significantly more NE when sputum was pre-treated with BLES™ (p=0.0173) Treatment of sputum with DNase caused a significant increase in NE activity compared to untreated sputum (p=0.0398) whereas treatment of sputum with BLES did not have this effect.

Conclusion DX-890 effectively inhibits elastase in CF sol. Pre-treatment of CF whole sputum with BLES™ enhanced the anti-elastase effect of DX-890 and treatment of sputum with BLES alone did not cause an increase in NE activity in contrast to treatment with DNase. DX-890 inhibitory capacity in whole sputum could be improved by co-treatment with surfactant which increases fluidity of sputum and could allow greater access of drugs to sputum components.

We would like to thank Prof. F. Possmayer, ON, Canada and BLES Biochemicals Inc., ON, Canada for the bovine lung extract surfactant, BLES™. DX-890 was a kind gift from Dyax Corp, MA who also funded part of this research. Table 1 Results are summarised as neutrophil elastase activity µg/ml and show the mean and SEM of n=27 experiments. CF patients infected with P. aeruginosa (PA) have increased levels of proinflammatory cytokines, such as TNF-α, IL-1β, IL-6, and IL-8; but have low levels of the anti-inflammatory cytokine, IL-10 in their airways. The proinflammatory mediators in the CF airway recruit a large number of polymorphonuclear neukocytes (PMNs), which release high levels of neutrophil elastase and human neutrophil peptides and damage the lungs. The intense neutrophil-dominated inflammatory response associated with chronic PA infection results in progressive airway obstruction, the cause of death of over three-quarters of CF patients.

Recent studies highlight the significance of sphingolipid metabolism in cell growth, differentiation, membrane traffic, apoptosis, senescence, and immune regulation. Acid sphingomyelinase (ASM) plays a critical role in sphingolipid metabolism by hydrolyzing sphingomyelin to ceramide. Ceramide self-associates to form ceramide-enriched membrane platforms(also called lipid rafts). These platforms have been shown to be involved in the internalization of PA; induction of apoptosis and immune regulation.

Our in vitro and in vivo studies of ASM confirm that there is a loss of induction of ASM after PA infection in both CF bronchial epithelial cells and CFTR knock out mice as compared with controls. In vitro, we examined the ASM activity in human bronchial epithelial cells with normal CFTR expression (S9 cells) and with mutant CFTR expression (IB3 cells ) after PA infection at 0, 10, 30, 60, 180, and 300 min. ASM activity was up-regulated in S9 cells following PA infection after 60 min and kept increasing until 300 min, while ASM induction was lost in IB3 cells. ASM protein and ceramide levels also increased in S9 cells following PA infection compared with IB3 cells.

In our in vivo studies, CFTR knock out mice and control C57BL/6 mice were both administered various doses of PA orally (2x106 cfu to 5x107 cfu) . Six hours after PA infection, lung homogenates from the C57BL/6 mice indicated increasing ASM protein and ceramide production, which corresponded to the dose of PA given. In contrast, ASM protein and total ceramide production remained at baseline levels in the lung homogenates of CF mice regardless of the dose of PA given.

RNA interference (siRNA) assays were performed by transfecting a 21 bp pre-designed siRNA specific for human ASM in S9 cells. These cells were then infected with PA for 5 hrs to see if they had similar IL-8 level compared with infected IB3 cells. We demonstrated a significantly elevated IL-8 expression in S9 cells tranfected with ASM siRNA compare with S9 cells transfected with a scramble siRNA (p<0.005).

Our results suggest that the loss of ASM induction, caused by mutant CFTR, is involved in PA clearance and contributes to the exaggerated proinflammatory cytokine response in CF airway. We are conducting research on defining the roles of ASM and the sphingolipids in PA infection; cell signaling; as well as cytokine induction, We are also looking at the therapeutic potential of AAV.ASM vectors in PA-infected CF bronchial epithelial cells and CFTR knock out mice.

Work supported by NHLBI Introduction: New therapies are critically needed to treat cystic fibrosis (CF) lung disease, which is characterized by chronic infection and inflammation. Novel therapies to target airway inflammation are likely to improve the clinical outcome in CF. As yet there is no consensus regarding the molecular pathway(s) mediating inflammation in the CF lung.

Objective: To establish whether Toll-like receptor (TLR) signalling is the key molecular pathway responsible for the inflammation seen in CF lung disease and to determine whether inhibitors of TLR signalling can reduce this damaging inflammatory response.

Methods and Results: To quantify the airway inflammatory response, we used established CF and CF-corrected lung epithelial cells (IB3-1 and C38). Since artefacts may alter the inflammatory phenotype of cell lines, we validated all findings with fresh peripheral blood mononuclear cells (PBMCs) from CF patients (n=23) and healthy controls. We measured the inflammatory response (IL6, IL8, TNFα) of these cells following stimulation with: (i) CF pathogens (P. aeurginosa, B. multivorans, B. cenocepacia); (ii) purified TLR ligands. Both CF respiratory epithelial cells and CF PBMCs had a greatly increased inflammatory response compared to matched controls when stimulated with whole bacteria and purified TLR ligands (P<0.05). Interestingly, the CF airway cells responded most vigorously to the TLR5 ligand, flagellin (P<0.001). To investigate TLR5 as a potential anti-inflammatory target, we found that TLR5 mRNA (P<0.001) and protein (P<0.05) were expressed at a higher level in CF compared to CF-corrected lung epithelial cells. Finally, to determine the contribution of TLR5 activation to CF lung inflammation we exposed the airway cells to wild-type (PAK & PAO1) and isogenic flagellin-deficient strains of P. aeruginosa (PAK∆fliC, PAO1∆fliC, PAO1∆flgE & PAO1∆fliM) . Strikingly, the absence of the TLR5-flagellin interaction significantly reduced the proinflammatory response of CF respiratory epithelial cells to P. aeruginosa (P<0.001).

We show that TLR-mediated innate immune responsiveness is increased in both CF respiratory epithelial cells and fresh blood cells from CF patients. Moreover, inhibition of the TLR5-flagellin interaction markedly reduced proinflammatory response of CF respiratory epithelial cells following exposure to predominant CF pathogen P. aeruginosa. These data suggest that TLR5 activation may represent a novel anti-inflammatory target for improving CF lung disease.

Supported by the Canadian CF Foundation.

CF airway cells (IB3-1) were stimulated with wild-type (PAK) and flagellin-deficient P. aeruginosa (PAK∆fliC). Introduction: Genetic polymorphisms for TGF-β 1 have recently been identified as a significant modifier of CF lung disease severity. However, at present little is known about protein measurements of TGF-β 1 in bronchoalveolar lavage fluid (BALF) or serum obtained from pediatric CF patients during an acute respiratory exacerbation.

Hypothesis: TGF-β 1 in CF BALF and serum is increased in association with markers of CF lung disease severity, and is decreased after completion of IV antibiotic (IV abx) therapy.

Study Design and Methods: In a descriptive, prospective study of children with CF presenting for hospital admission to treat a pulmonary exacerbation, BALF specimens were obtained from patients undergoing clinically-indicated flexible bronchoscopy on admission. BALF was immediately analyzed for total and differential cell counts with the cell-free BALF supernatant stored at -80°C. Serum specimens were obtained at the initiation and conclusion of IV abx and similarly stored at -80°C. Measurement of total TGF-β 1 utilized a commercial ELISA kit. T-test statistical analysis assumed significance at p<.05.

Results: BALF was obtained from 8 CF patients with a mean age of 8.5±1.9 years (range 1-16 years). Three BALF cultures were Pseudomonas aeruginosa positive (PsA+). BALF TGF-β 1 levels ranged from 88-309 pg/ml (mean 168±26.4 pg/ml) with the highest TGF-β 1 measured in a patient with ABPA and culture positive for M. abscessus. In BALF, higher neutrophil counts (greater than the median, 2.58x10 6 PMN/ml) were associated with increased BALF TGF-β 1 (214±31.9 pg/ml vs. 121±28.8 pg/ml, p<.05). Serum TGF-β 1 was collected in 8 CF patients with a mean age of 7.5±2.3 years (range 1-18 years), four of whom were PsA+. Serum TGF-β 1 ranged from 21.4-99.7 ng/ml (mean 56.6±9.23 ng/ml). In both BALF and serum, neither patient age nor PsA culture status was associated with increased TGF-β 1 at admission. However, previous hospitalization in the last 12 months was significantly associated with an elevated admission TGF-β 1 in both BALF (233±38.2 pg/ml vs. 122±23.8 pg/ml, p<.05) and serum (75.8±8.19 ng/ml vs. 37.3±9.09 ng/ml, p<.02). After completing the clinically-indicated course of IV abx, 3/4 PsA-CF patients had a reduction in serum TGF-β 1 (mean 69.5±11.4 ng/ml to 48.1±4.52 ng/ml, p=NS). In contrast, 3/4 PsA+ CF patients had an increase in serum TGF-β 1 (mean 43.7±12.5 ng/ml to 62.3±17.1 ng/ml, p=NS) after completing IV abx therapy.

Conclusions: TGF-β 1 is measurable in BALF and serum from pediatric CF patients and varies in association with clinical parameters. These preliminary results support our hypothesis that increased TGF-β 1 may be associated with more severe lung disease as evidenced by 1) association of increased TGF-β 1 in BALF and serum obtained from CF patients with previous hospitalization in the last 12 months, and 2) association of increased TGF-β 1 in BALF with increased cellular inflammation. The direction of change in serum TGF-β 1 after IV abx therapy may be influenced by the presence or absence of Pseudomonas infection.

Acknowledgments: Supported by CFF (HARRIS07AO). Paula Murphy and Cassidy Henegar provided technical support. Background: Cystic fibrosis is one of the leading genetic causes of end stage lung disease for which the treatment is lung transplantation. However, mortality due to chronic rejection known as obliterative bronchiolitis (OB) remains high due to small airway obliteration. OB is associated with increased levels of matrix metalloproteinase-8 (MMP-8), an interstitial collagenase, expressed by polymorphonuclear cells (PMNs). PMNs are among the first inflammatory cells to be detected in OB but very little is known about the role of PMN MMP-8 or the mechanisms used by PMNs to migrate through extracellular collagen matrices. Objective: To identify the molecular mechanism for neutrophil recruitment into the airway. We hypothesize that MMP-8 promotes accumulation of PMNs in OB lesions by promoting their migration through the extracellular matrix protein barriers by its pericellular collagenase activity. Method: The heterotopic airway transplant model was used to study the role of neutrophil MMP-8 in promoting PMN migration through extracellular matrices. MHC-disparate Balb/c tracheas were subcutaneously transplanted into C57Bl/6 wild-type (WT) or MMP-8 deficient (MMP-8-/-) mice. To further dissect the mechanism of PMN migration in vitro their migration through collagen gels was studied. Collagen gels were made in transwells using rat-tail collagen providing a highly cross-linked thick collagen barrier. Bone marrow derived PMNs from WT or MMP8 -/-mice were fluorescently labeled and stimulated to induce the surface expression of MMP-8. PMNs were placed on thick collagen gels in traswells. The transwells had polycarbonate membranes at the bottom of the gels. The pore size of the membranes did not allow PMNs to pass through into the lower chambers that had fLMP a chemo attractant. At 4 hours and 24 hours the gels were fixed and the polycarbonate membrane at the bottom of the collagen gel was gently peeled off. PMNs on the membrane were counted under a fluorescent microscope. Results: Significantly decreased migration of PMNs into the airway lumen was noted in the MMP-8 deficient mice (P<0.05) 2 weeks post-transplantation compared to WT mice in the heterotopic airway transplant model. In addition there was more than a three-fold increase in the PMNs outside the lumen in the extracellular collagen matix (P<0.02). WT-PMNs were able to penetrate collagen barriers in greater numbers both at 4 and 24 hours compared to MMP8 -/-PMNs in vitro (P < 0.002 for 4 hours and 0.001 for 24 hours). In addition, MMP inhibitor GM 6001-treated WT cells demonstrated decreased migration through collagen gels (4 hours P<0.008 and 24 hours P<0.008). Conclusions: MMP-8 -/-mice has significantly less PMNs in the airway lumen due to decreased migration through collagen matrices providing protection against OB. This has great significance for conditions such as cystic fibrosis, lung transplant rejection (obliterative bronchiolitis) and other inflammatory diseases mediated by PMNs. MMP-8 blockade can potentially be a novel therapeutic target to decrease neutrophil efflux into the airways and thus reduce airway inflammation. Background: Our laboratory has demonstrated alterations in CF-related regulation of cholesterol processing. Hydroxycholesterols (oxysterols) are oxygenated derivatives of cholesterol formed from autoxidation of cholesterol and from de novo synthesis. Oxysterols are known to be key regulators of cholesterol metabolism and cellular signaling. The role of oxysterols in the cellular pathways involved in cholesterol processing in cystic fibrosis has not been elucidated. Based on our previous findings of increased de novo cholesterol synthesis in CF tissues, we hypothesize that oxysterol production in CF would be positively impacted. Increased oxysterol production may lead to abnormalities in cellular signaling associated with oxidative stress and may influence the inflammatory response in CF. The goals of this research are to investigate the effect of oxysterols on inflammation in CF and to determine if there is a difference in oxysterol production in CF compared to wild type controls.

Results: The goal of these studies is to determine if CF cells respond uniquely to oxysterols compared to control cells. Initial studies demonstrate that CF-model 9/HTEo-pCEPR cells failed to respond to 25-OH cholesterol (25-OH) with regards to NFκB activation. Control 9/HTEO-pCEP cells, however, exhibited a significant 2-fold increase in NFκB activity upon 25-OH stimulation (2.1-fold +/-0.1 increase from IL-1, TNF alpha alone). However, IL-8 promoter activation in response to 25-OH was not significantly altered in either CF or wt cells (2.5-fold +/-2.3 (CF) increase from IL-1, TNF alpha alone, 1.8-fold +/-2.3 (wt) increase from IL-1, TNF alpha alone). In order to verify IL-8 promoter results, IL-8 protein production was measured directly in response to Pseudomonas aeruginosa. IL-8 protein level in response to 25-OH was increased in CF cells (1.4-fold) and decreased in wt cells (1.2-fold), although not significantly in either group. Oxysterol content and de novo synthesis are currently being examined in vivo in CF and wild type mouse models. Mice lacking CFTR on a C57BL/6 background and wild type controls were obtained from our animal core and the amount of oxysterol present in these mouse models is currently being studied.

Conclusion: These findings suggest that endogenous oxysterol production or other components of oxidative stress are pre-stimulating NFκB activation via the oxysterol-sensitive pathway. Interestingly, although oxysterols robustly activate NFκB in wt cells, oxysterol-dependent pathways are not involved in IL-8 production. These data suggest that IL-8 is not a primary cause for increased oxysterol mediated inflammation. Elucidating the role of oxysterols in CF will be an important contribution to our understanding of the inflammatory pathways in CF and will contribute to the growing body of evidence for aberrant cholesterol regulation in CF. This work is supported by grants from the CFF and the NHLBI. Cystic Fibrosis (CF) is a common genetic disorder caused by a mutation in the CFTR, a chloride transporter of the ABC transporter family. Innate immunity, the first line in host defense against infectious agents, is abnormal in CF patients. CF patients are commonly colonized with bacterial pathogens, which is the major clinical problem of this disease. Despite a marked difference in the clinical outcome amongst CF patients infected with Pseudomonas aeruginosa and Burkholderia cepacia, the mechanisms under-lying these differences remain unclear. Understanding the innate immune response, specifically the role of CFTR, in P. aeruginosa and B. cepacia infections will lead to new insights into mechanisms that limit these infections.

Our objective is to develop a CF model in the simple organism, the worm, C.elegans, and to test the role of the CFTR orthologue, mrp-4, in host defense. Using this model, we have compared the survival of C. elegans between virulent P.aeruginosa (PA14) and B.cepacia (BC). Our experiments have demonstrated that, when compared to growth on their natural food source, E.coli OP50, worms fed on PA14 and BC die prematurely. Specifically, worms survive up to 140 hours on OP50, whereas PA14 fed worms die with in 40 hours. Death on PA14 also requires active infection, as heat inactivation of these bacteria destroys their killing capacity. Two BC strains have been tested, a virulent strain (25416) and an avirulent strain (17616). Worms infected with BC show an intermediate phenotype, surviving approximately 100 hours. Our future work will define whether synergy occurs during coinfection with P.aeruginosa and B.cepacia, as would be suggested by clinical data. In addition, we will define the transcriptional changes induced in C.elegans fed on PA14 and BC, and use these data to identify common and pathogen specific responses to these infections. Finally, mrp-4 knock-down worms will be generated using RNA interference, and the role of the CFTR orthologue in infection will be tested.

In conclusion, C.elegans offers a genetically tractable model organism in which the innate immune response to Pseudomonas aeruginosa and Burkholderia cepacia can be studied. In addition, this simple model system will allow us to define the role of the CFTR orthologue in host defense.

Supported Serine proteases released from neutrophils are considered central to the pathogenesis of CF lung disease, and have been obvious therapeutic targets. Although intracellular serine proteases are important in host defense and bacterial killing, extracellular enzymes may contribute to bacterial persistence and promote airway inflammation. Neutrophil elastase (NE) digests key opsonins present in the lung and has been shown to disrupt phagocytosis, allowing the bacteria to persist in the face of established pulmonary inflammation. In addition we have found that other proteases, like cathepsin G (CG), an abundant serine protease found in human and murine neutrophils, may also have other, critical roles in development of CF lung disease. Indeed, using murine models of endobronchial inflammation, CG inhibits airway defenses and interferes with the host's ability to clear Pseudomonas aeruginosa from the lung with effects quite distinct from NE. To test the hypothesis that differences in bacterial killing are due to defects in innate defenses created by excess, unopposed CG at the apical surface of the airway epithelium, we have examined profiles of proteins secreted into the airway lumen by epithelial cells, which are necessary for P. aeruginosa clearance and susceptible to CG proteolysis. Analysis of lavage fluid utilizing 2-D PAGE indicated a total of 88 spots which were increased in pooled lavage fluid from 6 infected CG-deficient mice as compared to 6 wild-type littermates. Tandem mass spectrometry analysis identified several novel proteins that were cleaved by CG, including inter-alpha inhibitor protein, selenium-binding protein, SEC14-like protein, beta-2-glycoprotein, and Annexin A5. Another protein candidate, serum amyloid P component (SAP), is a novel opsonin and may be involved in airway defenses. Immunoblot analy-sis demonstrates that CG cleaves SAP. These results suggest that CG, like NE, plays an important role in the pathogenesis of lung diseases. We have previously demonstrated in ∆F508-CF mice that in vivo treatment with azithromycin (AZM) attenuates cellular infiltration in baseline and lipopolysaccharide (LPS)-induced inflammation, and inhibits proinflammatory cytokine release in induced inflammatory condition (Legssyer et al, 2006) .

This study aimed at investigating macrophages isolated from wild type (WT) or ∆F508-CF mice (van Doorninck et al, 1995) and the effect of AZM on these cells.

Purified peritoneal and alveolar macrophages were stimulated with LPS (P. aeruginosa, 0.1µg/ml) plus IFNγ (0.1µg/ml), with or without AZM (1µg/ml). Macrophage inflammatory status was investigated by assessing different pro-and anti-inflammatory mediators at mRNA level 6h after stimulation.

Pro and anti-inflammatory markers seemed to be reduced in naive alveolar CF macrophages. By contrast, a trend to overexpression of these markers was seen in naive peritoneal CF cells. Pro-inflammatory status, as assessed by IL-1β, was induced in stimulated alveolar and peritoneal CF macrophages and in naive peritoneal CF cells. Moreover, NOS-2 was found to be overexpressed in stimulated alveolar CF macrophages while IL-10 was downregulated. AZM significantly reduced IL-1β expression in stimulated alveolar macrophages.

In conclusion, peritoneal and alveolar macrophages exhibit distinct phenotype. A pro-inflammatory status was more prominent in stimulated CF alveolar and peritoneal cells. Interestingly, AZM modulates IL-1β overexpression only in CF stimulated alveolar macrophages, supporting the antiinflammatory activity of this macrolide and identifying, at least partly, alveolar macrophages as possible target cells for its effects.

Supported It has been recently reported by Perez and Coll. that the inhibition of function of w/t CFTR produces an inflammatory profile that resembles that observed in CF patients [Am J Physiol Lung Cell Mol Physiol, 2007] , whereas we demonstrated that correction of F508del CFTR function with MPB-07 down modulates the Pseudomonas aeruginosa (P.aeruginosa) dependent expression of the pro-inflammatory mediators IL-8 and ICAM-1 in CF bronchial cells [Am. J. Resp. Cell. Mol Biology, 2007] . Since both evidence support a direct link between CFTR function and inflammatory response in respiratory epithelial cells, we extended our investigation to other F508del CFTR correctors, such as miglustat [Norez et al, FEBS letters, 2006] , which is an approved drug for Gaucher disease, in comparison with an isomer without any correcting effect, namely NB-DGJ. Human bronchial IB3-1 (CFTR F508del/W1282X), CuFi-1 (CFTR F508del/F508del) and NuLi-1 (CFTR w/t) cells were exposed to the laboratory strain of P.aeruginosa (PAO1) or TNF-alpha or IL-1beta and the transcription of ICAM-1 and IL-8 were quantitated by real time RT PCR. CFTR function was assayed by single-cell fluorescence imaging, using the potential-sensitive probe DiSBAC2 [Renier et al, Hum. Gene Ther. 1994 ]. Analysis of binding of NF-kB and AP-1 transcription factors to labelled DNA target was performed with Electrophoretic Mobility Shift Assay (EMSA) [Borgatti et al, J. Biol. Chem. 2003 ]. Miglustat significantly reduced the expression of IL-8 by 90% in IB3-1 and by 70% in CuFi-1 cells and of ICAM-1 by 80% in IB3-1 and by 50% in CuFi-1 cells, upon infection by PAO1. In parallel, correction of F508del CFTR function was observed after miglustat treatment in both IB3-1 and CuFi-1 cells. Miglustat had no major effects on overall binding activity of transcription factors NF-kB and AP-1, activated by PAO1 in IB3-1cells. The inflammatory response induced by TNF-alpha or IL-1beta was also significantly reduced in IB3-1 and CuFi-1 cells treated with miglustat. NB-DGJ, which is not a corrector of function of F508del CFTR, down modulated the induction of IL-8 and ICAM-1 in IB3-1 and CuFi-1 cells. In addition, both miglustat and NB-DGJ reduced the inflammatory response to PAO1 in non CF NuLi-1 cells. In conclusion, miglustat and NB-DGJ have an anti-inflammatory effect in bronchial cells independently of the correction of F508del CFTR, by interfering with the pro-inflammatory signaling downstream the receptors for pathogens and pro-inflammatory cytokines. Since miglustat is already approved for the treatment of Gaucher disease and other glycosphingolipidoses, it may be an interesting candidate to ameliorate lung inflammation in CF patients. Background: The primary cause of mortality and morbidity in Cystic Fibrosis (CF) is lung disease, which is characterized as a chronic cycle of infection and inflammation dominated by a neutrophilic infiltrate and increased levels of circulating mediators such as IL-8. A wide variation in the rate of progression of lung disease is observed in CF.

Hypothesis: We postulate the presence of a brisk systemic inflammatory response in CF patients and that this systemic response reflects the magnitude and progression of lung disease.

Methods: Lung function and clinical data were obtained from 164 CF patients treated and followed in the Adult CF Clinic, St Paul's Hospital. Forced expiratory volume at 1s (FEV1) and forced vital capacity (FVC) values dating back to 2000 were used in regression equations to group patients into quartiles based on their decline in predicted FEV1%. Rate of decline in lung function was compared with clinical characteristics and various systemic inflammatory biomarkers, including WBC, band cell count, CRP, IL-1β, IL-6, IL-8, MCP-1, and GM-CSF Results: The quartiles exemplified significantly different rates of decline in FEV1/FVC% values. Subjects with a rapid decline in lung function were younger and were colonized/infected more commonly with Methicillinresistant Staphylococcus aureus (MRSA) (p<0.05). In addition, altered calcium, magnesium, and albumin levels were found in subjects with increased rates of decline in lung function (p<0.05). Preliminary results show that patients with CF have significantly elevated levels of WBC (10.94E9/L), CRP (8.708E3ng/ml), and IL-8 (48.65 pg/ml) in serum when compared to control subjects (p<0.05). There were trends in relationships between levels of IL-6, IL-8 and MCP-1 and rates of decline in lung function, however none reached statistical significance.

Conclusion: There is a variable rate in the decline in lung function among CF patients and one factor may be a heterogeneous systemic inflammatory response. Circulating pro-inflammatory mediators may not only impact progression of lung disease, but could also be novel biomarkers to monitor and assess disease severity in Cystic Fibrosis. Nutritional status in addition effects the rate of decline in lung function perhaps by modifying the inflammatory response.

This research is supported by the BC Lung Association. Overexpression of the epithelial Na + channel β subunit (protein = βENaC, gene = Scnn1b) in transgenic mice results in CF-like lung pathology, characterized by neonatal mortality, mucus obstruction and airway inflammation (Mall et al., Nat. Med. 2004) . Breeding βENaC mice into gene-deleted mice enables quick and efficient determination of the specific pathways relevant to the development of lung pathology. TNFα is a pleiotropic pro-inflammatory cytokine released by many different cell types, including T cells, macrophages, granulocytes, and epithelial cells. TNFα levels are significantly elevated in bronchoalveolar lavage (BAL) from βENaC mice in comparison to wild-type (WT) littermates. We crossbred βENaC mice (heterozygous, inbred line C57-6608) with TNFα knockout mice, generating four types of mice: WT, TNFα +/-; WT, TNFα -/-; βENaC, TNFα +/-; and βENaC, TNFα -/-. All mice had comparable, high survival, ranging between 80-90%. The lack of TNFα did not prevent neutrophil and eosinophil infiltration and did not significantly modify the lung pathology characteristic of βENaC mice, namely mucus plugs, mucous secretory cell hyperplasia, airway inflammation and emphysema. The absence of TNFα in BAL from βENaC, TNFα -/-mice, in comparison to βENaC, TNFα +/mice, was confirmed by Luminex cytokine assay. The levels of KC, a potent neutrophil chemoattractant, were significantly elevated in BAL fluid from βENaC mice, regardless of the presence or absence of TNFα. In WT mice, no granulocytic infiltrate or morphologic changes were observed and TNFα and KC were undetectable. Similarly, the pathologic changes in βENaC mice were not mitigated by crossbreeding to knockout mice for TNFα receptor 1 (TNFαR1), the major mediator of biologic responses classically attributed to TNFα. These results suggest that TNFα per se does not have a critical pro-inflammatory role in the development of inflammation in the βENaC mouse model. The production of alternative cytokines may compensate for the loss of TNFα bioactivity in the TNFα and TNFαR1 -/-mice.

Since TNFα has been suggested to play a role in the regulation of ENaC, Ussing chamber studies are underway to test whether the bioelectric features of tracheobronchial epithelia from βENaC mice are altered by the absence of TNFα in vivo. These data suggest that a TNFα-independent signaling cascade links airway surface dehydration to airway inflammation in the βENaC mouse model. Innovative pharmacological approaches to control the excessive neutrophil infiltrates into the bronchial lumen of CF patients are thought to be beneficial to reduce the extensive airway tissue damage. The activation of expression of proinflammatory genes by P.aeruginosa with bronchial epithelial cells is a central mechanism to be targeted with novel therapies. Medicinal plants from the socalled traditional Asian medicine are attracting a growing interest because of their potential safety, already tested in large scale applications in human diseases. However, due to the presence of different active principles in each plant extract, whose multifunctional effects may even result contradictory, understanding the effect of each component is mandatory to pursue selective and reproducible applications. A panel of medicinal plant extracts have been firstly screened for their capacity to interfere in the binding of nuclear transcription factor proteins (TF) with DNA consensus sequences identified in the promoters of the pro-inflammatory genes, thus for their potential inhibitory action on gene expression. Extracts from several medicinal plants have been screened for their ability to interfere with the TF NF-kB, AP-1 and CREB, which are induced by P.aeruginosa, and some of them, such as Emblica officinalis (EO), Aegle marmelos (AM), Polyalthia longifolia (PL), have been shown to inhibit TF/DNA interactions, opening the possibility of potential applications to downregulate expression of pro-inflammatory genes. Extracts from EO, AM, PL were tested in IB3-1 CF bronchial cells exposed to the P.aeruginosa laboratory strain PAO1. EO, AM and PL strongly inhibited the PAO1-dependent transcription of IL-8 in IB3-1 cells. Pyrogallol, one active principle of EO, was tested in IB3-1 cells, where it inhibited the transcription of the neutrophil chemokines IL-8, GRO-alpha and gamma, of the Intercellular Adhesion Molecule ICAM-1 and the pro-inflammatory cytokine Interleukin 6, similarly to the whole EO extract, whereas a second active principle from EO, namely 5hydroxy-isoquinoline, had no effect. Similar results were obtained with EO and pyrogallol in the monocyte-derived macrophage cell line THP-1 exposed to PAO1. In conclusion, extracts from plants of the traditional medicine can inhibit expression of pro-inflammatory genes and screening active principles purified from medicinal plants could result useful to identify safe and innovative pharmaceutical molecules to control lung inflammation in the lung of CF patients.

Supported by Italian Cystic Fibrosis Research Foundation and by Fondazione CariVerona -Bando 2005 -Malattie rare e della povertà. Massive infiltrates of neutrophils in the mucosal wall and lumen of the conductive airways of CF patients contribute to the progressive lung func-tion decline by releasing different proteases responsible for the progressive airway tissue damage. Bacterial products and pathogens themselves within the mucopurulent material of the airway surface fluid induce the activation of transcription factors such as NF-kappaB, AP-1, Sp1, NF-IL6, NF-AT, Elk-1, CREB resulting in expression of chemo/cytokine genes driving the recruitment of leukocytes inside the bronchial lumen. To find new treatment options focused on the reduction of neutrophil chemotaxis, without abolishing the immune response against pathogens, we are exploring the transcription factor (TF) "decoy" strategy, in which oligodeoxynucleotides (ODN) mimicking the consensus sequences for the TFs proteins identified in the promoters of different chemo/cytokines are delivered inside the cell in order to interfere with gene transcription. CF bronchial epithelial cells IB3-1 have been exposed to the (i)P.aeruginosa(/i) strain PAO1. A clear PAO1-dependent activation of TFs such as NF-κB, Sp1, AP-1, NF-AT, NF-IL6 was confirmed by Electrophoretic Mobility Shift Assay (EMSA). In parallel, transcription of genes involved in innate immune response has been quantified by real-time RT PCR. Transcription factor "decoy" ODNs directed against the consensus sequences identified in the promoters of different genes have been designed and validated by testing their interference in TF protein/DNA binding assays. Transfection of IB3-1 cells with HIV-1 LTR and IgK chain NF-κB ODN "decoys" complexed with Lipofectamine, performed 30 hrs before challenge with PAO1, was shown previously to inhibit strongly PAO1-dependent transcription of IL-8 but not GRO, IL-1β, IL-6 and ICAM-1. Therefore other TF "decoy" ODNs have been also tested: a) ODN for NF-κB from IL-8 promoter inhibited IL-8, GRO-gamma and IL-6 by 50%, b) ODN for Sp1 from HIV-1 genome inhibited IL-6 by 50%, c) ODN for AP-1 from IL-8 promoter inhibited both IL-8 and GRO-gamma by 50%. A TF "decoy" molecule designed as peptide-DNA chimera mimicking the consensus sequence of HIV LTR NF-κB strongly and selectively inhibited IL-8 transcription. In conclusion, transcription of chemo/cytokines induced by (i)P.aeruginosa(/i) in CF bronchial epithelial cells (i)in vitro(/i) can be inhibited with different efficiency and selectivity by TF "decoy" molecules. These results provide useful hints for a gene-targeted anti-inflammatory approach and add further information on the regulation of expression of pro-inflammatory genes induced by (i)P.aeruginosa(/i) in bronchial epithelial cells.

Supported by Italian Cystic Fibrosis Research Foundation and by Fondazione CariVerona -Bando 2005 -Malattie rare e della povertà. Background: CF is characterized by hypersecretion of the pro-inflammatory cytokine IL-8 from airway epithelial cells. However the mechanism by which IL-8 gene expression is dysregulated in CF is not known. The expression of cytokine and chemokine genes is known to be regulated at multiple mechanistic steps including transcription, mRNA decay, translation, and various post-translational steps. Sequence-specific mRNA degradation is now recognized to be an important site controlling the expression of several chemokine mRNAs. This selective behavior is conferred by cisacting elements in the mRNA composed of AU-rich sequences (AREs). The regulatory function of AREs is thought to be mediated via RNA-binding proteins that specifically recognize the ARE motifs. The steps required for mRNA decay is comprised of deadenylation, decapping and body decay. Deadenylation or removal of the poly (A) tail by poly(A)-specific ribonuclease (PARN) appears to be the first and perhaps the rate-limiting step. This is accompanied by decapping or enzymatic removal of the 5' methylate guanosine cap. Subsequently, exonuclease activities in the 5' to 3' or 3' to 5' direction predominate to degrade the remaining mRNA.

Hypothesis: We have hypothesized that defects in transcription as well as modulation of the post-transcriptional stability of IL-8 mRNA might contribute to hyper-production of IL-8 protein in the cystic fibrosis.

Methods: IL-8 mRNA stability was assessed in cystic fibrosis IB3-1 lung epithelial cells and in AAV-CFTR-repaired IB3-1/S9 cells, by measuring residual IL-8 mRNA levels at various intervals of time after addition of actinomycin-D to the culture. Protein S100 extracts, prepared from the CF cell line as well as the CFTR-corrected cell line, were analyzed by western blot. The expression levels of the various factors known to participate in ARE-mediated mRNA decay, including TTP (an ARE-binding protein), PARN and the Exosome were compared in the two sets of S100 extracts.

Results: We find that the levels of IL-8 mRNA in the CF cells are greater than the levels found in the CFTR-repaired cells. In addition, we found that the rate of decay of IL-8 mRNA in CF cells was significantly less than that in the repaired cell line. The levels of TTP, PARN and Exosome proteins were significantly reduced in the CF cells compared to the CFTRrepaired controls. However, expression of TTP in IB3-1 cells causes significant destabilization of IL-8 mRNA.

Conclusion: We conclude that the high levels of IL-8 protein expression in CF lung epithelial cells can be partly due to enhanced stability of the IL-8 mRNA. Consistently, the actual levels of IL-8 mRNA are higher in the CF cells compared to controls. Understanding the mechanism by which TTP promotes enhanced destabilization of IL-8 mRNA in CF cells may be important for developing novel therapeutic targets to alleviate the pulmonary pathophysiology of this disease manifested by hyper-secretion of IL-8 protein. Understanding the mechanism of airway remodeling could lead to the identification of novel therapeutic targets for the prevention of irreversible lung damage in CF. We have investigated transcriptional responses to epithelial injury in a mouse model of CF that was developed in our laboratory (F508del, Cftr tm1EUR FVB backcross F12). To induce transient epithelial lung injury, homozygous normal and mutant age matched littermates were treated in parallel with Naphthalene (200 mg/kg IP) or carrier (corn oil) as control (N= 4x12). Naphthalene causes an almost complete selective ablation of CLARA cells overnight, followed by migration of resistant ciliated cells and rapid proliferation of undifferentiated progenitor cells. Two and seven days after treatment, lungs were collected for histological and transcriptome analysis. Based on a previous Affymetrix microarray analysis (N=6), a quantitative PCR array was designed containing 56 genes differentially expressed after naphthalene injury. As expected, naphthalene injury results in a transient increase of cell proliferation markers (Mki67, Cdc20, Cdc2a etc), and a strong (90%) reduction of CLARA cell markers (CC10/Scgb1a1; Claudin10/Cldn10; Cyp2f2). In addition, we have now identified twenty-five genes that are increased (P< 0.001) two-to fifty-fold after naphthalene injury in both normal and mutant mice. This includes known and novel markers of epithelial tissue injury such as Timp1, Retlna, Mmp8, Serpin3n, and Lipocalin. In particular, a group of EGF receptor agonists is strongly induced: Amphiregulin (Areg, 13 fold), Epiregulin (Ereg 50 fold) and Heparin binding EGF (3 fold). Amphiregulin mRNA can be detected by in situ hybridization only in airway epithelial cells after injury. These EGFR agonists are involved in stimulating epithelial repair, but can also activate cells in the underlying mesenchyma. Indeed, we observe a substantial increase of major extracellular matrix mRNA's two days after injury (Co1a1: 5-fold, Col3a1: 10-fold, Elastin: 3-fold). Whereas expression levels were substantially reduced in normal animals seven days after injury, in CF mutant animals two to three fold higher levels were observed for these three genes at day seven (P<0.04). CONCLUSIONS. We have designed and validated a gene array that can be used to study distal lung injury and repair. EGFR agonists produced by epithelial cells are the most prominent growth factors after injury, involved in both epithelial repair and extracellular matrix production by mesenchymal cells. ECM production in CF mouse lung is sustained compared to normal mice, suggesting an inherent tendency towards fibrosis in CF lungs. Our data suggest that the mesenchymal EGF receptor and regulation of its agonists are important future targets of novel therapeutic strategies.

Supported by EEC 6th FW projects EUROCARECF, IMPROVED PRECISION The mucosa of the proximal airways defends itself and the lower airways from inhaled irritants, allergens, and microbial and viral infections by several mechanisms. Sensory nerves monitor the luminal microenvironment and trigger reflexes in the central nervous system (CNS) that alter breathing, induce cough and stimulate mucus secretion when challenged with noxious stimuli. Sensory nerves also release the tachykinins substance P (SP), neurokinin A and calcitonin gene-related peptide through axon reflexes in neighboring tissues, and these locally released tachykinins stimulate mucus secretion by binding to neurokinin 1 receptors on submucosal glands. Recently we reported that local fluid secretory responses to noxious stimuli are dependent on the Clchannel CFTR in mouse airway submucosal glands and are defective in glands from CFTR knockout mice (Ianowski et al., J Physiol, 2007, 580, 301) . We have now tested the effects of SP directly and examined the possible role of CFTR in mediating these responses using tracheas from congenic wild-type and CFTR knockout mice (Cftr m1UNC /Cftr m1UNC ) that had been bred onto C57BL/6 J or Balb/c genetic backgrounds. We compared single gland secretion rates using optical methods as described previously. The cftr genotype of each mouse was assessed by using PCR to amplify genomic DNA from tail clippings obtained at age 18 days. After the CFTR knockout mice were weaned, intestinal obstruction was minimized by supplementing their water with Peglite. Capsaicinoids (chili pepper oil) increased fluid secretion from glands of wild-type mice from 10 ± 2 pl/min (n=9 tracheas, 17 glands) to 50 ± 10 pl/min (n=4 tracheas, 9 glands, Anova p<0.05, Tukey-Kramer multiple comparison test p<0.05). This response was abolished by exposing the basolateral surface of the tracheas to L-732,138 (10 µmol/l), a known SP (NK-1) receptor antagonist (15 ± 5 pl/min, not different from control, Tukey-Kramer multiple comparison test p>0.05). Secretion was stimulated from 4 ± 1 pl/min to 100 ± 30 pl/min (n=12 tracheas, 55 glands, Student's t-test p<0.05) by the direct application of SP, and this response was strongly inhibited by pre-incubation with the CFTR inhibitor CFTRinh172 (saturating concentration, nominal 100 µmol/l; n=6 tracheas, 21 glands). Finally, submucosal glands from CFTR knockout mice failed to secrete when exposed to SP (1 µmol/l) whereas wild-type littermates were responsive. These results indicate that SP mediates local responses to capsaicinoids through a CFTR-dependent mechanism. Loss of this local regulation in CF may contribute to the susceptibility of CF airways. Support: Canadian CF Foundation, Canadian Institutes of Health Research, NIH (DK51817), and the CFF (USA). Chronically infected/inflamed CF human bronchial epithelia (HBE), or normal HBE exposed to supernatant from mucopurulent material (SMM) from human CF airways, exhibit expansion of the endoplasmic reticulum (ER) Ca 2+ stores and amplification of Ca 2+ -mediated inflammatory responses. We have shown that infection/inflammation of HBE triggers an unfolded protein response (UPR) coupled to mRNA splicing of X-box binding protein-1 (XBP-1). Spliced XBP-1 (XBP-1s) is a transcription factor that promotes ER expansion to augment the protein folding capacity during increased protein synthesis. Because we have shown that HBE inflammation couples to increases in protein synthesis, we hypothesized that HBE infection/inflammation-induced ER Ca 2+ store expansion is mediated by XBP-1s. To test this hypothesis, we constructed four retrovirally-tranduced stable 16HBE14o-cell lines containing empty vector (EV), XBP-1 unspliced (XBP-1u), XBP-1s, or dominant negative XBP-1 (DN-XBP-1). Cells were grown to confluence under an air-liquid interface and ER Ca 2+ store expansion and IL-8 secretion studied in the absence or following 24 hr mucosal PBS or SMM exposure. Consistent with our hypothesis, expression of XBP-1s in the absence of stimulation induced ER expansion and increased mucosal UTP-sensitive ER Ca 2+ stores ( To confirm that these effects were mediated by UPR activation, cells were transfected with a UPR response element luciferase reporter plasmid, and luciferase activity was measured in the absence or presence of the UPR inducer tunicamycin (TM). Cells expressing XBP-1s exhibited higher baseline luciferase activity than cells expressing EV or XBP-1u, whereas luciferase activity was decreased in the DN-XBP-1 expressing cells. Furthermore, while TM increased luciferase activity in cultures expressing EV or XBP-1u, its effect was blocked in DN-XBP-1 expressing cells. These findings suggest that XBP-1s is the major trigger of ER Ca 2+ store expansion, which mediates the amplified Ca 2+ -dependent inflammatory response in infected/inflamed airway epithelia. Although it remains to be established whether UPR-dependent XBP-1s is a beneficial or a maladaptive response in infected/inflamed airways, therapies aimed at manipulating this UPR pathway may be beneficial for patients with chronic inflammatory lung diseases. Luminal exposure of well-differentiated normal human bronchial epithelia (HBE) to supernatant from mucopurulent material (SMM) from human CF airways increases the secretion of inflammatory mediators and triggers an unfolded protein response (UPR) mediated by the mRNA splicing of the X-box binding protein-1 (XBP-1). Spliced XBP-1 (XBP-1s) is a transcription factor that expands the endoplasmic reticulum (ER) and protein secretory pathway during ER stress induced by accumulation of unfolded proteins due to increases in protein synthesis. Our previous studies in primary cultures of HBE and 16HBE14o-cells stably expressing empty vector, inactive unspliced XBP-1 (XBP-1u), XBP-1s or a dominant negative XBP-1 (DN-XBP-1) suggest that induction of XBP-1s by SMM exposure promotes ER Ca 2+ store expansion, which mediates a Ca 2+ -dependent hyperinflammatory response. The present studies were designed to determine if Pseudomonas aeruginosa PAK strain (P.a.) would reproduce the effects of SMM in both in vitro and, importantly, in vivo models. For in vitro studies, HBE were exposed to broth (TSB) or to a 20% P.a. extract for 24 hr, and inflammation, XBP-1s, and ER Ca 2+ store expansion investigated by IL-8 secretion, Southern blots and calreticulin expression, respectively. Similar to SMM, P.a. induced a 4.5 fold increase in IL-8 secretion coupled to a 2 fold increase in XBP-1s and larger ER Ca 2+ stores as compared with TSBexposed HBE. These data demonstrate the link between P.a. infection, UPR activation and XBP-1s in vitro. We next tested the relevance of these findings in vivo. Wild-type mice airways were challenged with PBS or 10 6 cfu of P.a., and airway epithelial ER density was assessed by calreticulin staining 24 hr later. In comparison with PBS challenges, P.a.-challenged mice exhibited airway epithelial ER expansion, which correlated with the degree of airway inflammation, based on the presence of inflammatory cells. To test whether this P.a.-induced ER expansion was linked to XBP-1s, we utilized "ER stress activated indicator" (ERAI) mice expressing a fusion protein consisting of XBP-1u and the fluorescent protein Venus. In ERAI mice, UPR activation-dependent XBP-1 splicing leads to Venus expression; hence, Venus fluorescence is an index of XBP-1s. Consistent with our hypothesis that P.a. infection-triggered inflammation would induce UPR-dependent XBP-1s, Venus fluorescence, after 24 hr challenge with 10 6 cfu of P.a., was increased in inflamed as compared with non-inflamed airways. These findings suggest that 1) airway epithelia respond to bacterial infection-induced inflammation by up-regulating the ER Ca 2+ stores and 2) activation of the XBP-1s pathway by bacterial infection may be relevant to airway inflammatory responses in vivo. Funded by the CFF.

We have shown that luminal exposure of well-differentiated primary cultures of normal human bronchial epithelia (HBE) to supernatant from mucopurulent material (SMM) from human CF airways increases total cellular protein synthesis, which reflects the increased secretion of inflammatory factors induced by the infectious and inflammatory process. In the present studies, we first investigated whether these HBE responses to SMM were linked with an increased metabolic rate by measuring lactate accumulation into the serosal media. SMM increased lactate production, and this effect was maximal within 24 hrs (9.0+1.0 vs. 13.2+0.8, and 7.1+0.9 vs. 13.1+0. 4 mmol/L in 24 and 48 hr PBS vs. SMM, respectively; n=5-10), suggesting that the increase in protein synthesis couples to a hyper-metabolic state in infected/inflamed HBE. In agreement with this notion, 6 or 24 hr SMM, as compared with PBS exposure, induced the expression of genes associated with amino acid transport and metabolism (n=4). In addition, 6 or 24 hr SMM exposure up-regulated genes involved in oxidative stress (n=4). We hypothesized that these HBE responses were linked to an unfolded protein response (UPR) mediated by activation of the PKR-like ER kinase/pancreatic eIF2α kinase (PERK)-induced activating transcription factor 4 (ATF4), since this pathway has been shown to confer protection against amino acid loss and oxidative stress in other cells. We first tested whether SMM induced activation of PERK/ATF4 in HBE by performing Western blot analyzes of the components of this pathway. Twenty-four hr SMM exposure induced PERK activation, as indexed by phosphorylation of PERK, in comparison with PBS-exposed HBE. On the other hand, total PERK protein levels were unchanged in SMM-treated HBE. Phosphorylation of eIF2α, the downstream effector of PERK, and increased ATF4 protein levels, which depend on the phosphorylated status of eIF2α, provided additional evidence that the PERK/ATF4 pathway was activated by SMM. These data are consistent with the hypothesis that induction of ATF4 is triggered by UPR activation resulting from increased synthesis of inflammatory factors. We next utilized RNA microarrays to test whether ATF4 target genes were induced by SMM. Six or 24 hr SMM exposure induced ATF4 target genes (e.g., ERO1, an oxido-reductase that provides protection against the accumulation of endogenous peroxides during ER stress; stanniocalcin 2, whose expression is associated with anti-apoptotic functions; and heme oxygenase 1). These findings suggest that 1) activation of the UPR-dependent ATF4 pathway is a compensatory component of the airway epithelial adaptive response to luminal infection/inflammation, and 2) activation of the ATF4 pathway protects against inflammation-induced amino acid loss and oxidative stress by up-regulating genes involved in amino acid transport/metabolism and oxidative stress responses. Unraveling the functions of ATF4 should help determine if therapies targeted to manipulate pathway activity would be likely to improve lung function in patients with CF or other chronic inflammatory airway diseases. Funded by the CFF.

Miller, T.J.; Perez, A.; Qian, Y.; Davis, P. Pediatrics, Case Western Reserve University, Cleveland, OH, USA FXYD5 is a cell surface protein originally identified in a screen for molecular markers of tumorigenesis. Increased FXYD5 expression was found in tumors from stomach, thyroid, colon, pancreatic, breast and lung cancers and correlated with down-regulation of E-cadherin and poor patient prognosis. Recent studies have shown that overexpression of FXYD5 promotes cell motility, decreases cell-cell attachment and increases tumor metastasis. FXYD5, also known as Dysadherin, is a member of small family of proteins known to regulate the Na,K-ATPase. We now report that FXYD5 is upregulated in cystic fibrosis (CF) airway epithelia and modulates wound healing. We show by immunohistochemistry and immunoblot analyses that FXYD5 is increased in the lungs of S489X CF mice, and demonstrate an almost 3-fold increase in FXYD5 expression in the nasal epithelia of CF mice compared to wild-type littermates (P<0.001). Furthermore, we show that FXYD5 is upregulated in nasal scrapings from human CF patients compared to controls (P<0.02). Immunofluorescence data show that Flag-tagged FXYD5 co-localizes with the Na,K-ATPase in epithelial cells, suggesting that FXYD5, similar to other members of the FXYD family, may regulate Na,K-ATPase function. It has previously been shown that expression and localization of the Na,K-ATPase is required for efficient polarization and suppression of cell motility in epithelial cells. The recurrent remodeling of pulmonary epithelium as a result of bacterial infection in CF requires that airway epithelial cells polarize and migrate to wound sites in order to maintain lung integrity. Thus we hypothesized that FXYD5 may be involved in wound healing after infection. Laser-capture microdissection and microarray analysis of murine lung epithelia after 3 hours treatment with P. aeruginosa indicated a significant, 5-fold increase in expression of FXYD5 that was confirmed by immunoblot analysis. Others have shown that FXYD5 may mediate expression of MCP-1, a critical determinant in monocyte recruitment, through activation of the NF-kB pathway. Treatment of human tracheal epithelial (HTE) cells with a CFTR inhibitor (172) confirmed that loss of CFTR function correlated with increased FXYD5 expression by quantitative RT-PCR (P<0.001), an effect that was abrogated with treatment of PDTC, an inhibitor of NF-kB (P<0.01). We speculated that FXYD5-induced increases in cell motility may be due in part to phosphorylation at serine 163. In a murine airway epithelial cell wound healing model, serine to alanine (S163A) mutations at serine 163 inhibited wound healing compared to wild type FXYD5 overexpression, whereas aspartic acid (S163D) mutations increased wound healing (P<0.005). Immunoblot and immunofluorescence analyses of these mutants suggest phosphorylation at Ser163 regulates membrane localization. We conclude that FXYD5 is increased in cystic fibrosis epithelia due to increased inflammatory mediators and suggest that FXYD5 may modulate airway epithelia wound healing after infection with P. aeruginosa through phosphorylation at Ser163. The inflammatory response to bacterial infection in the CF airway is exaggerated compared to normal, leading to the accumulation of millions of necrotic neutrophils. Extracellular neutrophil elastase (NE) activity in the CF airway not only compromises innate defences by cleaving opsonins and reducing ciliary activity, but amplifies the inflammatory response by stimulating expression of the neutrophil chemoattractant IL-8 and increases mucus production, in addition to degrading the tissue matrix leading to fatal bronchiectasis. NE therefore represents an important target for the development of new therapies. However, this strategy requires consideration of the normal physiological function of this enzyme, since previous studies in knock-out mice indicated an essential role for NE in cell migration, bacterial phagocytosis and killing.

Our approach was to use the intracellular neutrophil elastase inhibitor GW311616A to 'knock-out' NE activity in neutrophils in normal human blood and test the function of isolated cells in chemotaxis, bacterial phagocytosis and killing assays.

Whole normal human blood was incubated with GW311616A, or PBS control, for 1h at 37°C. Neutrophils were isolated by sedimentation of red blood cells (RBC) on Dextran 70, purification on Lymphoprep and hypotonic lysis to remove contaminating RBC. Chemotaxis towards IL-8 was measured using a modified micro-Boyden chamber. Phagocytosis was assayed by the depletion of Staphylococcus aureus (SA), Pseudomonas aeruginosa (PA) and E. Coli (EC) in supernatants following culture in a 10:1 ratio with PMN for 2h at 37°C. Supernatants were diluted and remaining organisms were plated on agar and grown overnight to count viable colonies. Bacterial killing was assayed by incubating bacteria and cells in a 1:1 ratio for 15 minutes to allow phagocytosis, washing off remaining organisms, and incubating for 4h at 37°C. Neutrophils were lysed with water and lysates plated on agar to test for bacterial growth overnight.

Results; GW311616A inhibited intracellular NE dose-dependently and at 10 µM 0.168±0.062 % NE activity remained (n=5). There was no significant effect of 10 µM GW311616A on neutrophil chemotaxis, bacterial phagocytosis or killing of any organism compared to PBS-treated controls. Phagocytosis data is shown in the table.

Thus, in the absence of NE activity, human neutrophils remain wellequipped with other defence molecules including myeloperoxidase and defensins to successfully maintain the role of the neutrophil in innate immunity. However, mouse neutrophils which lack defensins require NE activity for optimal intracellular bacterial killing, and mice are not a perfect model for studies of human infection. The development of novel inhibitors of NE to treat lung disease in CF therefore remains an important goal.

Supported by the CF Trust of Great Britain. Background: Sphingolipid signalling may differ between individuals with CF and healthy controls. The response to bacterial inflammation is different, and uptake and inactivation of sphingosine-1-phosphate, an intracel-lular pro-inflammatory mediator, is reduced in CF cells. It may therefore continue to act on G-protein coupled receptors in the plasma membrane. (Boujaoude et al, J Biol Chem 2001) . Furthermore, ceramide originating from basolateral sphingomyelin hinders augmentation of CFTR-mediated anion conductance across the apical membrane, resulting in reduction of transepithelial airway anion secretion (Ito et al, BBRC 2004) .

Aim of study: To determine if there is a difference in the levels of alkaline, neutral or acid sphingomyelinase (SMase), or in the levels of neutral or acid ceramidase, in the intestinal or bronchial mucosa and some other tissues, between wildtype, homozygous (+/+) and heterozygous (+/-) delta-F508 CFTR mice.

Methods: Enzyme activities (Duan and Nilsson Meth Enzymol 2000) were determined in intestine (and content) divided into four regions, liver, lungs, kidney and spleen from deltaF508-CFTR mice (+/+) and controls (wildtype, +/-).

Results: There was an increased amount of neutral ceramidase in spleens from deltaF508-CFTR mice (+/+) in comparison to control mice (p=0.0278). No other significant differences were seen.

Conclusion: Delta-F508 mutation did not influence the levels of alkaline SMase and neutral ceramidase acting as ectoenzymes, or the levels of intracellular SMases and ceramidases, which may all generate bioactive sphingolipid metabolites in intestine and lungs. The implications of the increased level of neutral ceramidase in spleen are not known. 

In children with cystic fibrosis (CF) there is a clear correlation between the development of chronic P aeruginosa infection and acceleration in the decline of lung function. When chronically present, P aeruginosa takes on a mucoid phenotype and is impossible to eradicate. Prior to this, when colonisation is intermittent, it is possible to eradicate it with aggressive antibiotic regimes. We sought to examine the degree of inflammation and innate defence status in the lungs of children with cystic fibrosis in various stages of colonisation by looking at a range of proteases, innate defence proteins and markers of inflammation in broncho alveolar lavage (BAL).

Children with CF were allocated to one of three groups in relation to P aeruginosa infection; chronically colonised, intermittently colonised and non-colonised, on the basis of the Leeds criteria. BAL was collected as per ERS guidelines as part of each patient's routine clinical care. BAL was collected from control patients undergoing elective non-pulmonary surgical procedures. Differential cell counts in BAL were performed manually. Secretory leukocyte Protease inhibitor (SLPI), Elafin, Alpha-1 antitrypsin (A1AT) and Lactoferrin concentrations were measured by ELISA. Neutrophil elastase activity and Cathepsin Activity were assayed by colorometric activity assays.

Fifty two patients were included in the study ranging in age from 3 months to 18 years (11 Chronic, 10 Intermittent, 17 Non colonised and 14 Controls). Neutrophil counts, neutrophil elastase activity and Cathepsin activity were markedly increased in children chronically colonised with P aeruginosa compared to those in the intermittent and non colonised groups. In contrast, levels of the antiproteases SLPI and A1AT and the antimicrobial peptides Elafin and lactoferrin were highest in the control group and decreased as colonisation progressed, with levels in the chronically colonised group markedly lower than those with intermittent colonisation.

This study demonstrates that in children with chronic P aeruginosa colonisation, there is a marked decrease in antiproteases and antimicrobial factors and a marked increase in protease activity and neutrophil influx in comparison with those who are non-colonised or intermittently colonised. These findings underline the importance of careful microbiological surveillance and early aggressive treatment of P aeruginosa infection in children with CF in order to avoid chronic colonisation. Background : Abnormal bronchial angiogenesis is responsible for hemoptysis in cystic fibrosis (CF). Expression of VEGF-A in airway epithelium induces bronchial angiogenesis in animal models. We have recently found that VEGF-A and EGF receptors (EGFR) are increased in the airway epithelium of subjects with advanced CF lung disease.

Aims: To examine the effects of PA bacterial products and EGFR inhibition on VEGF synthesis in airway epithelium. Methods: Culture of non CF (NCI-H292) and CF (CFTE29o-) human airway epithelial cell lines. Stimulation with PA lipopolysaccharide (LPS). Assessment of VEGF mRNA and protein by RT-PCR and ELISA. Use of chemical inhibitors, blocking antibodies and SiRNA.

Results: PA LPS increased VEGF gene expression and protein production time-and dose-dependently in both cells lines. Using chemical inhibitors, we show that EGFR and ERK1/2 activation are required for LPSinduced VEGF production. Using blocking antibodies to EGFR and its ligands, we show that TGF-alpha-dependent EGFR activation mediates PA LPS-induced VEGF gene and protein synthesis. Using pharmacological inhibitors (an ROS scavenger and an NADPH oxidase inhibitor) and using small interfering RNA of Dual oxidase (Duox) 1 and TNF-alpha converting enzyme (TACE),we show that LPS-induced VEGF upregulation is dependent on Duox1-mediated ROS release and TACE activation. Thus, PA products induce VEGF synthesis in airway epithelium via a Duox1-ROS-TACE-TGF-alpha-EGFR-ERK1/2 cascade.

Conclusions: These results describe a novel pathway by which bacterial products induce angiogenic signaling in CF and non CF airway epithelium. Background: Unlike bronchoalveolar lavage (BAL), the airway mucosa has been under-investigated in cystic fibrosis (CF), despite the fact that irreversible airway wall changes (bronchiectasis) are a feature of endstage disease. CF is characterized by a neutrophil-dominated inflammation in BAL, but little is known about the pattern of inflammation in the airway mucosa, especially in children with relatively early stage disease. We aimed to assess whether the pattern of inflammation seen in CF BAL was also found in the airway mucosa in CF children.

Methods: To date, endobronchial biopsies and BAL from 46 children (0-16 years) with CF and 8 control children (0-16 years) without lower respiratory disease have been assessed. BAL cell differential was assessed on May-Grünwald-stained cytospins. Endobronchial biopsies were stained for neutrophils (neutrophil elastase, NE), T-(CD3) and B-(CD20) lymphocytes, eosinophils (EG2), and macrophages (CD68). Area profile counts of immunopositive cells in subepithelial tissue were performed by investigators blinded to disease group.

Results: All cell types were increased in CF BAL compared to controls. CF BAL was characterized by an abundance of neutrophils (837 x 10 3 /ml vs. 3 x 10 3 /ml in controls, p<0.0001) with moderate numbers of lymphocytes (48 x 10 3 /ml vs. 3 x 10 3 /ml in controls, p<0.001). In contrast, CF subepithelial tissue was characterized by a lymphocytic infiltrate (961 cells/mm 2 vs. 575 cells/mm 2 , p<0.01) with only very few neutrophils (12 cells/mm 2 vs. 0 cell/mm 2 , p<0.05). The lymphocytic infiltrate in CF consisted mainly of T lymphocytes (91%). Eosinophil counts in subepithelial tissue did not differ between CF and controls. For all cell types, there was no correlation between counts in BAL and counts in subepithelial tissue. Conclusions: In contrast to the neutrophil-dominated inflammation in the airway lumen, CF is characterized by a lymphocytic inflammation in the airway mucosa. The lymphocytic infiltrate consists mainly of T lymphocytes, the pathophysiological function of which may be important and is being investigated in future work. Support: ERS long-term fellowship and Swiss National Foundation grant to NR The epithelium serves as a barrier to the penetration of foreign antigens, particles, and infectious agents across the airway. The integrity of this barrier is dependent, in part, upon the apical junctional complex consisting of the tight junction (TJ) and the adherens junction. Alterations in TJ permeability have been linked to the pathogenesis of inflammatory bowel disease and this increased intestinal permeability may actually precede the onset of chronic inflammation. In cystic fibrosis (CF), the airway lumen is filled with high concentrations of inflammatory cells, bacteria, and inflammatory mediators. Since TJ barrier function can be significantly reduced by inflammatory mediators, we hypothesized that measures that enhance airway TJ barrier function will decrease airway responses to the continuous presence of inflammatory mediators in the lumen. To test this hypothesis, we examined the relationship between lung inflammation and epithelial permeability in vivo using a lipopolysaccharide (LPS) model of lung inflammation. Pseudomonas aeruginosa LPS was instilled intratracheally into the lungs of C57Bl/6 mice which were then euthanized at 24, 48 and 72 hrs. Lung inflammation was assessed by total cell counts using a hemacytometer and differential counts by Wrights staining of cytospin preparations of the bronchoalveolar lavage fluid (BALF). LPS increased total cell counts and neutrophil concentrations that peaked at 48 hours after LPS administration, compared to saline controls. Measurements of the proinflammatory murine cytokine KC in BALF, by a cytokine antibody bead technique, showed an increase in murine KC at 48 hr following LPS administration, which correlated with the substantial increase in neutrophil concentration. Changes in lung permeability with inflammation were assessed by ELISA measurements of the levels of serum protein murine albumin in BALF. Correlating with changes in cellular inflammation and murine KC levels, albumin concentration peaked at 48 hr after LPS administration. This increase was subsequently resolved, consistent with the restoration of barrier function. An examination of frozen sections of lung from LPS-treated animals showed a redistribution of the tight junction protein ZO-1 consistent with the disruption of barrier function. Since the p38 MAP kinase signaling pathway has been implicated in LPS-induced airway inflammation, and an inhibitor of this kinase, SB203580 has been shown to reduce this inflammation, the effect of this inhibitor on barrier function is being investigated. In initial studies, SB203580 appears to reduce total and neutrophil cell counts by 50% in vivo. The effect of SB203580 on murine albumin concentrations in BALF and on TJ protein localization is currently being evaluated. A reduction in these parameters will be used as indices of improved barrier function with SB 203580. These studies will determine whether a reduction in lung inflammation correlates with a restoration of barrier function. The degree of protection provided by the p38 MAP kinase inhibitor SB 203580 could have important implications for inflammatory lung diseases such as CF. Background: Chronic pulmonary inflammation in CF is characterized by a robust neutrophil response associated with airway damage and failure to eliminate the pathogen, P. aeruginosa (PA). PA is a highly adaptable opportunist which quickly develops resistance to antimicrobials. Thus, the development of specific immunotherapy targeting the neutrophil recruitment without ablating the host's immune response to infection or promoting pathogen resistance would be ideal. Our group has identified IL-17 as a prime target for the development of immunotherapy due to the central role that the IL-23/IL-17 proinflammatory axis plays in neutrophil recruitment. However IL-17 does not mediate the early neutrophil recruitment seen in response to infection. Hypothesis: IL-23, acting synergistically with IL-1, is critical to early neutrophil recruitment during pulmonary PA infection. The primary effector cells are the IL-23-producing antigen presenting cells: alveolar macrophages (AMs) and myeloid dendritic cells (DCs). Methods: WT and IL-23-deficient mice were infected with PA at 1x10 6 cfu/50ul by intratracheal (IT) inoculation for 3 hours. BAL inflammatory cell counts, and cytokines and chemokines were measured. AM and DC cultures were infected for 3 hours in vitro and supernatant cytokines/chemokines were measured by luminex and ELISA; IL-23 levels were measured by Taqman. These studies were designed to elucidate the role of IL-23 in the early neutrophil peak and define AM-and DC-mediated cytokine/chemokine production. Recombinant murine IL-23, IL-1, and IL-23 + IL-1 were instilled via IT into WT and IL-23-deficient mice. BAL inflammatory cell counts and cytokines/chemokines were measured at 3 hours. These studies were designed to elucidate the role of IL-23 and IL-1 in the early neutrophil peak and define AM-and DC-mediated cytokine and chemokine production. Results: At 3 hours post-infection, IL-23 deficient mice had significantly lower percent neutrophils (p<0.02) and lower MIP1α, KC, and IL-6 (p<0.001) in the BAL. IL-17 was undetectable. There was no significant difference in bacterial load that could account for these cytokine/chemokine differences. Infected WT AMs elaborated significantly more MIP1a, GM-CSF, MCP-1, IL-1, G-CSF, IP-10, KC and IL-6 than the IL-23-deficient AMs (p<0.001) and the response was inoculum-dependent (p<0.01). DCs elaborated no IL-1 and exhibited IL-23-dependent differences in Mip-1α and MCP-1 production (p<0.05). In vivo studies of IL-23 and IL-1 effect demonstrated a synergistic increase in BAL neutrophil recruitment (p<0.01) and cytokine and chemokine induction (p<0.01). Conclusion: The first wave of neutrophil recruitment seen during PA infection is IL-23-dependent and IL-17-independent. AMs and DCs are critical to IL-23 and IL-1 indcution of this neutrophil recruitment. These studies identify IL-23 as a key mediator of neutrophil recruitment in the early stages of infection as well as the proximate mediator in the IL-23/IL-17 pro-inflammatory axis and suggest IL-23 as a potential target for anti-inflammatory therapy in the treatment of PA pulmonary infection.

Supported by the Cystic Fibrosis Foundation, American Lung Association and the NIH The airways are under constant assault from air-borne pathogenic material. Despite the intake of up to 100,000 bacteria per hour, the airways are sterile below the larynx in healthy individuals. The task of maintaining this sterility falls to the airway surface liquid layer (ASL), the protective twophased system consisting of the viscoelastic mucus layer and the periciliary layer (PCL) through which cilia beat, sweeping the mucus layer away from the lungs. The mucus layer, which is responsible for trapping pathogenic material, is comprised of mucins (high molecular weight glycoproteins), cellular debris, DNA, neurtifils, and more than 100 other proteins. This chemically heterogeneous mixture forms a viscoelastic gel that is thick enough to trap pathogens of various sizes and surface chemistries, while not sticking to the underling cellular / cilia layer, allowing the transport of trapped pathogenic materials away from the lungs. The performance of this trapping / transportation system is defined by the rheological properties of the mucus layer and the force imparted on the mucus. Therefore, understanding how mucins and other chemical components of the mucus layer interact with each other to form a successful mucus gel (i.e. one that is cleared from the airways) is crucial to understanding airway defense.

Here we present the results of physical and chemical composition studies of sputum samples collected from patients with chronic obstructive pulmonary disease (COPD) and cystic fibrosis (CF). The rheological properties of each sample was assayed using parallel plate rheolometry, probing the materials non-linear viscoelastic properties such as viscosity, elasticity, and yield stress. The physical properties of the sample are then correlated to the sample's chemical properties such as percents solids (divided between salts, proteins, and mucins), as well as the relative concentrations of the key airway mucins Muc 5B and Muc 5AC. Our results indicate that the physical properties of sputum are not well predicted from the total amount of biosolids in a given sputum sample, but by the relative concentrations of Muc 5B and Muc 5AC and the interactions of these molecules with themselves and the other proteins present. Further, we establish that the heterogeneous physical properties within a given sputum sample correlate to differences in the Muc5B and Muc 5AC concentrations. The Gene Modifier Study (GMS) was established as an effort to identify potential genetic modifiers of cystic fibrosis pulmonary disease and survival. During the course of this study over 1000 delta F508 homozygous CF patients classified as having mild lung disease, severe lung disease, or increased survival, have submitted both clinical data and blood samples for single nucleotide polymorphism (SNP) analysis. The original SNP analyses have shown a significant association between variants in the endothelin receptor A (EDNRA) gene, and CF survival, most markedly in female CF patients. Sixteen additional SNPs within and around the EDNRA gene have now been genotyped, and have implicated the 5' and 3' untranslated regions of the gene as having the most significant association with pulmonary disease in females (p<0.000001), suggesting quantitative differences as a possible mechanism for the association with pulmonary phenotype. We are in the process of saturating the 5' and 3' regions of EDNRA with an additional 20 SNPs to further delineate the genetic association. EDNRA binds endothelin-1 (ET-1) in airway smooth muscle cells, causing increased cell proliferation, smooth muscle contraction, and stimulation of inflammatory molecules. Because each of these effects is known to be deleterious to the CF lung, we hypothesize that the EDNRA variants found more commonly in "severe" CF females ("severe" alleles) are marking increased EDNRA expression compared to alleles found more commonly in "mild" CF females ("mild" allelels). Because the genetic association was strongest in CF females, we used the Matinspector software to analyze 2kb of EDNRA promoter sequence and found several putative binding sites for both estrogen and progesterone. We then used a BrDU assay to measure cell proliferation after stimulation with both estrogen and progesterone. These experiments showed that the ASM cells with the "severe" 5' EDNRA genotype proliferated at levels approximately twice that of the ASM cells with the "mild" 5' EDNRA genotype following stimulation with either estrogen or progesterone. Using a single base extension protocol and quantitative PCR on human airway smooth muscle cells, we were also able to compare EDNRA expression from the "severe" allele, and the "mild" allele. These comparisons of EDNRA expression demonstrate that expression levels appear to be approximately 20% higher from alleles found more frequently in the "severe" CF females. In addition, preliminary data suggest that stimulating the ASM cells with estrogen increases EDNRA expression by approximately 10 fold, and like the cell proliferation experiments, these increases are most pronounced in cell lines with the "severe" genotypes. These data suggest that the "severe" genotypes are marking alleles with increased expression, perhaps due to estrogen binding, that leads to increased ET-1 functional effects that over time are deleterious to the CF lung. CF patients do present with variable spectra of lung disease, of which infections are most life-threatening. β-defensins have an antimicrobial activity against a broad spectrum of microorganisms and are chemotactic agents for cells of the adaptive immune system, and therefore assist in combating these infections.

β-defensins 2-6 are part of a repeat region. This repeat region is polymorphic between individuals and therefore the dosage of these defensin genes/proteins varies.

We developed a real time PCR assay to quantify the number of βdefensin repeats in this region. Appropriate controls are needed for an accurate quantitative assay. Therefore we made 6 concatemeric constructs with 1 copy of DEFB1 and a particular number of DEFB4 copies, which ranged from 1 to 6 copies. Using these controls as standards, the number of defensin repeats could be accurately determined in DNA samples.

We then tested 146 F508del homozygous CF patients from Belgian (57 patients), Czech (52 patients) and South-Italian (37 patients) origin. The diploid number of repeats varied between 3 and 10. For each patient group, a higher number of repeats was found in the group of patients with milder disease (FEV1 >70%) compared to the group of patients with more severe disease (FEV1 < 70%) (student T test, P-values of 0.0006, 0.03 and 0.019 respectively). Moreover, in our cohort of 112 Belgian CF patients, CF patients of 25 years or older have a significant higher number of repeats than the CF patient group below 25 (P-value 0.05).

To evaluate this at the functional level, we cultured nasal epithelial cells from 3 individuals with a low number of repeats (i.e. 3 or 4 repeats) and 4 individuals with a high number of repeats (i.e. 8 repeats). The cells where grown in air liquid interface cultures. After differentiation, the cells were stimulated with 10ng TNFα.

In cells with a high number of repeats, DEFB4 expression, as measured by the extent of transcription, was strongly upregulated by TNFα. In cells with a low number of repeats, DEFB4 was not upregulated (P-value = 0.015).

We also tested the antimicrobial activity of epithelial cells. We challenged epithelial cells from 3 individuals (8 repeats) with a laboratory strain (PA01) of Pseudomonas aeruginosa and a clinical isolate (30 -300 CFU), either in combination with TNFα or without TNFα. After 3h, surviving bacteria were counted by a plating out method. Cells that were stimulated with TNFα 12h prior to the bacterial challenges were more bactericidal. The clinical strain was more vulnerable to the surface liquid than the laboratory strain. In epithelial cells from individuals having a low number of repeats, these effects were very variable from individual to individual.

In summary, the β-defensin region is a modulator of cystic fibrosis lung disease.

The pro-inflammatory response in cultured epithelium cells strongly correlates with the number of β-defensin repeats. Cells with a higher number of repeats respond to TNFα treatment, which in turn results in a better antimicrobial activity of the surface liquid. Rationale: Studies of affected twins and siblings demonstrate that modifier genes are major contributors to variation in cystic fibrosis (CF) lung disease severity. We performed genome wide linkage analysis to identify regions likely to contain modifier genes affecting severity of CF lung disease.

Methods: 683 individuals with CF from 360 families were analyzed. Pulmonary function data were collected from patient chart review and were supplemented with data from the US Cystic Fibrosis Foundation Patient Registry. To minimize environmental variation, only data obtained while subjects were living with an affected twin or sibling were analyzed. The pulmonary phenotypes were defined using the best CF-specific percentile for FEV1 (Kulich, et al) within the last year of available PFT data as a crosssectional measure (MaxFEV1CF%) and using two longitudinal measures: the lifetime average CF-specific % for FEV1 (AvgFEV1CF%) and the estimated percent-predicted FEV1 at age 20 (EstFEV1%pred@20yrs, Schluchter, et al). Longitudinal measures were derived from a minimum of 4 years of PFT data. Short tandem repeat markers (STRs) were typed in all affected individuals and their parents (Marshfield Genotyping Center: 402 markers or DeCode Genotyping Center: 1030 markers). Two-point and multipoint linkage analyses were performed using Sequential Oligogenic Linkage Analysis Routines (SOLAR).

Results: Patients represented the spectrum of lung disease severity, with MaxFEV1CF%'s ranging from 0 to 1, mean 0.70 ± 0.26 and AvgFEV1CF% ranging from 0.01 to 0.99, mean 0.59 ± 0.24. The MaxFEV1CF% was predictive of AvgFEV1CF% (r=0.89, p<0.0001) for the 486 individuals for whom both measures were available. The two longitudinal measures were also highly correlated (r=0.80, p<0.0001). Linkage was found at chromosome 5 for all three phenotype definitions. Peak multipoint LOD scores on chromosome 5 occurred at 196 cM for MaxFEV1CF% and AvgFEV1CF% (LOD 3.0 and LOD 3.4, respectively) and at 191 cM for EstFEV1%pred@20yrs (LOD 2.8). Single point LOD scores on chromosome 5 peaked at marker AAAT072 (3.3 for MaxFEV1CF%, 3.4 for AvgFEV1CF%, and 1.88 for Est-FEV1%pred@20yrs). The region of linkage encompasses approximately 6 megabases near the telomere of chromosome 5q.

Conclusions: Chromosome 5 appears to contain one or more genetic modifiers of CF lung disease severity.

Supported by the NHLBI, CFF and Genome Canada through the OGI. Cystic fibrosis-related diabetes (CFRD) is the most common extrapulmonary complication of CF and is an increasingly important contributor to morbidity and mortality as CF patients live longer. While pancreatic fibrosis and loss of exocrine and endocrine tissue are common in CF, 20-30% of CF adults develop defects in insulin secretion and accumulation of islet amyloid polypeptide, features typical of type 2 diabetes (T2DM) in the general population. To test whether modifier genes play a role in CFRD, we compared concordance rates for CFRD in 68 pairs of monozygous (MZ) twins, 22 sets of dizygous (DZ) twins, and 470 sets of 2 or more siblings (1176 individuals with CF). Criteria for defining CFRD included physician diagnosis, treatment with insulin/oral agent, and 2 episodes of glucose ≥200 mg/dl. MZ twins were highly concordant for CFRD (9 of 12 pairs, 75%). The young age of DZ twin recruits precluded analysis of this group in isolation (0 of 4 pairs were concordant). Twelve of 71 (14%) sibling pairs were concordant for CFRD. With heritability defined as: h 2 =2*(MZ concordance -DZ concordance), and including siblings as a proxy for DZ twins, heritability is estimated as ~1.0. The same results were obtained considering only same-sex DZ twins and siblings, correcting for differences in age and duration of clinical follow-up, or restricting analysis to ∆F508 homozygotes. These data support a significant role for one or more modifier genes in development of CFRD. We then tested whether CFRD correlated with a strong family history of adult-onset diabetes (at least 1 first-degree or 2 second-degree relatives on the same side of the family). Of those reporting family history of diabetes, 20 of 69 had CFRD, compared to 34 of 271 with no family history (OR=2.84 [1.5-5.4 ]; p=0.001). This correlation persisted after adjusting for age, sex, and pancreatic insufficiency (OR=2.6; p=0.028). Thus, family history of diabetes correlated with increased risk of CFRD. We then tested whether variants in TCF7L2, a transcription factor in the Wnt signaling pathway, that have been reproducibly associated with T2DM in the general population were associated with CFRD in our study subjects. Genotyping of four single nucleotide polymorphisms associated with T2DM (rs4506565, rs7903146, rs12243326, rs12255372, here termed snpA-D) and transmission disequilibrium testing (TDT) of 53 parent-parent-child trios revealed significant overtransmission for snpB (31:16, p=0.03) and snpC (28:15, p=0.047), and possible overtransmission for snpA (33:22, p=0.1) and snpD (31:18, p=0.06). In every case, the TCF7L2 allele overtransmitted to patients with CFRD is the same allele that confers increased risk for T2DM. Furthermore, individuals with CFRD who were homozygous for risk alleles were diagnosed at a significantly earlier age (average 14.6 vs. 20.1; p=0.03). These data support a key role for modifier genes in development of CF-related diabetes, and demonstrate that CFRD and type 2 diabetes may share disease mechanisms such as alterations in Wnt signaling. Supported by NIH DK076446, HD27799, DK44003 and HL68927, and CF Foundation grant CUTTIN06P0. Cystic Fibrosis (CF) phenotypes and survival are highly variable among DF508 homozygous patients, pointing to the existence of modifier genes and/or environmental factors that contribute to this disease. Studies to identify genetic modifiers of CF are being carried out using DNA from homozygous DF508 CF patients. Important clinical features, such as severity of lung disease, liver disease and meconium ileus (MI) status, are well defined. TGFβ1 has been previously identified as a modifier of CF lung disease (Drumm et al., NEJM, 353(14) : [1443] [1444] [1445] [1446] [1447] [1448] [1449] [1450] [1451] [1452] [1453] 2005) , but it does not explain all of the genetic heterogeneity in this population. Current evidence suggests that mucus is involved in the progression of CF, making the MUC genes prime candidates as modifiers of lung disease and/or other phenotypes. METHODS: Our approach to evaluate MUC genes utilizes both variable number tandem repeat (VNTR) polymorphisms and single nucleotide polymorphisms (SNPs). VNTR polymorphisms in the MUC2 and MUC5AC genes (n=405 and n=366 patients, respectively) were detected by Southern blotting under conditions fully optimized to maximize the allele size resolution. To minimize gel to gel variation, a genomic DNA mixture with the most common MUC2 and MUC5AC alleles was used as internal markers. Accuracy and reproducibility were evaluated by duplicating the Southern on the critical DNA samples for MUC2 and MUC5AC genes (n=85 and n=63 patients, respectively). Fifty SNPs in MUC5AC, MUC5B, MUC2, MUC1 and MUC4 genes are being tested using Illumina technology in 808 CF patients. RESULTS: Preliminary analysis suggests there are significant differences between CF patients with "severe" and "mild" lung disease for both MUC2 and MUC5AC allele distribution, which is mainly driven by the male population; exhaustive statistical data analysis still is underway. The VNTR data also suggest significant association between the larger MUC2 allele size and CF patients with MI. The ongoing VNTR analysis will be complemented by MUC gene SNPs being genotyped. CONCLUSIONS: Initial results indicate that we can reproducibly characterize MUC2 and MUC5AC VNTR alleles. Additional characterization of MUC2 and MUC5AC VNTR alleles, coupled to SNP data, will allow us to better define the significance of MUC gene variations as modifiers of different CF phenotypes.

Supported by CFF PEREZV06G0 (JPV), CFF KNOWLE00A0 (MK), CFF R026-CR02 (WKO), NIH RR00046, R01 HL68890, and CFF DRUMM00A0.

Reporting for the Gene Modifier Study group (MRK). We have previously reported that βENaC transgenic mice, which overexpress the beta subunit of the amiloride sensitive sodium channel (Scnn1b) specifically in the airways, share common features with CF, including increased ENaC activity, reduced airway surface liquid, mucus accumulation and obstruction, inflammation, and death. Interestingly, analysis of this model also suggested the existence of potential genetic modifiers of phenotype severity, and we speculated that identification of these modifiers would provide novel insights into disease phenotype. To establish a set of reagents that could be used to uncover genetic modifiers, we have bred βENaC transgenic mice from two independent founder lines (6608 and 6047, B6:C3 background) onto several strains of inbred mice, including C57BL/6N, C3H/HeN, BALB/cJ, FVB/J, and 129/SvJ. These studies revealed dramatic phenotypic differences as measured by survival (0 to 80%) among strains and between lines. All lines thus far tested show ~2-3 fold increases in amiloride sensitive short-circuit current as measured by Ussing chambers in the trachea. Complete phenotypic characterization of C57BL/6N Line 6608 at backcross generation 12 reveals high survival (66±5% in comparison to 54±4% of the mixed B6:C3 background), yet the mice maintain the pulmonary features of the originally reported mice, including increased mucus plugging, mucous cell hyperplasia, neutrophilic and eosinophilic inflammation peaking at early timepoints (5 days -2 weeks). Lymphocytic nodules, which are not commonly seen in 4-6 weeks-old animals with mixed strain background, are a common feature in the C57BL/6N congenic line. Emphysema and early airway epithelial cell necrosis, two phenotypes initially not strongly associated with transgene expression, are also observed. Line 6608 on BALB/cJ and C3H/HeN backgrounds has reduced survival compared to the C57BL/6N background, and generation 8 backcross lines are now being evaluated in these two strains for other phenotypic characteristics. Analysis of backcross data from Line 6047, which has low survival on all genetic backgrounds tested to date, including C57BL/6N, suggested that the transgene may have integrated onto a C3H locus with a dominant negative effect on survival. Genetic analysis using genome-wide SNP genotyping revealed a region of Chromosome 4 linked to the transgene in Line 6047 (LOD score > 3.0). Further analysis of this region is underway. In summary, backcrossing onto different genetic backgrounds is revealing genetic modifiers for phenotypes in the βENaC overexpressing model. Characterization of these phenotypic and genetic differences should provide clues about the mechanisms relevant to disease development. Furthermore, inbred lines with variable phenotype will likely be an important reagent for the CF community as the utility of this model is evaluated in future studies. Supported by NIH (SCOR P50 HL60280) and CFF (MALL04GO, ONEAL07GO). Introduction: New York State screens newborns with immunoreactive trypsinogen levels within the top 5% of all infants, for 32 common CF gene mutations. Infants found to be heterozygote carriers are referred to a CF Center to determine sweat chloride concentration. The proposed abnormal sweat chloride value for this group of newborns is ≥ 40 mmol/L, ≥ 3 standard deviation (SD) above the mean (Farrell, 1996) . This retrospective study reports on the mean sweat chloride value + 3 SD in CF heterozygote newborns who have been referred for evaluation to SUNY Upstate Medical University Cystic Fibrosis Center, and regarding the genotype of these infants with abnormal sweat chloride levels.

Method: From October, 2002 to December, 2006, 404 infants were referred for positive CF screening, and 300 of these (75%) were identified as heterozygote carriers by the screening program. At our Center, these patients underwent pilocarpine iontophoresis, followed by collection of sweat (≥ 20 µl) in Macroduct ® coils. Sweat testing was performed successfully in 279 of the 300 (93%) patients. The patients' age (mean ± SD) at the time of sweat testing was 45.8 ± 19.0 days. For infants with an initial sweat chloride level ≥ 24 mmol/L, the sweat test was repeated within a week and a complete gene sequencing was requested (Quest Laboratory, CA). Results: The sweat chloride level (mean ± SD) in newborns who were heterozygous for a CF mutation (excluding those who were found to have an additional mutation or deletion) was 10.4 ± 4.5 mmol/L (n = 273). The mean + 3 SD was 24 mmol/L, which defined our minimal value for an abnormal test. Eleven infants had sweat chloride values of 27-91 mmol/L (Table) ; 6 of them (55%) were subsequently diagnosed with CF by complete gene sequencing. Four of the 5 remaining patients (80%) had ∆F508 mutation coupled to the 5T variant on the opposite chromosome 7 (sweat chloride levels, 24-41 mmol/L).

Conclusions: The reference range for sweat chloride in CF heterozygote infants appears to be significantly lower for some Centers than previously reported. Thus, each CF Center should consider evaluating the cutoff values for the test at their site. Moreover, the 5T polymorphism may account for sweat chloride elevations in heterozygote infants. Method: We reviewed the charts of all 28 patients with CF who were referred as a result of the New York State newborn screening program to the SUNY Upstate Medical University CF Center from October, 2002 through April, 2007. We included the patients who were identified by the State as heterozygote carriers of one of 32 common CF gene mutations, and whose second mutation was identified only after complete CF gene sequencing.

Results: Seven of the 28 patients (25%) met inclusion criteria. Six of the 7 patients (86%) were compound heterozygous for a novel or a rare CF gene mutation. One patient was compound heterozygous for a large deletion in the CF gene. The genotype and clinical status of the 7 patients are shown in Table. All patients have been pancreatic sufficient to date.

Conclusions:The clinical effect of compound heterozygosity as a result of novel or rare CF gene mutations appears to be mild in early childhood. Prior to newborn screening (with the exception of the patient with CFTR deletion), these patients may not have been diagnosed with CF in the first few years of life. These patients may have subclinical airway inflammation and thus benefit from early treatment. Patients with CF manifest symptoms in the pancreas, respiratory tract, male reproductive tract and sweat gland due to mutations in CFTR. Patients with non-classic CF have disease in a subset of these organ systems. Most non-classic CF patients have two disease-causing mutations in CFTR and at least one mutation permits residual CFTR function. A subset of non-classic CF patients have only one CF-causing mutation after screening for a panel of common CF-causing mutations or following mutation scanning of the coding region of CFTR. These patients present a diagnostic dilemma and a challenge for genetic counseling. We evaluated 10 CF patients with only one CF-causing mutation identified after a screen of 97 CFTR mutations (3 patients) or scanning of the coding region of CFTR (7 patients). Nine of these patients, including one set of siblings, have non-classic CF with borderline or elevated sweat [Cl -] plus lung disease (Table) . One patient is pancreatic insufficient and has classic CF. Many of these patients have features which are consistent with CFTR dysfunction including a CF-like nasal potential difference (NPD),P. aeruginosa infection, or congenital bilateral absence of the vas deferens (CBAVD), suggesting that they have a second CFTR mutation. Mutations that are not detected by screening methods include insertions or deletions, mutations outside of the CFTR coding region that affect RNA splicing or expression, or mutations in the coding region of CFTR that were missed by screening methods. To exclude the third possiblity, DNA sequencing of the 27 exons and flanking introns of CFTR was performed. A second mutation was identified in the coding region of CFTR in 6 of the 10 patients; 3 had screening for 97 known CFTR mutations (Genzyme), while the remaining 3 had comprehensive scanning of the coding region of CFTR by modified TGGE (Ambry). Each of the mutations identified by sequencing has been previously described in patients with CF and is predicted to cause CFTR dysfunction. These results reaffirm that patients with one CFTR mutation who have biochemical and clinical features of CF are likely to have a second mutation in the coding region of CFTR. Thus, we suggest sequencing CFTR in patients with only one mutation after mutation screening before employing other more complex genotyping methods (insertion/deletion or RNA analysis). Diagnostic criteria for confirming CF in symptomatic individuals includes two positive sweat tests or two known disease-causing CFTR mutations. Accurate sweat testing is performed at accredited CF Centers, while CFTR testing is available through national and specialty genetic labs. Most labs offer analysis of a basic panel of CFTR mutations as recommended by the American College of Medical Genetics (ACMG), while specialty laboratories may offer an expanded panel or full sequencing/scanning. We report on one Center's use of these methodologies to confirm the diagnosis of CF presenting in adulthood.

The charts of 20 patients diagnosed with CF at 〉18 years of age were reviewed. 19 were sweat tested. All were genotyped. Results: Twelve (63%) pts had two positive sweat tests (>60 mmol/L). An additional five (25%) had at least one borderline result (40-59 mmol/L). Two pts had negative results (35-39 mmol/L). One refused sweat testing because DNA analysis through the ACMG panel had confirmed the diagnosis prior to initial consultation. Genotyping results for the 20 pts are summarized in the table below.

Rare mutations identifiable only through gene sequencing accounted for 18/40 alleles (45%) in our population. Importantly, among this group, only one pt had a negative sweat test, a suggestive 39 mmol/L. Two pts with positive sweat tests and clinical symptoms failed to reveal any CF mutations after sequencing. We continue to follow pts without genotypic confirmation, based on their clinical presentation and sweat chloride levels, and have recommended additional evaluation, including nasal potential difference studies

In our series of adult-diagnosed patients, sweat test results were positive, borderline, or suggestive in all cases tested. Sweat testing costs $100-$250 and results are ready in a day. Genotyping costs $400-$2500 and takes several weeks. We acknowledge that circumstances may arise where reliable sweat testing is not conveniently available; but in our series, genotyping with the ACMG panel would have diagnosed 10% of pts; using an expanded panel would have diagnosed 35%. Genotyping is an important tool for genetic counseling, determination of eligibility for research studies, furthering knowledge of CFTR dysfunction and CF pathophysiology, and for confirming a CF diagnosis after borderline or suggestive sweat test results. Based on our findings and the dramatic difference in cost, we conclude that sweat testing should remain the first approach in the diagnostic workup of adult patients with a clinical presentation suggestive of CF.

In Colorado, 318 infants with CF (non-meconium ileus) have been diagnosed with CF by a two tiered immunoreactive trypsinogen (IRT/IRT) based newborn screening approach. The IRT/IRT algorithm has been recently adopted by other screening programs with two mandatory screening tests. While most infants in Colorado have been successfully identified, the program has had a missed case rate of approximately 5%. The more common approach to CF newborn screening is the IRT/DNA method in which the blood spot of infants with an initial elevated IRT is tested for the most common CF mutations. The initial IRT cutoff is lower in the IRT/DNA programs than in the IRT/IRT programs, resulting in a lower missed case rate. The considerable number of carriers identified through the IRT/DNA approach puts a significant burden on the genetic counseling community, as carriers are identified at a rate of 1/20-1/25 of positive IRTs We propose an IRT/IRT/DNA newborn screening algorithm that will maximize sensitivity and specificity while minimizing the number of identified carriers. Using new database technologies in the newborn screening lab we will be able to identify those infants with an elevated first IRT (>60ng/ml, approximately 97th percentile). All infants with an IRT >60ng/ml will have a repeat IRT on their second state mandated blood-spot. If the second screen is also elevated (>60ng/ml), the blood spot will be tested using a panel of 43 mutations, including mutations specific to the Hispanic community. Infants with one or two CFTR mutations will have a sweat test to confirm the diagnosis, or rule out CF. We compared the projected statistics of our current method IRT/IRT to the new IRT/IRT/DNA method, and to IRT/DNA is presented in the table, based on 70,000 births per year in Colorado. Four infants (1.5%) identified under the current IRT/IRT protocol would not have been identified by the mutation panel proposed in the new algorithm, out of 288 genotyped, non-meconium ileus infants. Three of these missed cases are Hispanic. Two would be identified using an extreme IRT cutoff of the 99.9th percentile (150ng/ml). The projected missed case rate would be <0.7% (0.2 -2.5%, 95% CI), using the IRT/IRT/DNA algorithm, with 23 carriers identified, maximizing both sensitivity and specificity. This algorithm may provide a better alternative to the IRT/IRT screening methods in states with two mandatory screening tests, and has advantages over both the IRT/IRT and IRT/DNA methods. Newborn screening for cystic fibrosis (CF) is rapidly expanding and has been implemented in at least 30 states. Although most newborn screening assays are done using biochemical testing, many laboratories screening for CF include both biochemical and molecular testing of multiple alleles in the Cystic Fibrosis Transmembrane Conductance Regulator gene. In response to the growing need for proficiency testing (PT) materials for molecular testing, the Centers for Disease Control's Newborn Screening Quality Assurance Program (NSQAP) in collaboration with the University of Wisconsin School of Medicine and Public Health, the Johns Hopkins Hospital, and Case Western Reserve University, created a repository of dried-blood spot specimens with known mutations in the CFTR gene to be used in a PT program. Twenty milliliters of blood was collected voluntarily from adult donors with CF and sent to the NSQAP laboratory. Each specimen was adjusted to a hematocrit of 55% before being spotted onto Whatman 903 paper (75 µL per spot), dried, and stored at -20°C with desiccant. Proficiency testing (PT) panels consisted of 5 to 7 blind-coded specimens from adult donors. The panels were sent quarterly to laboratories worldwide that test specimens for CF using molecular methods. Laboratories were asked to report the genotype, method used, and the presumptive clinical assessment of each specimen. Twenty-two laboratories participated during both Quarters 1 and 2, 2007. The laboratories used 16 different methods ranging from in-house assays to commercially available kits. Most reporting laboratories tested the following 3 alleles -∆F508, G542X, and G551D. Another twelve alleles were detected by most participants. Nine more alleles were common among commercially available kits. Laboratories were evaluated based on the clinical assessments. Mutations that were not detected by a particular method were not evaluated. Overall, the laboratories performed well. Data compiled from both quarters demonstrated that there was 1 incorrect clinical assessment and 2 amplification failures. Developing a PT program for DNA-based testing is complicated by the number of methods and different alleles each laboratory chooses to test. Though molecular testing for CF may be complex, PT monitors the laboratory's ability to test multiple alleles, including uncommon alleles, the limitations of various assays, and the different algorithms used for screening. The repository will also allow storage and access to rare specimens that may be useful for future research but are not readily available. who did could not reproduce. Genetic counseling focused on a patient's parents, who were counseled about their recurrence risk at the time of the child's diagnosis. Today, CF is a disease of adulthood. In 2002, >40% of CF patients in the US were >18; by 2010 it will be >50%. Together with advances in assisted reproductive technology (ART), reproduction and recurrence risk are now important issues for adolescent and young adult CF patients.

Methods: A 19 item questionnaire was developed from the results of prior semi-structured interviews with 18 CF patients age 16-25 years. Knowledge based questions (medical issues, inheritance, and reproductive options/risks) as well as communication patterns (preferred resources for learning about CF and preferred people with whom to talk about reproductive issues) were addressed. Recruited from the UAB CF clinic population, 51 patients age 15-29 (mean 21), 24 male (47%), 27 female (53%), completed the questionnaire.

Results: Regarding autosomal recessive inheritance of CF, only 33% knew that two carriers have a 25% chance of having a child with CF, and 25% knew that two carriers have a 50% chance of having a child who is a carrier. However, 82% knew that two carriers could have a child who did not have CF, and 52% knew that two carriers could have a child who did not carry CF.

On their own reproductive risks, 59% knew that a CF patient had a 0% chance of having a child with CF if their partner was not a carrier, but only 26% knew that all their children would be carriers even if their partner was not a carrier. In the scenario of a CF patient with a CF carrier partner, 44% knew that a child had a 50% chance of having CF, and 24% knew that a child had a 50% chance of being a CF carrier.

Most patients knew about their reproductive potential, as 96% responded that CF patients are able to have children. However, when asked about whether the chance for having children was different for males and females with CF, 65% answered that it was more difficult for males, 8% that it was more difficult for females, and 27% answered "not sure." While 62% reported that they knew that there were options for male CF patients who wanted to have children, only 26% knew of ART.

Conclusions: Despite widespread availability, the lack of knowledge of adolescents and young adults with CF about the genetics of their disease continues. Furthermore, these patients are unaware of both modern technologies that could enable them to have biological children and the risk of those children having CF. This study illustrates the changing needs of patient education as medical knowledge progresses. CF patients would benefit from further genetic knowledge and counseling to enable them to make informed decisions about reproduction as they mature into adulthood. Center at the University of Minnesota is one of three sites in the state providing confirmatory testing and follow up services for newborns identified by screening. While CF NBS identifies children with CF, most of the infants with positive screening results are carriers. Our goal is to provide genetic counseling to every CF NBS patient seen at our center, and we believe that a protocol incorporating genetic counseling in the initial care plan for infants both with and without CF is imperative. The literature has shown that families who obtain genetic counseling through the CF NBS process recall genetic information more easily and accurately and are more likely to have testing to determine parental carrier status. For parents of a child that is determined to be a CF carrier, it is especially important to find the optimal method and timing of genetic counseling as many of these families are from several hours away and therefore less likely to return to clinic. To provide genetic information and emotional support for the families of infants screening positive on CF NBS, our center has a genetic counselor who serves as the CF NBS coordinator and clinical contact for the family. This allows many opportunities to speak with the genetic counselor and ask questions, as well as learn about their child's diagnosis or carrier status and the subsequent carrier testing recommendations for the infant's parents and families. To assess the success and impact of the Minnesota Cystic Fibrosis Center's NBS follow-up program and the incorporation of genetic counseling, an anonymous questionnaire was developed for parents of infants who were seen at our center due to a positive CF NBS result. Questionnaires were mailed to parents of all infants seen at our center for a sweat test and genetic counseling due to a positive cystic fibrosis newborn screening result. As of the last mailing, this totals 54 families. Two questionnaires were returned due to incorrect address and 28 questionnaires were returned answered, indicating a response rate of 54% (28/52). Responses overwhelmingly indicated that parents were satisfied with our center's algorithm for the CF NBS follow-up program and found the information and support provided through genetic counseling to be a useful and recommended portion of the program. As CF NBS continues, it is critical that we learn about the patient's experience with genetic counseling and the NBS program, as well as identify areas needing improvement. Genetic counseling is vital to the comprehensive success of our center's program, and we will report on the responses gathered from the families identified through CF NBS this first year, as well as discuss the lessons learned from setting up such projects on a state-wide basis. Background & Aims: Aberrant splicing and nonsense mediated decay (NMD) lead to dysfunctional mRNAs by skipping exons and to a reduced number of functional mRNA respectively. Both mechanisms have a strong quantitative aspect and may determine whether a CF patient develops a classic or atypical disease phenotype. In order to approach these highly important questions we wanted to establish a new quantitative real-time PCR based assay which allows allele specific quantification on cDNA level. Using this assay we like to determine the exact proportions of the F508del and non-F508del CFTR mRNA in CF patients compound heterozygous for the F508del mutation (for example in CF patients carrying the F508del and a nonsense mutation such as the R553X).

Material & Methods: Materials: We used genomic DNA (gDNA) and total RNA (extracted from white blood cells and nasal epithelial cells respectively) from CF patients with compound heterozygosity for the F508del mutation, homozygosity for F508del mutation and from healthy individuals (controls).

Methods: The LASQ (Ligation dependent Allele Specific Quantification) assay comprises 4 reactions: 1. Reverse transcription of CFTR mRNA into cDNA using gene specific primers (all RNA specific). 2. Overnight hybridization (12-16h) of the CFTR cDNA with either the F508del specific or the wt specific oligo probe pair provided by Jan Schouten (MRC Holland). 3. Ligation of the hybridized oligo pairs using the Ligase 65 enzyme from MRC Holland. 4. Quantitative real-time PCR of the allele specific ligation products on the LightCycler (Roche).

Results: In order to establish the LASQ assay we first validated it using gDNA instead of cDNA as template. Mixing experiments were performed to verify the accuracy of the assay. In brief, gDNA (c=20mg/L) of a F508del homozygous and a F508del compound heterozygous CF patient were mixed in such a manner that 0.1, 0. Amplification products of the F508del and the wt allele (both 328bp long) were analyzed by gel electrophoresis (PAGE) and direct sequencing (ABI 3100) to control specificity. Our results using gDNA and cDNA showed that there occurs unspecific hybridization/ ligation for both probe pairs. The proportion of unspecific amplification products varies between 0.01 and 0.005 and increases the lower the initial number of templates is. However, the specificity of this assay can be significantly improved by increasing the hybridization temperature and/or decreasing the ligation time.

Conclusion: Although some minor limitations concerning allele specificity the LASQ assay has been proven to be an accurate, reliable and reproducible method for allele specific quantification and may be applied for several important questions in Cystic Fibrosis such as the exact determination of the amount of NMD of CFTR mRNA containing a premature termination codon (PTC) or the allele specific determination of aberrant splicing of CFTR mRNA . Background & Aims: As clinical presentation varies significantly among CF patients with the same genotype, e.g. in the F508del homozygous, it is evident that factors in addition to the CFTR genotype such as modifier genes, are involved in determining disease severity. However, only one of several previously postulated modifier genes, the TGFβ1 gene, could recently be confirmed in a large association study. Hence the identification of new modifier genes is a very important task in order to find new explanations for the heterogeneity of pulmonary disease in CF patients. We decided to search for new potential modifier genes applying a quantitative proteomic approach comparing the proteomes of a wild type (16HBE14o-) and a F508del homozygous bronchial epithelial cell line (CFBE41o-). The main goal of this study is the identification of up or down regulated proteins in the CFBE cell line which may act as modifiers of CF disease.

Material & Methods: Materials: We used two bronchial epithelial cell lines, e.g. a wild type (16HBE 41o-) and a F508del homozygous (CFBE 14o-) cell line which we obtained from Dr. Gruenert (California, USA). Additionally, we also used nasal cells from F508del homozygous CF patients obtained either from nasal brushings or nasal polyps.

Methods: Proteome analysis was performed by making 2D-Gels using high sensitive staining protocols (Ruthenium and Deep Purple). Quantitative analysis was accomplished applying the powerful DIGE (Difference in gel electrophoresis) method. For each DIGE experiment we made 4 gels whereby each cell line was twice labelled with Cy3 and Cy5 (=4 technical replicates). Finally, identification of the protein spots was done by the use of a MALDI-TOF mass spectrometer.

Results: In a first step we established the proteomes of the two bronchial epithelial cell lines. We were able to optimize protein extraction and 2D gel electrophoresis in such a manner that the proteomes of the two cell-lines looked very similarly and the assignment of spots could easily be done. Protein spots from both cell lines were analyzed using our mass spectrometry (MALDI-TOF MS) and allowed the identification of more than 60 different proteins so far. In a next step we quantitatively compared the proteome of the two cell lines using the 2D-DIGE method leading to the identification of 8 proteins which are down regulated and 5 proteins which are up regulated at least twofold in the CFBE cell line. Out of the aforementioned 13 differently expressed proteins 3 could already be identified because they were among the 60 previously determined proteins. While Glutathione S-Transferase P and Protein S100-A11 (S100 calcium binding protein) were down regulated (3.13 and 2.08-fold respectively), Superoxide Dismutase was up regulated (2.62-fold) in the CFBE cell line.

Conclusion: Comparative quantitative proteomics using the DIGE method is a promising tool in search of potential new modifier genes which may unravel one of the key problems in CF: the large heterogeneity of pulmonary disease in F508del homozygous patients. Background: Cystic Fibrosis is one of the most common autosomal recessive disorders among Caucasians, and manifests a wide range of disease severity. Although this range of disease expression can be attributed, in part, to specific mutations within the CFTR gene, much of this variability has not been adequately explained. From the Gene Modifier Study (GMS-a multicenter study of 1,306 CF patients), ten genes were tested as potential modifiers of CF, and the TGFβ1 codon 10 CC genotype was associated with lung disease severity. Objective: To test whether the adverse codon 10 CC genotype is associated with higher circulating levels of TGFβ1, compared to the TT genotype in CF patients and healthy controls. If true, then a link will be established between genotype, disease severity, and circulating levels of TGFβ1, and have implications for novel treatment of CF patients. Methods:

The study includes 60 clinically stable CF patients and 60 healthy control subjects equally distributed between the CC and TT genotypes. The CF patients enrolled are age 8 and older, of both genders, and all ethnicities that fulfilled the standard diagnostic criteria for CF, using the genetic information from the GMS. Healthy controls are age 18 and older, Caucasian males and females, obtained from the Environmental Polymorphism Registry (EPR), a DNA registry of 20,000 self-reported normal volunteers. We genotyped 485 blood samples from the EPR to define 30 healthy control subjects for each of the CC and TT cohorts. Subjects have a blood draw of 35 ml. Blood is divided into 4 tubes and each tube is used to measure a different parameter: CBC with differential; TGFβ1 levels in platelet poor plasma by Quantikine Human TGFβ1 ELISA kit; TGFβ1 levels in the buffy coatwhich includes platelets; and TGFβ1 mRNA levels in lymphocytes from the buffy coat, using real-time PCR Roche Light Cycler. CBC with differential is performed to quantitate lymphocytes and platelets in order to reference TGFβ1 protein and RNA levels to the number of circulating blood cells. Analyses will include graphical comparisons between the two groups, Chi Square analysis, student's two-sample t test, and one way Analysis of Variance (ANOVA). Results: Pilot studies show that ELISA TGFβ1 measurements are reproducible and mRNA levels can be quantified. We have currently enrolled 14 CC genotype and 26 TT genotype of the 60 CF subjects and 11 CC genotype and 19 TT genotype of the 60 healthy controls. Blood samples have been collected and processed. Conclusion: Genetic variants that predispose to more severe CF disease are potential targets for new therapies. We are testing the hypothesis that the adverse (codon 10) CC genotype is likely to reflect increased transcription and/ or TGFβ1 protein synthesis/ secretion. If true, "anti-TGFβ1" therapies could provide a novel therapy in CF. *Reporting for the Gene Modifier Study group. Supported by CFF KNOWLE00A0, CFF DRUMM00A0, GCRC RR00046, NIH 5R01 HL68890. Aim: Centralized periodic evaluations of data from the screening laboratories and CF centers by AFDPHE (French association for screening and prevention of infant handicaps), were analysed to optimize the efficiency of the program.

Methods: The strategy combined d3 IRT assay/DNA analysis (Kit Elucigen CF-30 ARMS) /d21-fail-safe IRT. Revised IRT-cut off levels were decided in order to maintain the percentage of positive screens around a 0.5% target. A questionnaire yearly sent to the CF centers collected the CF false negative cases.

Results: From 2002 to December, 31 2005, 625 CF cases were detected through NBS (2 717 905 screened infants). The I period (P) (n=553 248); d3-IRT:60 µg/l and d21-IRT: 30 µg/l showed a) 0.82% infant above the d3 cut-off, generating an excess of costly DNA tests b) 20% had a d21-IRT above the cut-off leading to a very high number of ST (n=808) with an extremely low rate of CF (n=1). By increasing slightly both d3/d21 IRT cutoff levels (65 µg/l, 40 µg/l) during the II P (n=1 172 868) a) the number of positive screens decreased to 0.64% b) 11.5% of infants with elevated d21-IRT had to be referred for ST (n=768) with 9 CF diagnosis. Since the risk to had true CF remained very low among infants with no detected mutations, during the III P (n= 991 789), d21-IRT concerned only the ones with d3-IRT>100 µg/l and the percentage of infants requiring a ST was reduced to 2.5% (n=146, 5 CF). The incidence of CF detected during these 3 P did not vary significantly (1/4289-1/4635-1/4081). Another point of concern was the false negative cohort; with a follow-up period over 18 months, 23 CF were detected on clinical symptoms (3.4%) at a mean age of 10 months. Only 3 were directly related to the modifications of the strategy.

Conclusion: centralisation of data made possible changes in the flow charts of the screening strategy to limit the number of false positive cases without significant alteration of the global performance of the program. There were three infants of mixed AA/Caucasian origin diagnosed with CF, all had ∆508/genotype on screening.

Mutation data on the 53 additional AA CF patients followed at the 11 CF Centers in New York was collected. Six patients have not been genotyped. There were 6 patients who were homozygous for ∆508, 14 patients had only one mutation identified and 5 were not found to have any CF mutations. ∆508(27%) and 3120+1G>A(11%) were the most common mutations. There are five CF patients (including 3 infants diagnosed by NBS) with mixed racial origin who are not included in this analysis.

Mutations Results: The median (range) MBL plasma level was 2.46 (0.04-11.14) µg/mL in CF patients, compared to 1.43 (0.02-11.30) µg/mL in controls. MBL2 genotype frequencies were similar in patients and controls. Lung function level was not correlated with MBL genotype or plasma level. The frequency of colonization with Pseudomonas aeruginosa was 41% in MBLdeficient (XA/O and O/O genotype) children with CF and 15% in MBL-sufficient (A/A and YA/O genotypes) children (p=0.04). We found a trend of a decreased age of first onset of colonization with Staphylococcus aureus, Haemophilus influenza and Pseudomonas aeruginosa in MBL-deficient CF patients (p=0.19, p=0.10, and p=0.14, respectively).

Conclusions: MBL-deficiency was associated with an increased frequency of Pseudomonas aeruginosa colonization in children. MBL deficiency was not associated with lung function deterioration. In a larger cohort we hope to confirm that MBL deficiency influences age at onset of bacterial colonization in CF patients.

Acknowledgments Although there is some evidence that CFTR gene mutations may be associated with respiratory diseases, little is known about the relationship between CFTR gene mutations and idiopathic bronchiectasis (Ngiam et al. 2006) . We have recently showed that a rare allele (-33G>A) in the minimal CFTR promoter, previously reported in patients with idiopathic bronchiectasis ( Using supershift assays, we demonstrated that these transcription factors bind CFTR promoter in vitro. A functional analysis, by using co-transfection assays with expression vectors of each transcription factor in pulmonary epithelial cells, showed that NRF2, IRF1 significantly decrease CFTR expression, whereas IRF2 and Sp1 increase it.

In an attempt to further elucidate the mechanisms involving these factors in the CFTR transcriptional regulation, for instance, to determine whether these factors could interact together in order to regulate CFTR transcription, we started several experiments such as co-immnoprecipitation, multiple co-transfection and RNAi.

Taken together, these data evidence that the variant -33A in CFTR promoter should be considered as an important risk factor in bronchiectasis pathogenesis. Furthermore, we have identified a novel regulatory complex on the minimal CFTR promoter, which will enlighten the understanding of the transcriptional regulation of the CFTR gene. A better knowledge of CFTR cis-and trans-acting elements will allow to consider new approaches to modulate and/or control more specifically CFTR expression.

This work is supported in part by the association Vaincre La Mucoviscidose. (Nat Genet, 1999) . A recent study reports strong genetic influences for MI (Blackman et al., Gastroenterology, 2006) , but it did not replicate the modifier locus on chr. 19q13. Inconsistency of results may reflect the variability of data reporting and different classifications of MI (i.e. surgically or medically treated). We tested the accuracy of reporting of MI on Case Report Forms (CRFs), as compared to primary source documents. Methods: The CRF for the GMS has a checklist for past medical history. For example, we requested "Yes" or "No" for MI (an obstruction of the terminal ileum at birth), but did not require source documents or identification of the type of treatment. To evaluate CRF reporting of MI, we requested source documents for patients with reported MI, including a surgical or medical treatment report. If a written report from the time of birth was unavailable, a clinic note, detailing MI at birth, treatment, and evidence of a surgical, abdominal scar (if applicable), was required. Verbal confirmation by the patient and evidence of a surgical scar was also accepted, if no written documentation was available. Results: On CRFs, 233 of 1371 patients (17%) were initially reported to have MI. To date, source documentation has been obtained for 140 patients, and 112 of those (80%) have been confirmed to have MI (92% by written report, 8% by verbal report). Of the 112 with confirmed MI, 92 (82%) had surgery, and 20 (18%) had medical treatment. There were 10 false reports of MI (7.1%). MI could not be confirmed or refuted for 18 patients (12.9%) because of insufficient information (n=16) or confounding circumstances (n=2). To date, documentation for 243 patients who were reported to have no MI has been obtained, and no false negatives have been found. Additional documentation for MI from other sites is expected. Conclusion: At least 7%, and perhaps as many as 18%, of the reports of MI on CRFs were inaccurate. Gene modifier studies must include rigorous documentation of MI to ensure an accurate correlation between phenotype and genotype. Studies should also characterize different classifications (i.e. surgical vs. medical treatment) to assist the detection of gene modifiers of MI.

Reporting for the Gene Modifier Study group (MRK); Supported by CFF KNOWLE00A0, CFF DRUMM00A0, NIH 5R01 DK66368, NIH 5R01 HL68890, and GCRC RR00046. Any mutation that disrupts or diminishes the efficiency of the splicing process will have an impact on disease manifestation. In the Cystic Fibrosis (CF) transmembrane conductance regulator (CFTR) gene more than 1,500 mutations were identified, most of them being disease-causing [1] . Among these,~42% are classified as missense and about 13% are classified as splicing mutations, given that they disrupt the consensus splice sites [1] . However, the splicing mutation concept is evolving, as it nowadays also includes mutations other than just those within or close to the consensus splice sites, rendering prediction of mutation consequences a hard task. Nevertheless, it is still very important for the clinical settings to determine the functional effect of gene mutations.

Our aim here was to study the effect of I1234V, a rare CFTR missense mutation in NBD2 (exon 19), directly in native tissues of a CF patient bearing F508del in the other allele, so as to gain insight on how it influence the disease outcome.

To look for a possible effect at the RNA level, total RNA was extracted from native nasal cells and colonic tissue, and cDNA producd using random primers. RT-PCR amplification was performed in the region spanning exons 18-20, being one of the primers fluorescently labelled. The products were analysed as described before [2] . The I1234V(3832A>G) mutation which creates both a novel acceptor and a novel donor, was found here to cause alternative spliced CFTR transcripts lacking the last 18 nucleotides of exon 19, thus showing that only the novel donor is used in vivo. Moreover, our data show that no normal (only alternatively spliced) transcripts result from this allele. We have also analysed the I1234V mutation at the protein level by producing a stable BHK cell line expressing the I1234V-CFTR, after generating the respective mutant cDNA construct by site-direct mutagenesis. Protein expression and function was determined by immunoblot and iodide efflux assay, respectively. Results show that I1234V-CFTR protein is processed and functional. However, given the above-described absence of normally spliced mRNA coding for this CFTR variant, we have to conclude that this protein is not produced in vivo. Indeed, since only the mRNA coding for CFTR lacking the last six aminoacids of exon 19 was detected, we are currently characterizing in vitro the proprerties of this truncated protein.

Altogether, our data clearly demonstrate that the functional effect of this mutation is not due to the amino acid change but to abnormal splicing.

In conclusion, characterization of the consequences of mutations in native affected tissues is important, not just because this provides unexpected information about the mechanisms underlying the basic defect but also for disease diagnosis and prognosis. (Darrah et al, NACFC 2006) . The purpose of this study was to examine the influence of common genetic variation in the endothelin pathway on the CF phenotype.

Methods: Patients were recruited from the CF clinics in Dublin, Belfast and Seattle. Serial clinical data were abstracted from medical charts and local clinical databases. Studentized residuals of maximum FEV1, after adjusting for age, gender and height, was the phenotype of interest. Twentyone maximally informative tagSNPs were identified in the EDN1, EDN3, EDNRA and EDNRB genes using the NIEHS Environmental Genome Project and were genotyped using the Illumina BeadArray system. Genotype and haplotype analysis were carried out using HelixTree genetic analysis software.

Results: Clinical and genetic data were available on 544 CF patients (239 from Seattle and 305 from Dublin/Belfast). In the combined cohorts, tagSNPs in the EDNRA gene were significantly associated with differences in CF lung disease severity (p=0.000026). This was observed in both the Irish cohort (p=0.0049) and the Seattle cohort (p=0.00507). The effects were independent of gender and CFTR genotype. Four common haplotypes (haplotype frequency>5%) were identified in the EDNRA gene. There was significant association between EDNRA haplotypes and CF lung disease severity (Table 1 ). There was no association between genetic variants in EDNRB, EDN1 and EDN3 and CF lung disease severity.

Conclusions: The EDNRA gene is a genetic modifier of the CF phenotype. Our findings were seen in two independent cohorts and verify existing associations found in other populations. The endothelin pathway may be a novel therapeutic target for the treatment of CF lung disease. Background: Cystic Fibrosis (CF) is a recessive "monogenetic" disorder, but there is heterogeneity of lung disease severity and survival reflecting environment and non-CFTR genetic modifiers. The UNC/CWRU multisite Gene Modifier Study (GMS) identified patients who were "severe" or "mild" as teenagers or young adults and pertinent cross-sectional data was collected; however, we did not collect all pertinent information about early (age <8 years) clinical features. Objective: We sought to retrospectively evaluate the early clinical features of "severe" and "young mild" patients enrolled in the GMS. Methods: We obtained all CFF Registry data available on 596 patients. There were 303 "severe" (worst 25 th percentile of birth cohort, age range 8-25) and 293 "mild" (best 25 th percentile, age range 15-28). Initial analyses focused on cross-sectional plots of patient age versus multiple clinical features, including age at diagnosis, hospitalizations, CDC height and weight percentiles, presence/absence of Ps. aeruginosa from respiratory cultures, and FEV 1 (% Pred). Results: There were 15.5 and 17.6 years of CFF Registry data per patient for "severes" and "milds," respectively. Preliminary results indicate "severe" patients were diagnosed earlier in life than "mild" patients (mean: 1.2 vs. 1.9 years, p=0.004). This difference is greater after omitting patients who had meconium ileus (diagnosis: 1.4 vs 2.3 years, "severes" and "milds" respectively, p=0.003). Between 1 and 7 years of age, "severe" patients were hospitalized more frequently than "mild" patients and, from age 8 onwards, the disparity in frequency of hospitalization increased. "Severes" and "milds" had similar CDC height percentiles (~35 th percentile) until age 8, at which point percentiles increased more for "mild" versus "severe" patients. By age 3, "severe" patients already had lower CDC weight percentiles than "milds" (32 nd versus 41 st percentile) and this disparity increased throughout adolescence. From age 1 to 8, "severe" patients had a 2-3 fold higher prevalence of Ps. aeruginosa than "mild" patients. As early as 6 years of age, FEV 1 (% Pred) was ~30 points lower in "severes" as compared to "milds." However, some "severe" patients had normal lung function at ages 6-8 and overlapped with mild patients; thus phenotyping of lung severity for young patients is not optimal. All of these results were similar for males and females. Summary: Retrospective analysis of the CFF Registry data indicates that patients classified later in life as being "severe" experience a worse course of disease from early in life. Understanding the early clinical course may prove helpful in defining surrogate phenotypes for modifier studies, and help define appropriate therapeutic targets. Cystic fibrosis is mainly caused by small molecular defects of the CFTR gene; despite the genotype is defined in the majority of patients, a number of CF cases still remain uncharacterised. The CF mutation database lists more than 35 large rearrangements that may escape detection using PCRbase techniques. The Innogenetics assay, the DHPLC and sequencing screening showed a mutation detection rate of 92.6% in our population. We report here the results of MLPA screening for CTFR gene rearrangements, performed on the unidentified alleles of our CF patients.

Our sample population consists of 691 unrelated Italian CF patients (for a total of 1378 alleles), followed at CF Centres of Lombardia Region. MLPA analysis was performed in 40 patients who still had one or two unidentified alleles after extensive analysis of CFTR gene. All patients studied had classical clinical CF symptoms. Subjects presented with persistent or recurrent respiratory symptoms, failure to thrive, salt loss syndrome and gastro-intestinal findings.

We characterized 10 different deletions and a new duplication (dup Promoter-ex.3). Thus, 26.2% (21/80) tested alleles had a large gene rearrangement. The deletion of exons 22-23 (6/80) was the most frequent in our cohort. All patients had positive sweat chloride values (above 60mEq/L), except the patient carrying duplication who has borderline sweat chloride value.

Out of 24 patients, 6 (25%) had fecal elastase levels consistent with a preserved pancreatic function: of these patients, 3 had mild mutation, 1 had severe, and 2 had unknown mutation. Six patients present liver involvement.

The results of the present study could indicate that compound heterozygosity for large rearrangements in CFTR gene, is strongly associated with severe pancreatic disease as mutations in classes I, II, and/or III. L997F is a missense substitution which changes from leucine to phenylalanine at position 997, resulting from a G/C transition at position 3123 in exon 17a of the CFTR gene.

It has been described in patients with disseminated bronchiectasis, recurrent idiopathic pancreatitis, sarcoidosis, newborns with hypertripsinemia,. Recently Derichs et al. reported one healthy 3-year-old girl homozygous for L997F; therefore the pathogenic role of the variant is still unclear.

In this study we present 6 subjects compound heterozygotes with L997F. No other mutations have been identified after molecular analysis performed using sequencing analysis of the whole coding region of CFTR gene and MLPA technique in order to search for gene rearrangements.None of them presented IVS8 5T allele.

Three of them had a severe mutation in trans (R553X, 2183AA>G, and N1303K), while the other three had a mild mutation on the complementary allele (D1152H, R334Q, and R334W). Individuals with a severe mutation in trans presented a remarkably different clinical picture compared to those with a mild mutation in trans. The formers were diagnosed at a mean age of 2.8 years, had an average sweat chloride of 55.9 mEq/l; besides, all of them have respiratory symptoms, Staphilococcus aureus in sputum cultures (one had Pseudomonas aeruginosa as well), and in two the chest x-rays was abnormal. The three L997F/PS mutation individuals were diagnosed at a mean age of 20.1 years, had an average sweat chloride of 39.1 mEq/l; besides none of them had respiratory symptoms, abnormal chest-Xrays, or positive sputum cultures, but two had a history of pancreatitis.

These data seem to suggest that L997F cannot in any case be considered a neutral polymorphism. The variability of its clinical expression seems to be influenced by the mutation in trans.

Further studies are needed in order to support our results. Expression of the cystic fibrosis transmembrane conductance regulator (CFTR) is tightly regulated both spatially and temporally, yet the molecular mechanism of this regulation is not well understood. Because no tissue-specific regulatory elements were recognized in the basal CFTR promoter, the crucial cis-regulatory elements are likely to be located elsewhere within the CFTR locus. A number of the non-coding regions of the CFTR gene were found by our laboratory to contain DNase I hypersensitive sites (DHS), suggesting the presence of regulatory elements at these sites. Studies presented here investigate the role of an intron 1 DHS (DHS1) in the tissue-specific regulation of CFTR gene transcription. The elucidation of molecular mechanisms underlying the temporal and spatial expression of CFTR may aid in developing more specific targeted therapies for cystic fibrosis (CF).

Footprinting analysis of CFTR intron 1 revealed a protected region within the core of DHS1 at 185 + 10 kb. In silico analysis of this sequence uncovered a number of transcription factors binding motifs, including a consensus binding sequence for hepatocyte nuclear factor 1 (HNF1). We have previously identified a role for this factor in the regulation of CFTR expression. Results of electrophoretic mobility shift assays (EMSA) demonstrate that HNF1α specifically binds to the motif in CFTR intron 1 in vitro. In addition, chromatin immunoprecipitation (ChIP) analysis of cells expressing both CFTR and HNF1α factor shows that this factor binds to the intron 1 site in vivo. When cloned as an enhancer, the DHS1 element was found to augment minimal CFTR promoter activity in a luciferase reporter based assay. This increase in luciferase activity was abolished when two nucleotides within the core HNF1 binding site were mutated, suggesting a functional role for HNF1α in CFTR gene transcription. Further experiments are underway to determine whether additional transcription factors can be recruited to the core of the intron 1 DHS regulatory element and can interact with HNF1 and the CFTR basal promoter to modulate tissue-specific CFTR gene expression. The promoter of the CFTR (cystic fibrosis transmembrane conductance regulator) gene is not solely responsible for its complex pattern of expression. To identify potential regulatory elements for CFTR we previously mapped DNase I hypersensitive sites (DHS) across 400 kb spanning the locus. In addition to intronic DHS, a number of sites were observed that flank the CFTR gene. We hypothesized that these may include insulator elements that establish the chromatin expression domain, within which the CFTR gene is regulated. Insulators are elements that shield against the effects of regulatory elements from adjacent genes, and they may also block silencing of integrated transgenes.

Using a well-established insulator assay, in which DNA regions of interest are assayed for their ability to interfere with communication between a chicken β-globin enhancer and a neomycin resistance gene, we demonstrated that two DHS, at -20.9 kb from the CFTR translational start site, and at +15.6 kb from the 3' end of the gene, exhibit enhancer blocking activity comparable to known insulator elements. Electrophoretic Mobility Shift Assays (EMSA), demonstrated in vitro binding of the well-characterized insulator protein CTCF (CCCTC-binding factor) at the -20.9 kb DHS. This was confirmed in vivo in both CFTR-expressing and non-expressing cell types using Chromatin Immunoprecipitation (ChIP). In contrast, although the +15.6 kb DHS did not bind CTCF, we obtained in vitro evidence for the interaction of other factors that may be involved in insulator activity. Furthermore, ChIP analysis of histone modifications across the CFTR locus revealed striking differences between the -20.9 kb and +15.6 kb DHS, again suggesting mechanistic differences between these elements.

The characterization of insulator elements flanking the CFTR locus may be of direct practical relevance in the design of vectors for effective CF gene therapy. One of the problems encountered in gene therapy protocols is the relatively rapid loss of expression from the CFTR cDNA once it is introduced into mammalian cells. Incorporation of the -20.9 kb and +15. Aminoglycosides, particularly tobramycin, are primary antibiotics used to treat the airway infections in cystic fibrosis patients. Lifetime, systemic exposure to these antibiotics is significant and can be associated with significant nephro-and ototoxicity. These toxicities have the potential to significantly reduce the quality of life in the aging adult CF population. We previously reported an incidence of aminoglycoside ototoxicity of greater than 50% with sensitive audiometric studies (i.e. pure tone audiometry and distortion product otoacoustic emissions). These studies indicated considerable variability in ototoxicity in patients with similar systemic aminoglycoside exposure. The literature suggests that genetic variability in mitochondrial DNA can partially explain these differences. Single nucleotide polymorphisms (snps) in the mitochondrial 12SrRNA gene are associated with aminoglycoside ototoxicity. We genotyped 90 unselected adult CF patients for four of the mitochondrial 12S-rRNA polymorphisms associated with aminoglycoside ototoxicity by direct sequencing of DNA harvested from peripheral blood. Four patients exhibited polymorphisms in this gene. Two patients possessed the A1555G transition, while another patient each revealed a polymorphism at A827G and C1494T transitions. The patient with the A827G polymorphism died prior to audiometric testing. Both patients with the A1555G had audiometric studies consistent with moderate to severe ototoxicity. while the patient with the C1494T had mild aminoglycoside ototoxicity. All four patients had severe airway obstruction and at least one full course of parenteral tobramycin at 8mg/kg/d. There was no history of toxic serum levels aminoglycosides during this therapy. Both patients with the A1555G polymorphism had a family history consistent with the expected maternal inheritance pattern associated with the mitochondrial 12S-rRNA gene. We found no evidence of the delT961 polymorphism in this population. These studies indicate a higher than expected frequency (4%) of mitochondrial 12S rRNA polymorphisms associated with aminoglycoside ototoxicity. These studies demonstrate that genetic screen-ing provides valuable susceptibility information and may change clinical decision-making. Support: Rising Hope Foundation/National CF Foundation. Cystic fibrosis is a genetic autosomal recessive disease that is caused by deleterious mutations in the CFTR gene. In its most severe form, CF results in abnormal sweat electrolytes, sino-pulmonary disease, male infertility, and pancreatic exocrine insufficiency. In fact, CF is the most frequent cause of pancreatic insufficiency in humans. Carriers of CF (i.e., heterozygotes) do not express any of these classic symptoms. Recent studies indicate that significant numbers of non-CF patients diagnosed with congenital bilateral absence of the vas deferens (CBAVD) or idiopathic chronic pancreatitis (ICP) are compound heterozygotes and carry two CFTR mutations of varying severity. In dogs, especially German Shepherds and rough-coated Collies, exocrine pancreatic insufficiency (EPI) has been observed with symptoms similar to those seen in human patients. EPI in dogs is most often caused by pancreatic acinar atrophy (PAA), a degenerative disease of the exocrine pancreas that has been shown to be inherited as an autosomal recessive trait. The locus responsible for canine PAA has not been identified to date. We considered the possibility that CFTR mutations might be responsible for this disease. A dog that had been diagnosed with PAA and one that was a known carrier of PAA were screened for CFTR mutations. Our samples also included dogs diagnosed with idiopathic pancreatitis as well as healthy dogs. We have established protocols for detecting putative mutations in the canine CFTR gene using temporal temperature gradient gel electrophoresis (TTGE). In the dog with PAA and in the carrier of PAA, our screening methodology identified mutations in 8 of the 27 amplicons that span the 27 exons of the CFTR gene. DNA sequencing revealed silent mutations in exons 17, 26 and 27. In exons 17 (L888) and 26 (L1402), single nucleotide polymorphisms converted CTT codons into CTC codons, both leucine codons. In exon 27 (P1447) another single nucleotide polymorphism converted a CCC proline codon into a CCT proline codon. The other identified mutations (amplicons 15, 16 and 17, 20, 22, 24, 25, 26 and 27) were found in intronic sequences and have no predicted effect on CFTR expression or function. Based on these findings, we conclude that CFTR mutations are not responsible for PAA and the EPI that it causes in dogs. Hypothesis: Specific polymorphisms in genes of the innate immunity may modify the severity and progression of CF pulmonary disease and influence Pseudomonas aeruginosa colonisation.

Methods: Single Nucleotide Polymorphisms (SNPs) analysis of multiple genes contributing to the innate immunity ( (MBL2), MASP (MBL associated serine Protease) 1Ç3, FCN (Ficolin) 1Ç2, LBP (Lipopolysaccharide -binding Protein), CD14, TLR (Toll-like receptor 1Ç10) ) was performed in 116 CF patients (age 6-44 years).

Spirometric data and bacterial colonisation status with Pseudomonas aeruginosa were collected retrospectively. A decline in FEV1 of 14 % over a 6 years period and a percent predicted FEV1 value of 70 % were used to discriminate mild from more severe affected adult CF patients (≥ 16 years).

The frequency of single and combined SNPs in well-defined subgroups of the CF population was compared.

Results: In adult CF patients with a mean FEV1 lower than 70%, the combinations of SNPs of CD14 (promoter) with TLR 10 (promoter), FCN 1 with TLR 5 (exon 6), and MASP 3 with TLR 9 (promoter) were more common than in adult CF patients with mean FEV1 > 70 % (odds ratio (OR) 6,2 (95% confidential interval (CI) : 1,7-22,2 ) , OR 6, 2 (95 % CI: 1, [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] 2) and OR 5,6 (95% CI: 1,7-18,4) respectively).

The frequency of combination SNPs of MASP2 with TLR2 and MASP2 with TLR4 was significantly higher in CF adults with more than 14 % decline in FEV1 compared to CF adults with less than 14 % decline in FEV1 (OR 6,5 (95% CI: 1,6-25,8) and OR 5,3 (95% CI: 1,5-18,6) respectively). However, the single SNPs of MASP2, TLR2, TLR4 were not more frequently found.

In Pseudomonas aeruginosa colonised CF patients, an increased frequency of combined SNPs of TLR1 (exon 4) with TLR6 (exon 1), LBP (exon 10) with TLR6 (exon 1) and MBL2 with TLR 6 (exon 1) (OR 3,6 (95% CI: 1,3-10,1), OR 3,6 (95% CI: 1,3-10,1), OR 3,4 (95% CI: 1,1-10,2) respectively) was found. In contrast, for single SNP TLR6 (exon 1) no statistically significant differences were found between the groups (OR : 2,0 ( 95% CI: 0,9-4,6)).

In general, the odds ratios for the combined SNPs were higher than for the single SNPs and remained significant when the different groups were subdivided according to CFTR genotype.

Conclusion: Certain combinations of SNPs of genes of the innate immunity are more frequently found in CF patients with a lower FEV1, a stronger decline in pulmonary function and Pseudomonas aeruginosa colonisation. However, for the single SNPs this trend was less pronounced or absent.

Congenital bilateral absence of the vas deferens (CBAVD) is a rare condition associated with mutations in CFTR. CBAVD lacks extra-genital organ system abnormalities; it has been viewed as a mild (variant) form of CF. The disease serves as an excellent model for studying the clinical and experimental ramifications of marginal CFTR activity, and the functional thresholds necessary to confer a CF-related phenotype. Recently, Drumm et al. (NEJM, 2005) found that the codon 10 CC genotype of TGFβ-1 is associated with more severe pulmonary disease, particularly in individuals homozygous for ∆F508 CFTR. TGFβ is also known to play a critical role during normal development and differentiation of the human vas deferens (Han-nema et al., Horm Res, 2007) . Because 70% of CBAVD subjects carry at least one defective CFTR allele (but millions of males worldwide are heterozygous for CFTR mutations without developing CBAVD), the present study was designed to test whether TGFβ polymorphism might serve as a genetic modifier underlying isolated vas deferens loss in the setting of CBAVD. We sequenced the 5' end of the TGFβ gene in eighty CBAVD individuals possessing at least one CFTR mutation, and compared genotype frequency with results previously published from demographically similar controls. When CC frequency in CBAVD subjects was compared to pooled estimates from earlier trials, no significant differences were seen using a non-dominant model. Compared to frequencies reported by Arkwright et al. and Garrote et al., a greater number of homozygous CC subjects (23.8% vs. 11.4%, p<0.005) was observed. Our preliminary results therefore indicate that if TGFβ-1 codon 10 polymorphism is associated with CBAVD penetrance, it does not represent the major modifier of disease phenotype. In order to evaluate this possibility further, we plan analysis of a larger (matched) control set of CBAVD and non-CBAVD individuals, studies of TGFβ-1 activity in vas deferens epithelial cells with and without CFTR, and experiments to investigate the effects of TGFβ on CFTR biogenesis and function. Supported by NIH and CFF.

Newborn screening (NBS) for CF has been successful in identifying CF infants, particularly those who have positive sweat tests immediately following the positive screening result. Tracking by personnel at the central NBS program facilitates resolution of diagnostic status because CF NBS presents a challenge to CF Center (CFC) systems and primary care providers when apparently asymptomatic infants have inadequate (quantity not sufficient (QNS)), borderline or indeterminate sweat tests. Short-term follow up is more problematic when the screening algorithm requests earlier sweat tests and defines a lower cutoff for indeterminate sweat test results than is used for older infants, children and adults. The Massachusetts (MA) CF NBS Workgroup chose a low cutoff for indeterminate results of 30 mEq/L up to 6 months of age, reverting to the accepted 40 mEq/L cutoff thereafter. Follow up by the central NBS program identified infants whose diagnostic status remained unresolved more than 6 weeks after a QNS or borderline sweat test and eliminated the need for individual CFCs to follow infants tested at more than one CFC. The Program sent informational letters and a reminder recommendation to finalize the diagnosis to primary care providers of infants with unresolved diagnoses. Of the 189 CF infants identified by MA CF NBS, 25 had an initial sweat result that was indeterminate and another 20 had an initial result that was QNS. The central screening program has the capacity to track infant and child outcomes across CFCs; for example, we have been tracking growth outcomes of all CF infants identified by MA CF NBS, despite 31 of the 189 CF infants having moved from one CFC to another during their first 5 years of life. Tracking and outcomes data will be presented. When more than one CFC operates within a screening program, centralized tracking ensures capture of quality data. Despite several educational campaigns CF is often not diagnosed in early infancy in Poland (the mean age of diagnosis 42 months, median is 12 months). After several years of discussions,in September 2006 a neonatal screening for CF supported by the Ministery of Health was started as the third obligatory neonatal screening in Poland. More than 124,000 neonates were screened in the first seven months using the existing screening infrastructure and spare blood spots after PKU and congenital hypothyreosis screening. The protocol was two stage IRT/DNA (covering the 47 most common mutations in the Polish population). The final diagnosis was based on sweat tests (conventional pilocarpine-iontophoresis and conductometric nanoduct) as well as clinical examination. Values of Chloride in sweat >40 mmol/l and NaCl >80 mmol/l were considered diagnostic for CF. 12 children were confirmed as CF. Although delta F508 is the most common mutation in the Polish population, only two infants were delta F508 homozygotes, six were delta F508 heterozygotes and the remaining 8 had different mutations (3 children had 3849+10kbC>). Half of the diagnosed group was still pancreatic sufficient. The children had confirmed CF diagnosis at the mean age of 6 weeks. Only 3 children were free from any radiological changes in lungs. The most common pathogen in this group was Staphylococcus aureus. At the first visit information for parents about CF was given along with physiotherapy and dietary education. Anthropometric measurements, airways bacteriology and elastase -1 in stool were performed. Psychologist consultations were available. Neonatal screening for CF enabled earlier diagnosis and the introduction of complex therapy in comparision with symptomatic screening. Objectives: The clinical course of cystic fibrosis (CF) varies widely among patients carrying the same CFTR gene mutations supporting that additional genetic modifiers could affect the CF phenotype. As inflammation is a central contributor to the pathogenesis of CF lung disease, genes involved in the inflammatory response are potential modifier candidate genes.

Methods: We examined the influence of 13 polymorphisms in 7 genes involved in the inflammatory response (TNF, IL-1β, IL-1 receptor antagonist (RN), IL-6, IL-8, IL-10 and TGF-β1), on disease progression in a group of 329 Caucasian children with CF. The genotypes were tested for an association with changes in lung function tests, Pseudomonas aeruginosa colonization and nutritional status by multivariable analysis.

Results: We found a significant association between TGF-β1 +869T/C variants and decline in lung function measured as the forced expiratory volume in one second (FEV1) and the forced vital capacity (FVC) (p=0.004 and 0.02 respectively). IL-8 gene polymorphisms (-251A/T, +396G/T and +781T/C) were associated with the occurrence of Pseudomonas aeruginosa colonization (P<0.02).

Conclusions: This study suggest that IL-8 variants may influence Pseudomonas aeruginosa colonization in CF, however this need further confirmation. Moreover, we provide additional support to the results of other trials and strongly suggest an association between TGF-β1 variants and lung disease phenotype in CF. If such a role of TGF-β1 is confirmed by functional studies, it may have important clinical impact for the identification of patients at risk to develop more severe respiratory manifestations and for the development of new therapeutic strategies aimed at adequately balancing TGF-β1 production and action Introduction: Cystic fibrosis (CF) is the most common monogenic disease in Caucasian population, its frequency is about one in 2500 live born. The estimated frequency in hispanic population is about one in 8500, with differences in mutation frequency especially for low frequent alleles. Heterogeneity in pulmonary manifestations cannot be explained by the CFTR mutation genotype. Recent studies are focused in the study of modifier genes. These are genes different from the primary CFTR gene that could influence the FQ phenotype, acting trough mechanisms like inflammation, repair and remodeling, protease-anti protease balance, innate immune response, etc. This work shows some advances in CF diagnosis and modifier genes analysis in Mexican population. Methods: We tested the effectiveness of INNOLiPA CFTR36 kit (Innogenetics) in 23 clinically diagnosed CF patients. Thirty six frequent mutations in CFTR gene were analyzed by PCR and reverse hybridization with Allele Specific Oligonucleotides probes in DNA samples extracted from blood or oral swabs. For frequency analysis of modifier genes we tested about seventy DNA samples from a DNA bank of patients previously tested positive for CFTR mutation. The polymorphism analysis of modifier genes was made by PCR-RFLP. For Alpha 1 Antitrypsin (AAT), Z and S alleles were analyzed according Stanford and cols. (1999) . For Tumor Necrosis Factor Alpha (TNFα) -308 polymorphism in the promoter region was analyzed according Chen and cols. (2006) . Results: Mutation was detected in in 20 of 23 patients (87% of effectiveness). DF508 allele frequency was 52% (Caucasian 60-70%) followed by 2789 +5G-A, S549N, G542X, G85E and 1078delT (20%). 28% of the mutant alleles remain undetected. Preliminary results in modifier genes show low frequency for AAT mutant alleles (1.4% for AAT S and Z alelle respectively) and 8.7% for TNF-308. Conclusions: DF508 alelle frequency was 52%, and we found the low fequency alleles S549N and 2789+5G-A. Frequency of modifier genes mutant alleles in CF patients are low. This work is in progress of selection and recruitment of patients and controls, as well as standardization of the other genes modifiers, along with some other biomarkers that will allow to understand the physiopathology of CF and to define a possible genetic risk profile of severe pulmonary disease. Over 1300 putative Cystic Fibrosis (CF)-causing mutations have been reported to the CF Mutation Database and almost half (~630) are rare missense mutations that are predicted to substitute a single amino acid. To determine if any of these rare missense mutations cause CF by altering localiza-tion of CFTR, we examined mutations in the cytosolic loops of CFTR since localization signals have been identified in cytosolic loops of other ion channels. Amino acids alignments comparing human CFTR sequence to that of 36 species revealed that cytosolic loop 4 (CL4) was the most conserved loop by identity. Furthermore, CL4 had the most naturally observed mutations: 33 of the 68 amino acids were found to be mutated in CF patients and 10 sites had multiple missense mutations at the same amino acid. In addition, different substitutions of the same CL4 amino acid have been associated with different disease consequences, such as the missense mutations in the arginine at codon 1070. Patients with the R1070P mutation have pancreatic insufficient (PI)-CF, those with R1070W mutation have pancreatic sufficient (PS)-CF or CBAVD, while patients with R1070Q primarily had PI-CF.

To determine whether the R1070 mutations cause disease by affecting CFTR localization, we used the FLP-In system to create stable polarized MDCK-GFP-CFTR cells to produce isogenic clones with CFTR expressed from the same integration site. Confocal microscopy studies of stable MDCK-FRT-GFP-CFTR-R1070 mutant cell lines revealed that CFTR-R1070P was cytoplasmic, CFTR-R1070W was apical and cytoplasmic, and CFTR-R1070Q was apical. To confirm these observations, quantitative biotinylation studies were performed. Biotinylation assays of MDCK cells stably expressing CFTR-R1070P confirmed that this mutant was absent from the apical surface, CFTR-R1070W had a low level of fully glycosylated protein at the apical membrane, while CFTR-R1070Q had fully glycosylated protein at the apical membrane comparable to wild-type CFTR. Thus, CFTR-R0170P and CFTR-R1070W displayed properties in polarized cells that were distinctly different from wild-type CFTR consistent with their associated CF phenotypes. However, the profile of CFTR-R1070Q (apical localization and Cl-channel function (Seibert et al. 1996 , Mickle et al. 2000 ) appears inconsistent with a PI-CF phenotype. Indeed, re-analysis of CFTR genes bearing R1070Q revealed that 11 of 12 carried a second mutation in cis (S466X). In nine of these patients in whom CFTR genotype and pancreatic status is known, presence of the nonsense mutation S466X is associated with the PI-CF phenotype. Since nonsense mutations are known to cause severe gene dysfunction by promoting decay of their RNA transcript, we concluded that the S466X mutation rather than R1070Q is responsible for the PI-CF phenotype.

The finding of S466X explained the otherwise enigmatic functional studies of R1070Q and clinical observations in patients bearing this mutation. This study also demonstrates that stable expression of CFTR mutants in polarized MDCK cells using the FLP-In system provides a useful screen for evaluating the disease potential of rare missense mutations. The clinical course of CF is characterized by recurrent pulmonary infections and chronic inflammation. In CF patients, up-regulation of toll-like receptor-2 (TLR2) gene in airway epithelial cells is believed to enhance proinflammatory responses towards bacterial TLR2 ligands. We have recently shown that decreased methylation (demethylation) of the TLR2 promoter is responsible for CF-related up-regulation of TLR2 in bronchial epithelial cells (Shuto. T. et al., FASEB J, 2006) . However, the molecular mechanisms responsible for DNA demethylation-dependent TLR2 gene upregulation in CF cells remain unknown. Here, we identified minimum region of the TLR2 promoter critical for its expression at -120/+110, which contains one putative binding site for Sp1. SP1 inhibitor mithramycin A treatment reduced TLR2 promoter activity and its expression in CF bronchial epithelial cells (CFBE41o-), suggesting the importance of this SP1 site for the TLR2 gene regulation. Moreover, bisulfite sequence analysis revealed the hypomethylation of adjacent this SP1 binding motif within TLR2 promoter in CFBE41o-cells, implying the CF-specific demethylation of adjacent SP1 binding motif. Although mithramycin A rarely affected basal expression of TLR2 gene in 16HBE14o-cells (non-CF bronchial epithelial cells), mithramycin A inhibited TLR2 expression in 5-azacytidine (DNA-demethylating agent)-treated 16HBE14o-cells. Taken together, our results suggest that SP1 is crucial for DNA demethylation-induced gene upregulation of TLR2 in CF bronchial epithelial cells. Mucus hypersecretion due to goblet cell metaplasia is a critical feature of CF lung disease, affecting both mucociliary clearance and drug delivery. Various studies in CF lungs and CF primary cultures have shown increased expression of MUC5AC, the main gel-forming mucin produced by goblet cells; however, little is known about the role of MUC5AC in the progression of lung disease. To determine whether overexpression of MUC5AC alone is sufficient to induce lung pathology, we generated a mouse model overexpressing Muc5ac. The entire Muc5ac cDNA (minus 26% of the predicted VNTR sequence, final size ~9Kb) was cloned and tagged with an internal GFP epitope at a site close to the VNTR to preserve the integrity of mucin domains. GFP-Muc5ac was linked to the rat CCSP promoter to drive specific airway expression, primarily in Clara cells since these cells are capable of secreting mucins. Blastocyst injections were used to generate transgenic founder lines in C57Bl/6 background. Although Muc5ac-GFP mRNAs were expressed in transgenic mice, lung histology did not show significant increases in inflammation, AB/PAS-positive cells, or luminal secretion compared to wild-type littermates. GFP fluorescence from freshly excised lung was weak but immunohistochemistry showed that the distribution of GFP correlated with the expected expression pattern, with the majority of positive cells localized to the airways and few positive cells in the alveolar space. At higher magnification, GFP-positive granules were observed in dome-shaped cells that were also positive with CC10 antibody, suggesting proper Muc5ac-GFP packaging in Clara cells. Bronchoalveolar lavage samples separated by agarose gel Western blots and probed with GFP antibody confirmed that each line secreted Muc5ac-GFP, with a migration pattern comparable to a mucin. To further characterize the secreted transgenic mucin, a pool of lavage samples was separated into supernatant and pellet. The pellet was resolubilized in 4M Guanidine. Both elements were passed through a S1000 size-exclusion column. Analysis by light scattering and refractometry to determine molecular weight and concentration revealed that the solubilized pellet contained an enormous but compact complex (920 x 106 g/mol and 320nm radius of gyration). GFP-rich fractions were PAS-positive, indicating proper glycosylation of the transgenic Muc5ac-GFP. CsCl density gradient studies revealed that the GFP-rich fractions had a high molecular weight but did not appear to be a prominent part of the megacomplex retrieved in lavages, unlike other tested mucins. In conclusion, we showed that the overexpressed Muc5ac-GFP was a high-molecular-weight protein that was glycosylated, packaged in Clara cells and formed multimers. Lack of discernable phenotype in the transgenic lungs suggests that mucin overproduction alone is not sufficient to trigger goblet cell metaplasia and mucus accumulation, supporting the protective role of mucins. Currently we are crossing the Muc5ac-GFP with the Scnn1b mouse model to study Muc5ac-GFP under dehydrated conditions and we are conducting allergen-challenges in the transgenic mice. Results of such studies will provide novel insights into the role of secreted Muc5ac in development of lung disease in these models. Supported by grants from the CF Foundation. The major characteristic in cystic fibrosis (CF) is acquirement of chronic lung infections with P. aeruginosa. Once established the chronic infection can only be suppressed, not eliminated. The CF patients harbour one single strain of P. aeruginosa which only rarely is replaced by other strains. During the chronic lung infection there is a year long mutual impact of the bactaria on the host, and of the host response and antibiotics on the bacteria, resulting in a continous adaptation of the bactaria in the lungs of the CF patients with selection for different clones. In contrast, mouse models of the lung infection in CF is limited to two weeks! Therefore, we introduced a new strategy infecting different groups of BALB/c mice with pulsed field gel electrophoresis-(PFGE) identical clones isolated from one CF patient during 23-years of P. aeruginosa lung infection.

1) Infecting the mice with non-mucoid isolate showed that the early isolates were significantly better in establishing an infection with reduced clearance of the bacteria from the lungs (p<0.04), increased dissimination of the macroscopic pathology (p<0.002) and a more acute type and a higher degree of inflammation (p<0.05) as compared to mice infected with a late isolate. In addition, the pulmonary cytokine response was comparable with observations in CF where GM-CSF and IL-10 correlated to a milder disease, whereas G-CSF, MIP-2 (IL-8 analog), and IL-1b correlated to severe disease.

2) Infecting the mice with the mucoid PFGE-identical isolates from the same CF patient revealed a signicantly higher mortality in mice infected with the late isolates (p<0.005). In addition, the pulmonary G-CSF and MIP-2 response increased in the groups of mice that were infected with the late isolates as compared to the early isolate (p<0.05). Moreover, the G-CSF and MIP-2 increased during the first three days of infection with the late mucoid isolates (p<0.05), indicating an impairment of the host response in controlling the lung infection.

3) When tobramycin treatment was initiated 1h after infection of mice with a mucoid isolate, the number of bacteria were cleared or reduced to a significantly lower level (p<0.05), and a significant decrease of both G-CSF and MIP-2 at day one, two and three was observed (p<0.03). In contrast, when treatment was initiated after 24h the number of bacteria did not change significantly, and the reduction in inflammatory cytokine response became significant for both late isolates only at day three of treatment (p<0.03). Moreover, cytokine levels were significantly lower in the 1h as compared to the 24h group (p<0.04) confirming the importance of early and aggressive antibiotic treatment in CF. Non-mucoid isolates were reduced during the first three days of infection independent of treatment.

In conclusion, we have implemented a clinical important CF-time perspective (years) in experimental P. aeruginosa lung infection, useful for patho-physiological and treatment studies. Cystic Fibrosis (CF) is a complex disease that affects multiple organs and results in a wide range of phenotypes in humans. Mouse models of CF display many of the phenotypes observed in human patients and provide necessary tools to dissect this multifaceted disease. However, due to the involvement of Cftr in various organ systems, the ability to discern the contribution of each organ system to each phenotype of CF is unknown. For example, we have found that mouse growth is dramatically reduced compared to wildtype, and mechanisms ranging from neuroendocrine disruption, intestinal malabsorption and cacchexia have been proposed. We have found that CF mice consume more calories per gram body weight than non-CF mice and intestinal absorption of dietary lipids is similar between the two. CF animals absorb 94% and non-CF 97% (p > 0.5). To identify the source of the growth deficit, it would be advantageous to restrict expression of Cftr in specific tissues to determine each tissue's contribution to the phenotype. To accomplish this, we have created a conditional, null murine Cftr allele. A conditional null allele is one in which the gene of interest is induced to be non-functional in either a specific tissue or cell type, or at a specific developmental time. Examination of Cftr absence on a tissue-by-tissue basis will greatly facilitate our understanding of the disease. The conditional Cftr allele was created by "floxing" exon 10 of Cftr, known to be critical for function, with flanking LoxP sites that allow for the deletion of exon 10 when bacterial recombinase Cre is present. To date, mice stably transmitting the floxed allele have been created. These founder mice have been crossed with mice expressing Cre from the promoter of the villin gene, which restricts transcription of Cre to intestinal epithelium. First generation animals show appropriate recombination and deletion of exon 10 in the gut. However, these first generation animals are heterozygous for the floxed allele. We are currently backcrossing these animals with the founders to generate animals with homozygous deletion of Cftr in the intestinal epithelium. Similar crosses are planned to target other tissues. This new animal model of CF will be available to the CF research community to further our understanding of CF pathophysiology. This work was supported by a grant from the CF Foundation Increased ENaC-mediated Na+ absorption is a hallmark of cystic fibrosis (CF) airways. We demonstrated that mimicking Na+ hyperabsorption by overexpression of βENaC in mouse airways results in airway surface liquid (ASL) depletion and reduced mucus clearance causing a spontaneous CFlike lung disease with high pulmonary mortality, and airway mucus obstruction, mucous cell metaplasia and chronic airway inflammation in adult survivors (Mall et al., Nature Med.10:487, 2004) . The aim of this study was to identify the initiating lesions and investigate the natural history of lung disease caused by ASL depletion in βENaC overexpressing (βENaC-Tg) mice. To achieve this goal, we performed longitudinal studies on lung morphology, airway mucus obstruction, bronchoalveolar lavage inflammatory cell counts, expression levels of mucins (Muc5ac, Muc5b, Muc4) and proinflammatory cytokines (TNFα, KC, IFNγ, IL-13, Eotaxin-1), lung volume, lung mechanics, and airway Na+ transport in neonatal to adult βENaC-Tg mice and wild-type littermate controls. We show that airway mucus obstruction in βENaC-Tg mice originated in the trachea in the absence of mucous cell metaplasia in the first days of life, and was associated with hypoxic degeneration and necrosis of airway epithelium, and death. In surviving βENaC-Tg mice, mucus obstruction extended into the lungs and was accompanied by secondary mucous cell metaplasia, increased expression of Muc5ac, Muc5b and Muc4, and airway inflammation with transient increases in macrophages and TNFα, in eosinophils and IL-13, and persistent increases in neutrophils and KC in the lung. βENaC-Tg mice also developed emphysema with increased lung volumes, distal airspace enlargement, and increased lung compliance. Several disease surrogate markers waned in parallel with βENaC transcript levels and airway Na+ absorption in adult βENaC-Tg mice indicating that the level of ENaC overexpression was critical in determining disease severity. In summary, our results provide several novel insights into the in vivo pathogenesis of lung disease caused by ASL depletion. We show that (i) ASL depletion is sufficient to initiate severe airway mucus obstruction in the absence of mucous cell metaplasia and/or mucus hypersecretion; and (ii) airway Na+ hyperabsorption and mucus plugging-induced hypoxia is associated with airway epithelial necrosis constituting a mechanism that initiates airway inflammation in the absence of infection. Further, our results point to a novel link between airway Na+ hyperabsorption, eosinophilic inflammation, and emphysema. The high frequency of allergic airway inflammation and emphysema in CF patients warrants further studies of the mechanisms that link ASL depletion with these associated pathologies. The balance of airway surface liquid (ASL) at a height compatible with efficient clearance of mucus is a process regulated by two main pathways: the secretion of Clvia CFTR, and absorption of Na + via the epithelial Na + channel (ENaC). To study this interplay, an immortal cell line capable of growing at an air liquid interface and having properties of human bronchiolar epithelium would be desirable for its simplicity in preparation and ease of molecular manipulation.

An immortal human bronchial adenocarcinoma cell line (HBAC) was infected with a LCFSN retrovirus carrying either CFTR cDNA (HBAC CFTR ) or the blank vector (HBAC MOCK ). 8-10 day monolayers, grown on transwell clear membranes, were mounted in Ussing chambers and studied under open-circuit conditions. To study regulation of Clsecretion, the equivalent short circuit current was calculated in the presence of amiloride (apical, 100 µM).

In HBAC MOCK cells, forskolin (10 µM, basolateral) had no effect upon I SC . In contrast, HBAC CFTR cells demonstrated a robust increase in I SC ; a rapid increase to a peak (∆~20 µA cm 2 ), followed by relaxation to a stable and sustained plateau. Genistein (50 µM) also stimulated I SC in HBAC CFTR (∆I SC 13 µA cm 2 ) but not HBAC MOCK cells. Amiloride resistant basal current was consistently higher, by 2-3 µA cm 2 , in HBAC CFTR versus HBAC MOCK cells, presumably reflecting low basal CFTR activity. We next examined the ability of endogenous receptors present in HBACs to regulate transfected CFTR. Stimulation of I SC in response to apical adenosine (100 µM) in HBAC CFTR was qualitatively and quantatively similar to the forskolin response. While the above experiments provided evidence to suggest that HBAC CFTR cells possess the bioelectric capability of sustaining anion transport via nucleoside-stimulated CFTR activity, it does not guarantee a resultant vectorial flow of fluid. In the final series of experiments, we examined whether adenosine was indeed able to stimulate secretion across the HBAC-CFTR monolayer by assaying ASL height. Monolayers were loaded apically with FITC-labeled dextran (20 µl), and the majority of dye was then aspirated. Monolayers were then allowed to equilibrate over a 2 hour period, and ASL measured by XZ scanning on a Leica confocal microscope. ~100 µM adenosine and ~10 µM benzamil was then added apically in PFC and ASL height measured after 10 and 30 minutes. In both HBAC MOCK and HBAC-CFTR cells, the basal ASL height was ~4 µm. Over 30 minutes ASL height of HBAC MOCK cells did not change in response to adenosine. In contrast, adenosine stimulated HBAC CFTR cells demonstrated a doubling of ASL height (to 8 µm) over the same period. This response is consistent with the prolonged increase in I SC observed upon stimulation with adenosine.

In summary, the HBAC model which displays only absorptive currents endogenously, can be transformed into a cell line in which CFTR-mediated secretion can be stimulated and sustained by endogenous receptor systems present in the apical membrane. This new model will allow us to study CFTR-mediated fluid transport in a defined manipulatable system. Despite the identification of the CFTR channel more than 15 years ago, many questions remain about its regulation and in vivo function. In order to better understand how CFTR functions in vivo, numerous studies have searched for and identified potential interacting partners but to date, there is little evidence linking any of the known CFTR interactors to its activity.

During a forward genetic screen in zebrafish, aiming to identify genes regulating endodermal organ development, we identified a mutant, which we named baobab (bao), after the African tree that accumulates water during the wet season, which exhibits a dramatically enlarged fluid-filled gut tube. Most interestingly, we have shown that various CFTR inhibitors reduced the appearance of enlarged guts in bao mutant embryos, suggesting that the fluid accumulation phenotype resulted from an increase in CFTR activity. To test whether increased CFTR activity resulted from changes in CFTR protein levels or localization we raised antibodies against zebrafish CFTR. Western-blot and immunolocalization studies revealed that the channel was expressed and localized as in wild type in bao mutant embryos indicating that the increased CFTR function most likely resulted from the activation of the channel. These data, together with the recessive character of the mutation, suggest that the Bao protein is a negative regulator of CFTR activity.

We have positionally cloned the bao locus and isolated the affected gene. bao encodes a cytosolic protein that has not been previously shown to regulate CFTR. In agreement with the pharmacological data pointing to a gut-specific function of bao, in situ hybridization studies showed that bao is higly expressed in gut. We are currently investigating how the Bao protein regulates CFTR function. In addition, we are also investigating the function of CFTR during zebrafish development.

These studies lay the groundwork to take advantage of the zebrafish system to further investigate CFTR biology. We establish a genetic model system for studying the regulation of the CFTR channel that will contribute to the understanding of the pathophysiology of Cystic Fibrosis and several intestinal secretory conditions. Acknowledegements: this work was supported by an EMBO long-term postdoctoral fellowship to MB and by NIH grants to D.Y.R.S..

We thank A. Verkman for providing CFTR inhibitors and K. Brand for technical assistance. The trachea of the adult CF mouse (cftr tm1Unc or cftr tm1Kth ) expresses little CFTR and exhibits neither a defect in cAMP-mediated Clsecretion nor hyperabsorption of Na + , both signature abnormalities in human CF airways. We generated a mouse exhibiting hyperabsorption of Na + in the airways by over-expressing the β subunit of the epithelial Na + channel (βENaC, gene Scnn1b). βENaC mice exhibit airway pathology (decreased mucociliary transport, mucus plugging, and airway inflammation) similar to human CF airways. To determine whether inactivation of CFTR in βENaC mice results in a more severe phenotype, we generated a CF/βENaC mouse by crossing the ∆F508 (C57BL/6N) mouse (CF mouse) with the βENaC mouse (C57BL/6N). Survival analysis showed a drastic reduction in survival for CF/βENaC mice in comparison to CF mice or βENaC mice. We studied the airway bioelectrics of 3 day-old pups of each genotype. The basal short circuit current (I sc ) was ~5 fold greater in wild-type (WT) (50 ± 13 µA . cm -2 ; mean ± SEM, n=4) vs. CF (12.2 µA . cm -2 , n=2) neonatal tracheas. As previously reported, the basal I sc in βENaC preps was significantly elevated (99.2 ± 14 µA . cm -2 , n=7) vs. WT tracheas. However, the basal I sc in the CF/βENaC tracheas (26.6 ± 19 µA . cm -2 , n=4) was intermediate between the CF and the βENaC tracheas. The amiloride-sensitive I sc was elevated similarly in the tracheas of the βENaC and CF/βENaC pups and was ~ 3X greater than that exhibited by the WT or CF tracheas. The post-amiloride I sc was strikingly reduced in both CF and CF/βENaC tracheas compared to the WT and βENaC tracheas. We suggest that the post-amiloride I sc reflects constitutive CFTR-mediated Clsecretion, which may help hydrate airways in the neonate. Unlike the tracheas of adult CF mice, neonatal CF tracheas exhibited the classic defect in cAMP-mediated Clsecretion, with the forskolin response in the CF and CF/βENaC tracheas being reduced ~10fold compared to the response in the WT or βENaC preps. The response to UTP (Ca ++ -mediated Clsecretion) was similar in the four genotypes. Absence of CFTR-mediated Clsecretion does not appear to cause airway pathology in neonatal CF mice, but when this defect is coupled with Na + hyperabsorption, it results in a significant increase in neonatal mortality. Due to lack of CFTR-mediated Clsecretion, CF/βENaC mice appear to experience greater mucus adhesion, producing death due to asphyxiation. In addition, the tracheas of CF mice (especially congenic C57Bl/6N) often appear constricted or collapsed due to a cartilaginous defect resulting in increased compliance. This anatomical defect does not appear to be detrimental to the CF mouse. However, in CF/βENaC pups, that accumulate thick, sticky mucus in their tracheas, this tracheal constriction may contribute to their asphyxiation. These results demonstrate the importance of both abnormal Na + absorption and decreased Clsecretion in the pathogenesis of CF lung disease. Supported by NIH SCOR P50 HL060280. Recently, we reported that species-specific differences in the airway biology of rAAV transduction exist between mice and human (Am J Respir Cell Mol Biol, 2006, 34:56-64) . These differences significantly interfere with the use of mice as surrogate models for human lung gene therapy with certain rAAV serotypes. To survey which species might be an optimal animal model for rAAV preclinical testing, we utilized differentiated airway epithelia grown at an air-liquid interface (ALI) from five species (human, non-human primate [NHP] , pig, ferret, and mouse) to test the efficiency of transduction with three serotypes of rAAV (type 1, 2, and 5). Our results demonstrated significant species-specific differences in rAAV transduction biology in airway epithelia. The serotype preferences of rAAV transduction from the apical membrane in human, ferret, and pig epithelia was rAAV1>rAAV2≈rAAV5, different from that seen in mouse epithelia (rAAV1>>rAAV5>>rAAV2). Surprisingly, the profile of rAAV transduction with these three serotypes in NHP rhesus monkey airway epithelia (rAAV2≈rAAV5>>rAAV1) was less similar to human than pig and ferret. Additionally, rAAV1 lacked a bias in polarity of transduction in human airway epithelia, however, in NHP epithelia, rAAV1 transduced the basolateral membrane 10-fold more effectively than the apical membrane. Furthermore, differences in the properties of rAAV1 receptors between these species was also seen. Using neuraminidase treatment of the apical surface of ALI cultures, we tested whether N-linked sialic acids were potential receptors for rAAV1 binding and/or transduction. These studies demonstrated removal of N-linked sialic acids inhibited rAAV1 transduction of human, NHP, and mouse airway epithelia, but not ferret and pig. However, rAAV1 binding in this setting did not correlate with changes in transduction; neuraminidase treatment dramatically reduced rAAV1 binding to NHP epithelia, but enhanced binding to human epithelia (~100 fold). These findings suggest viral binding to airway epithelial cells may not be a major barrier for rAAV1 transduction. Evidence that proteasome-modulating agents enhancing rAAV transduction 2-3 logs for all serotypes and species tested, suggest that the ubiquitin/proteasome pathway represents a common intracellular block in rAAV transduction of airway epithelia. These findings support the notion that viral interactions at the surface of airway epithelial cells (with co-receptors and receptors) influence intracellular processing AAV virions that can alter the efficiency of productive transduction. ENaC surface density in cortical collecting duct (CCD) epithelial cells is established by cAMP/PKA regulated insertion and retrieval of the channel at the apical membrane, together with the recycling of internalized ENaC back to the apical surface (Butterworth et al. (2005) Journal of General Physiology 125: 81-101). Ubiquitination by the Nedd4-2 ligase elicits endocytosis of surface ENaC, which may lead to its degradation. However, ENaC recycling to the apical surface requires ubiquitin removal by a deubiquitinating enzyme (DUB).

Using a chemical probe approach, we identified ubiquitin carboxy-terminal hydrolase (UCH-L3) as the predominant DUB in vesicular endocytic and recycling compartments in mouse CCD epithelia. Specific inhibition of UCH-L3 (10µM of 4,5,6,7-Tetrachloroindan-1,3-dione) in filter-cultured CCD cells resulted in a rapid decline in ENaC currents, which was accelerated by forskolin stimulation, suggesting that UCH is responsible for recycling ENaC to the apical surface. Knockdown of UCH-L3 significantly reduced both basal and cAMP-stimulated ENaC currents (from 7.7±1.2µA/cm2 in control siRNA treated to 1.5±0.7µA/cm2 in UCH-L3 knockdown cells).

To determine whether the same DUB was responsible for the regulation of ENaC in human bronchial epithelial (HBE) airway cells, differentiated cultures derived from CF and non-CF lung tissue were treated with the UCH inhibitor. To address the possibility that protease cleaved ENaC may be regulated differently from unprocessed channels, differentiated HBE cultures were either maintained at air/liquid interface or were submerged under liquid/liquid conditions to alter the cleavage state of ENaC, as we have previously described (Myerburg et al. (2006) Journal of Biological Chemistry 281: 27942-49). In all cases, no significant difference in ENaC-mediated Na+ transport was observed between control and UCH-L3 inhibited conditions. These data are in sharp contrast with the mechanisms of ENaC regulation observed in the CCD model system, and they support our previous findings that ENaC turnover and mode of regulation in HBE are distinct from the recycling mechanisms defined in the CCD epithelia (Butterworth et al. Primary human tracheobronchial epithelial (hTBE) cells cultured at an air-liquid interface (ALI) regulate ion transport, mucin secretion, and cilia beating to recapitulate directional mucus clearance, a key innate defense mechanism that fails in cystic fibrosis (CF). Thus, ALI hTBE cell cultures are crucial for studying CF pathogenesis and for developing/validating novel therapies. Different research centers use alternative protocols that enable reproducible creation of ALI hTBE cell cultures. However, limited primary hTBE cell availability has driven the creation of cognate cell lines, accomplished mainly by the introduction of viral oncogenes. Viral oncogenes disrupt normal cell physiology causing cellular hyperplasia, abundant apoptosis, genomic instability and, sometimes, poor polarization/differentiation, rendering the cells less useful than primary cells for assessing CFTR function. Furthermore, genetically unstable cell lines will continuously acquire changes independent of CFTR genotype. Bmi-1 is a polycomb group protein that controls cell cycle and cell identity via epigenetic regulation of chromatin, maintaining stem cells by suppressing expression of p16, an inhibitor of cyclin dependent kinases. Bmi-1 expression has been used to create cell lines that recapitulate normal cell structure and function better than viral oncogene-immortalized cell lines. Using HIV lentiviral vectors, we introduced Bmi-1 and the catalytic subunit of telomerase to enhance the growth of 3 different ∆F508 homozygous CF and 3 non-CF primary human airway epithelial cell preparations. All 6 new cell lines grew for at least 40 population doublings, while their normal counterparts senesced prior to a maximum of 20 doublings. At passage 14-15, the new cell lines had a diploid karyotype compared to grossly abnormal chromosomes in cells immortalized by viral oncogenes. Ussing chamber analysis of ALI cultures at passage (p) 14-15 revealed variable transepithelial resistances among the cell lines ranging from 125 ->1000 Ω*cm2, but short circuit current (Isc) responses stimulated by forskolin and inhibited by CFTR172 were true to the cell's genotype and comparable to early passage primary cells, except for one non-CF cell line with relatively low CFTR Isc's at p15. Amiloride sensitive and UTP-stimulated Isc's were present but were more variable in magnitude in comparison to the currents observed in low passage primary cells. ALI cultures exhibited a pseudostratified morphology with prominent apical membrane polarization, few apoptotic bodies, numerous mucous secretory cells and occasional ciliated cells. All CF and non-CF cell lines in culture produced similar levels of IL-8 at baseline and CF and non-CF cells equally increased IL-8 secretion in response to IL-1β, TNFα and the Tolllike 2 receptor agonist, Pam3Cys. These novel cell lines will help fill gaps currently hindering CF research and therapeutic development.

Supported by Cystic Fibrosis Foundation Grant Randel04G0. We recently considered that domestic dogs could represent a large natural reservoir to search for CFTR mutations and, ultimately, to identify and breed animals with CF traits. The canine genome has been completely sequenced, and the canine CFTR gene, found on chromosome 14, consists of 27 exons spread over 163 kb. An animal model that faithfully expresses human CF lung pathology is critical for the development of effective treatments for this most lethal aspect of CF disease. While transgenic mouse models of CF replicate numerous aspects of CF disease, they unfortunately do not develop the characteristic lung pathology including mucus plugging of airways, chronic bacterial infections, and bronchiectasis. The reason for this is not completely understood but it is likely related to significant differences in lung morphology between these species. To date, no non-murine genetic model of CF has been developed, and naturally-occurring, CF-causing mutations have not been identified in any non-human animal species. Temporal temperature gradient gel electrophoresis (TTGE), developed as a screening tool to identify sequence variations and mutations in the human CFTR gene, was used to screen dogs for CFTR mutations. DNA was purified from whole blood received from veterinary clinics treating dogs for a variety of ailments. We have performed TTGE analysis on DNA from 179 dogs, which includes 24 dogs with pancreatitis and 4 dogs with bronchiec-tasis, and have found hundreds of potential mutations amongst the thousands of amplicons tested. TTGE can detect single base pair changes within a fragment of DNA. As the temperature increases, the migrating DNA becomes partially melted, creating single stranded regions (internal bubbles and branched ends) that decrease its mobility through the gel. The DNA fragments have distinct melting profiles that are sequence dependent; therefore, changes in sequence of even a point mutation can alter the profile and result in a change in the rate of migration through the gel that can distinguish wild type from even a homozygous mutant. Thus far 65 exonic mutations have been sequenced but none have caused changes to CFTR at the amino acid level (i.e., the mutations have all been silent). We conclude that TTGE is a sensitive and effective method for screening large canine populations for CFTR mutations. ( Introduction: The search for "correctors" of defective protein processing of ∆F508 CFTR has led to the development of high throughput screening assays to find compounds that promote the surface activity of the mutant protein. By necessity, primary screening assays typically use cell lines expressing recombinant ∆F508 CFTR. However, "hits" identified in primary screening require scrutiny for false positives whose activity is either dependent on the ∆F508 CFTR expression system or insufficiently robust to correct processing in the complex physiological setting of a native epithelium. To this end, we have developed a secondary functional assay that measures the shortcircuit current (Isc) responses in organ cultures of gallbladder mucosa from ∆F508 CFTR mice (cftr tm1Kth ). Previous studies have shown that murine ∆F508 CFTR recapitulates the protein processing defect including the capability to increase processing when mucosa is incubated at reduced temperature (27°C) (J. Clin. Invest. 98:1304 Invest. 98: , 1996 . Gallbladder mucosa was chosen for this assay because the murine gallbladder can be removed aseptically for organ culture, the epithelium is essentially a monolayer, and the mucosal wall is transparent which facilitates morphological assessment.

Methods: Using either wild-type (WT), ∆F508 CFTR (DF) mutant or CFTR knockout (CFTR KO) mice, excised gallbladders were opened longitudinally, rinsed and positioned on loose nylon mesh. The preparation was mounted between two halves of a polycarbonate culture cup with 4.5 mm aperture and placed in submersion culture for 0 -2 days at either 37°C or 27°C using supplemented Ham's F12 medium mixed 50:50 with NIH 3T3 fibroblast-conditioned DMEM containing 2% calf serum. After incubation, the culture cups were placed in custom designed Ussing chambers for measurement of Isc responses to a cocktail of 10 µM forskolin and 100 µM IBMX (cAMP Isc) at 37°C. Contribution of calcium-activated Clchannels to the cAMP Isc was eliminated by addition of the distilbene DIDS (100 µM) to the luminal bath.

Results: The mean ∆cAMP Isc responses (in µA/cm 2 ) after two days incubation were as follows: WT at 37°C = 21.1 ± 4.1 (n = 4); DF at 37°C =3.9 ± 0.3 (n = 5); DF at 27°C = 13.6 ± 2.7 (n = 6); and CFTR KO at 27°C = 2.8 ± 0.1 (n = 3). In a pairwise comparison, the WT (37°C) was significantly greater than the DF (37°C) and CFTR KO (27°C) but not significantly different from DF (27°C).

Conclusions: Incubation of ∆F508 CFTR mutant murine gallbladder mucosa at 27°C for two days in organ culture restored approximately 65% of the CFTR-dependent Isc responses to cAMP stimulation. The temperature-dependent effect is consistent with correction of murine ∆F508 CFTR processing, indicating a physiological screening assay that is useful for preclinical testing of candidate drugs targeted at correcting the protein processing defect of ∆F508 CFTR.

Supported by CFFT and NIH.

The lack of a large animal model that replicates the lung disease found in humans with cystic fibrosis is a major impediment to understanding disease pathogenesis and the development of effective therapies. Although cystic fibrosis is a multi-system disease, airway infection and inflammation currently cause most of the morbidity and mortality in patients. Several homozygote null and missense CFTR mutations have been developed in mice, but they do not exhibit the lung phenotype observed in humans. In contrast to mice, the structure and physiology of porcine lung and airways closely resemble those of humans. Therefore, the goal of this project is to develop a swine model for cystic fibrosis by combining gene targeting and nuclear transfer to produce a pig with a CFTR null genotype. We employed homologous recombination to disrupt exon 10 of pig CFTR with a neomycin resistance cassette. We generated primary male fetal fibroblasts that were positive for CFTR-targeting at one allele as detected by multiple PCR screens. Genomic Southern blot analysis confirmed proper gene targeting and revealed most clones were free of random integration events. These cells were used for somatic cell nuclear transfer that resulted in the birth of ten CFTR null heterozygote piglets. Subsequent breeding of these male pigs to wild-type females has yielded at least 9 female CFTR null heterozygotes. Breeding male and female heterozygotes will provide CFTRnull homozygotes. Our recent published and preliminary studies demonstrate that mouse Slc26a9, a newly discovered member of Slc26 family of anion exchangers, is expressed on the apical membrane of tracheal epithelial cells and gastric surface epithelial cells and in the tubulovesicle membranes of gastric parietal cells. The functional properties of mouse Slc26a9 were examined in Xenopus Oocytes and in stably transfected cultured cells. In Xenopus Oocytes injected with the Slc26a9 cRNA and studied with the voltage and/or pH sensitive microelectrodes, the rank order for anion selectivity at +80mV was Cl->I-> NO3->gluconate > sulfate, and that of the selectivity (Px/PCl) was I-> Cl-> NO3-= gluconate > sulfate. Inhibitor profile studies demonstrated that NPPB at 100 µM inhibited the current at 5 min by 40% (p<0.001 compared to control period, n=6). The effect of NPPB was partly reversible. DIDS at 500 µM had no significant effect on the Slc26a9-mediated current (p>0.05 compared to control period, n=6). Niflumic acid at 100 µM inhibited the current at 5 min by ~38% (p<0.001 compared to control period, n=5) which was partly reversible. CFTR inh 172 at 5µM inhibited the current at 5 min by ~23% (p<0.005 compared to control period, n=5) which was reversible. Clamping currents in a 85 mM K-gluconate medium and using 10 mM K-Cl or K-bicarbonate revealed very low bicarbonate conductance for Slc26a9 compared to that of Cl-. No significant pHi changes were observed in oocytes when favoring Cl-exit by substituting perfusate Cl-with gluconate in the presence of CO2/HCO3-. Conversely we observed significant alkalinization in AE1-expressing oocytes in the same experimental solutions. Interestingly, in HEK 293 cells stably transfected with the mouse Slc26a9, a significant intracellular alkalinization was observed upon switching to a chloride-free perfusate. The alkalization in Slc26a9-expressing cells was inhibited by ~55% in the presence of 500 µM DIDS. Baseline intracellular pH was significantly increased in Slc26a9expressing HEK293 cells vs. non transfected cells. However, baseline pHi was comparable in control and Slc26a9-expressing cells in the presence of 100 µM EIPA, indicating the activation of endogenous NHE by Slc26a9. In conclusion, Slc26a9 displays characteristics of an anionic channel in xenopus oocytes, with low bicarbonate permeability. In mammalian expression systems, Slc26a9 can also function in Cl-/HCO3-exchange mode with moderate sensitivity to inhibition by DIDS. We propose that Slc26a9 plays an important role in anion (chloride) secretion and/or pH regulation in tracheal and gastric epithelial cells. No cell lines exist at present that robustly express endogenous wt and ∆F508 CFTR on the same genetic background. Zinc finger nucleases (ZFNs) allow the introduction of specific genetic changes at loci within the human genome (Nature 435: [646] [647] [648] [649] [650] [651] 2005) . We have been using this technology in an attempt to generate human cells carrying the major CF-causing genotype -∆F508 -at the endogenous CFTR locus. Such lines could be expected to provide better understanding of the ∆F508 maturational processing defect, as well as new model systems for cell based assays of ∆F508 repair. We performed a bioinformatic analysis of CFTR exon 10, and designed multiple distinct ZFNs against the ∆F508 DNA sequence. All of these nucleases were assembled, cloned into mammalian expression vectors, and assayed for DNA binding affinity and specificity. The ZFNs were then assayed for endogenous CFTR locus editing in Calu-3 and HT-29 cells using a novel, massively parallel methodology applied for the detection of genome editing events. The effort demonstrated that specific ZFN pairs successfully bind to, and cleave, within exon 10 of the endogenous CFTR locus in both cell types. We will describe experiments that yielded optimized ZFNs for the CFTR locus, and a robust viral platform to transiently deliver genome editing reagents to the human pulmonary cell line, Calu-3. Supported by CFF. The localization of Slc26 anion transporters, as well as their function vary among members. Some (Slc26a2, Slc26a3, Slc26a6, Slc26a11) are widespread, implying specific physiological function in diverse tissues. Others are expressed only in a single epithelia, e.g., Slc26a5 (prestin) in the inner ear. We have previously found that Slc26a9 is a highly electrogenic Cl --HCO 3 exchanger with a large anion conductance. To determine the possible physiologic roles of Slc26a9, we developed a chicken antibody (designed against sulfate transporter and anti-sigma domain, STAS, of Slc26a9) to localize the protein. To test the antibody specificity for functional protein, we inject Xenopus oocytes with cRNA for Slc26a6, Slc26a7 or Slc26a9 and documented appropriate pH and membrane potential changes. Only oocytes functionally expressing Slc26a9 were stained using our primary Slc26a9 antibody. Further, specific staining was eliminated when either pre-immune serum or blocking peptide was used. Slc26a9 was only found in epithelia. In the respiratory system, Slc26a9 was in the apical membrane of nasal epithelial cells and tracheal submucosal glands. Membranes of CaLu-3 cells, derived from human tracheal submucosal glands, stained indicating that our antibody also recognizes the human SLC26A9 protein. Slc26a9 is found in the proximal parts of the gastrointestinal tract (esophagus, stomach), with lesser amounts in the small intestine and colon. However, Slc26a9 is not present in liver. Slc26a9 is present in ducts of secretory epithelia such as salivary glands and pancreas. In the urinary tract, Slc26a9 is localized to the renal proximal tubule apical membrane but not in urinary bladder. In the reproductive system, we found Slc26a9 in epithelia of the uterus and seminal vesicles but not the vas deference or the prostate. Finally, Slc26a9 is found on the eye surface, i.e., corneal epithelial cells. Together, these results and the varied Slc26a9 physiology suggest that several epithelial function models will need to be re-evaluated to incorporate an electrogenic Cl --HCO 3 exchanger and/or another apical anion channel. (support: ROMERO-06G0, SINDIC-06F0).

A TGFβ-1 polymorphism modifies severity of cystic fibrosis lung disease in the setting of ∆F508 CFTR homozygosity (Drumm et al., NEJM, 2005) . The mechanism(s) underlying this observation, and whether elevated or repressed TGFβ are responsible, are not known. In addition, the pathogenic role of TGFβ in other CF organs and in murine models of the disease has not been well studied. Because the TGFβ codon 10 CC genotype is associated with worsened CF pulmonary prognosis, and since high levels of TGFβ are also known to suppress CFTR mRNA expression and apical CFTR protein expression in colonic epithelia (Howe et al., Exp Cell Res, 2004) , we developed a mouse model to test TGFβ/CFTR interactions that may directly influence CF phenotype. We bred ∆F508 heterozygous mice (Cftr tm1Kth , ∆F508 CFTR) against the TGFβ receptor-deficient murine strain (MTRII28). MTRII28 mice express a dominant-negative mutant form of the TGFβ type II receptor which lacks the intracellular (kinase) receptor domain. The receptor binds TGFβ, but fails to transmit a TGFβ dependent signal. The TGFβ receptor transgene was inserted downstream of a metallothionein-derived promoter and expressed in numerous murine tissues following induction with 25 mM ZnSO 4 in drinking water. An antibody raised against active TGFβ (LC-1-30-1, gift of Dr. Kathy Flanders, National Cancer Institute) was used to establish TGFβ in murine tissues including lung and intestinal tract of the MTRII28 strain. RT-PCR confirmed high level transgene expression similar to gapdh internal control. Nasal potential difference measurements in CFTR +/+ mice expressing the TGFβ transgene indicated intact cAMP dependent Cltransport following induction of the dominant-negative receptor. Homozygous CF animals deficient for TGFβ signaling have recently been obtained and exhibit the expected litter sizes and distribution of the ∆F508 allele. Studies of CF homozygous mice following induction with ZnSO 4 are in progress, and will be used to determine whether a failure to signal through TGFβ in vivo 1) influences the severity of CF intestinal disease, 2) alters airway or intestinal bioelectric findings characteristic of CF, or 3) changes survival, histopathology or other features of the disease. These experiments are intended to provide an in vivo means of investigating TGFβ signaling and resultant CF modifier effects in vivo. Supported by NIH and CFF. OBJECTIVE: To determine if the ABPA phenotype observed in CFTR mice was preventable by corrective gene replacement, we delivered AAV5 CFTR to the lung. In addition to further assess whether CFTR-/-lymphocytes alone where responsible for the phenotype, we performed adoptive transfers of CFTR-/-lymphocytes into CFTR+/+ mice and proceeded to expose the mice to the ABPA model.

Design/Methods: CFTR-/-mice underwent IT administration of rAAV5 CFTR or rAAV5GFP. All mice were then sensitized with 20ug of crude Af antigen via intraperitoneal route. Non-injected mice were sensitized without prior virus exposure. Challenges were performed on all mice, using an Af solution delivered by aerosol inhalation in a closed chamber on 3 consecutive days. Serum IgE levels were then obtained, as well as bronchoalveolar lavage fluid for differential cell counts, histologic and cytokine analysis. Adoptive transfer experiments were done using spleens cells isolated from Af sensitized delaF508-/-and F508+/+ mice and transfer into C57/Bl6_Rag-/-mice. After 7 weeks post transfer C57/Bl6_Rag-/-mice with F508-/-and F508+/+ lymphocytes were then challenged with Af.

RESULTS: Following challenge with Af, rAAV5-CFTR mice had significantly lower total serum IgE levels (17,251) as compared to rAAV5-GFP controls (26,892) (p=0.037). Analysis of Af-specific IgE also revealed a two-fold reduction in the mice receiving the corrective gene therapy. Curiously, gene transfer to the lung lowered the secretion of inflammatory cytokines from activated splenocytes of rAAV5-CFTR when compared to rAAV5-GFP controls. These results suggest that correction of immune cells in the lung that traffic to the spleen may be partially responsible for the attenuation of the phenotype. In addition the transfer experiments showed that the hyper-IgE syndrome can be partially transferred into B6Rag-/-by F508-/-splenocytes.

CONCLUSIONS: Correction of CFTR deficiency by AAV lung transfer attenuates the inflammatory phenotype associated with ABPA. Epithelial cells lining the vas deferens express cyclooxygenase (COX) 2 when exposed to testosterone in vivo, although no functional link to ion transport has been established. We hypothesized that testosterone modulates CFTR-mediated anion secretion across vas deferens epithelia that is stimulated via COX-dependent pathways. Vas deferens epithelial cells were isolated from human and porcine tissues and cultured as monolayers on permeable supports until assayed in modified Ussing chambers. RNA was isolated concurrently for semi-quantitative gene expression analysis. Bradykinin (BK) caused an increase in anion secretion (measured as short circuit current) and this effect was inhibited by indomethacin, which indicates COX dependency. Testosterone treated monolayers exhibited a BKinduced response that is 58% greater than paired vehicle treated monolayers. qRT-PCR revealed that COX-2 mRNA is 19 times more abundant than COX-1 mRNA in cultured vas deferens epithelial cells. The effects of BK on ion transport are inhibited by the B2 receptor antagonist HOE140 while the B1 receptor agonist Des-Arg9-BK produced no effect. Prostaglandin E2 and selective agonists of the EP4 (1-OH-PGE1) and EP2 (butaprost) receptors produced concentration dependent increases in short circuit current whereas sulprostone, an agonist of EP1 and EP3 receptors, had no effect. Taken together, these results suggest that testosterone plays a key permissive role for BK-stimulated anion secretion by inducing COX-2 expression. Bradykinin acts at the B2 receptor to increase COX-2 activity and the synthesis of prostaglandins that ultimately bind to EP receptors to promote CFTR-dependent HCO3-and Cl-secretion that would be expected to modify the environment for sperm. The results further suggest that developmentally associated increases in testosterone levels during gestation and puberty will produce physiological changes in vas deferens luminal environment that are likely compromised in cystic fibrosis patients.

Supported by Cystic Fibrosis Foundation SCHULT06PO and KSU COBRE NIH RR-17686. Background: Recent studies in cystic fibrosis (CF) suggest a prominent T H 2 profile; a facet that may interfere with normal bacterial clearance. This phenotype appears to increase secondary to infection with Aspergillus fumigatus or Pseudomonas aeruginosa and also, may interfere with macrophage activation. These same characteristics are present in CFTR-deficient mice (CFTR S489X -/-FABP-hCFTR (+/+), hereafter designated CF mice). As a study of pathogenesis of CF related diabetes (CFRD), we previously observed increased IL-10 and splenocyte proliferation in CF mice subject to streptozotocin-induced hyperglycemia. Herein, we hypothesized that differences in glucose concentration would increase T H 2 profiles of splenocytes in vitro.

Methods: CF mice and C57BL/6J mice (B6 mice; n=3/group) were sacrificed at 14 wk of age. Splenocytes were collected and cultured at the following glucose concentrations: 10, 15, 20 and 25 mM. Each mouse and glucose concentration was stimulated with ConA, LPS and CD3/CD28. Supernatants were collected at 48 hr and measured by Luminex analysis. Statistical analysis was performed by a one-way ANOVA.

Results: Stimulation by ConA produced increased levels of IL-10, IL-4, TNF-a, GM-CSF and IL-2 in splenocytes of CF mice (p<0.01 for all). Stimulation by LPS produced decreased concentrations of IL-10 in CF mice(p<0.05). Stimulation of CF mice by CD3/CD28 produced decreased concentrations of IL-2 (p<0.001). Interestingly, no difference was observed in cytokine elaboration as a function of glucose concentration. Data are presented in table as fold increase of cytokine concentration for CF mice compared to B6 mice.

Discussion: Profound differences in cytokine production were measured in stimulated splenocyte responses of CF in comparison to B6 mice. The profiles did support the notion for a prominent T H 2 response, with the potential to interference of normal macrophage activation. In addition, differences in cytokine production profiles appear related to the nature of adaptive and innate stimuli provided (ConA vs. LPS). The absence of influence by glucose concentration in vitro suggests that local/immediate concentration of glucose is not relevant to prolonged expression of cytokine responses to stimuli. Further studies on the impact of chronic hyperglycemia and its relation to immune response are necessary. An improved understanding of these mechanisms holds hope for a better overall outcome in patients with CF and CFRD. Studies from different laboratories to characterize ∆F508 CFTR Clchannel activity have yielded disparate results, likely due to variations in technique, cell type, or method of analysis. In the present study, we compared ∆F508 CFTR activation in two well characterized polarizing cell models. Fisher rat thyroid (FRT) cells have been used in high-throughput drug screening to identify ∆F508 CFTR active agents, whereas CFBE41ocells have been transduced to provide robust ∆F508 CFTR expression in an airway epithelial background. ∆F508 CFTR biogenesis following low temperature correction was compared, indicating qualitatively similar levels of CFTR transcripts and Band B and C CFTR. In paired studies in Ussing chambers, distinct differences in Isc were seen following exposure to CFTR activating stimuli. In FRT cell monolayers grown at 37°C, forskolin (20 µM) or genistein (50 µM) were strong stimuli of Isc, producing mean Isc of 13.0 +/-6.6 (forskolin) and 18.6 +/-7.2 (genistein) µA/cm 2 . Regardless of exposure sequence, both agents induced equivalent changes in maximal current (59% of 42.2 µA/cm 2 by forskolin when administered prior to genistein; 54% of 44.7 µA/cm 2 by forskolin in converse order). Growth at 27°C for 24 hours increased currents approximately 3.7 fold, but relative contributions of the two stimuli were similar to cells grown at 37°C (42% of total current stimulated by forskolin). In CFBE41o -∆F508 cells grown at 37°C, forskolin-dependant currents were minimal and genistein-stimulated currents small (4.2 +/-1.4 µA/cm 2 , compared to zero current in parental, nontransduced CFBE41ocells, p<0.05). In temperature-corrected ∆F508 CFBE41ocells, forskolin was a poor stimulus of Cl-transport relative to genistein, contributing 15% to the total current produced by both agents (p<0.001). Genistein-stimulated currents did not require pre-treatment with forskolin to produce maximal effects (28.8 µA/cm 2 with genistein alone vs. 31.1 µA/cm 2 after forskolin and genistein), and forskolin did not cause further Isc when added after genistein (0 µA/cm 2 ). Levels of cAMP did not account for these differences, since forskolin markedly increased cAMP in both models. To determine whether excessive phosphodiesterase (PDE) or phosphatase (PP) activity limited ∆F508 CFTR activation by forskolin in the CFBE41o -∆F508 cells, monolayers were pretreated with the nonspecific PDE inhibitor papaverine (100 µM) or the PP2 inhibitor endothall (400 µM), each sufficient to activate Isc in CFBE41oexpressing wtCFTR (papverine: 59.8 vs. 1.3 µA/cm 2 with vehicle, p<0.005; endothal: 40.1 vs. 8.8 uA/cm 2 , p=0.05). Neither agent restored Isc in ∆F508 CFBE41oby forskolin (0.8 mV difference, p=NS). These results indicate that activity of ∆F508 CFTR in polarizing monolayers is exquisitely dependant on the model chosen for study. Substantial differences in ∆F508 CFTR behavior in two commonly used cell models, and suggest that the failure of cAMP to activate ∆F508 CFTR in CFBE41ocells is due to cell-specific factors that may 1) influence ∆F508 CFTR folding and tertiary structure, or 2) favor interactions of ∆F508 CFTR with inhibitory binding partners that limit activation by cAMP. Adverse immune responses to viral vectors or vector-encoded proteins may hinder the practical application of gene therapy. To date, little effort has been devoted to determining the immunological consequences of topically delivered lentiviral vectors to respiratory epithelia. Previously, we demonstrated that an integrating feline immunodeficiency virus-based lentiviral vector pseudotyped with a baculoviral envelope glycoprotein (GP64-FIV) readily transduced differentiated airway epithelia in vitro when applied to the apical surface. Furthermore, using a luciferase reporter gene and bioluminescence imaging, we observed in vivo gene transfer to murine nasal epithelia following a single application of GP64-FIV. Longitudinal bioluminescence analysis documented persistent expression in nasal epithelia for greater than 11 months without significant decline. By histological analysis, surface epithelial cells were transduced following a single nasal application of GP64-FIV expressing beta-galactosidase. Importantly, we observed that 7 consecutive doses of GP64-FIV delivered over 7 consecutive days greatly increased the number of beta-galactosidase positive epithelial cells. Using quantitative bioluminescent imaging, we observed that repeated doses given over consecutive days resulted in a linear additive increase in gene expression. In addition, we performed studies investigating gene transfer efficacy and immune responses following lentiviral vector readministration with greater intervening time periods. In control mice, we observed that giving 4 nasally instilled doses (2 weeks apart) of an E1/E3 deleted Ad5 vector resulted in significantly attenuated expression after the second dose. The attenuated expression correlated to the production of neutralizing antibodies. In contrast, using the same delivery protocol, 4 doses of GP64-FIV resulted in gene transfer without loss of expression of the forth dose. No significant production of neutralizing antibodies was observed following the lentiviral delivery protocol. We further report the additive increase in reporter gene expression following 7 doses of lentiviral vector delivered over 7 consecutive weeks (1 dose/week), as well as the lack of production of systemic and local neutralizing antibodies. GP64-FIV efficiently transduces and persistently expresses a transgene in respiratory epithelia and has the potential for repeat administration without eliciting adverse immune responses. These data have important implications for the application of lentiviral gene therapy to pulmonary diseases. Uniform delivery of gene therapy vectors to all lung lobes is important for CF gene therapy. However, this important objective has not been achieved in large animals. To address this obstacle, we have developed an approach to specifically deliver vector into all lung lobes. In this strategy, an intracorporeal nebulizing catheter (Aero-Probe) is inserted into a bronchoscope to permit visual targeted aerosolization of vector specifically into each lung lobe. Using this approach, 2 ml of 0.1% LPC (to transiently open tight junctions) containing 1x1012 vp of HDAd-K18LacZ was sequentially aerosolized into each of the six major lung lobes of a baboon. Aerosolization was triggered by the ventilator upon inspiration at 0.1 ml/pulse at a rate of 8 to 10 pulses/min. Halfway through the procedure a very slight, transient and fluctuating decrease in oxygen saturation of no more than 2% was noted that did not warrant supplemental oxygen. The entire procedure was otherwise uneventful and well tolerated with no clinical manifestations of toxicity. No changes in chest X-rays taken at 6 h, 24 h and at necropsy 72 h post-vector were observed compared to baseline. X-gal staining of the lungs at 72 h revealed extensive transduction of the epithelium in the large and small airways in all six major targeted lung lobes. X-gal histochemistry revealed substantial transduction that was exclusively restricted to the airway epithelial cells and submucosal glands, the target cells for CF gene therapy. Assessment of the proximal major bronchi from each lung lobe revealed that~20% of the airway epithelial cells were transduced in the Left Upper Lobe (LUL), ≥ 90% were transduced in the LML, LLL, RUL, RML and RLL, and~80% were transduced in the accessory lobe. Approximately 70% of the airway epithelial cells in the tracheal sections examined were transduced. We also investigated the duration of pulmonary transgene expression. In these studies, we aerosolized a HDAd expressing the baboon α-fetoprotein from the K18 promoter into the lungs of baboons. By measuring serum bAFP levels, we found that pulmonary transgene expression from transduced airway epithelial cells can be detected for at least 177 days post-vector. These results demonstrate for the first time that exceedingly high levels of transduction of the airway epithelial cells and submucosal glands throughout all lung lobes in a large animal can be achieved with negligible toxicity resulting in long term trangene expression and should pave the way towards successful clinical CF gene therapy. Inefficient gene transfer after repeat administration of most viral vectors has limited their suitability for CF gene therapy. However, lentiviral vectors are reported to be less immunogenic. This property combined with their capacity to integrate into the genome of transduced cells and, therefore, allowing gene expression for the life-time of the cell (approximately 3 months for airway epithelial cells and theoretically unlimited for stem cells) justify further investigation of these vectors for CF gene therapy. To ensure high transduction efficiency in the lung without the need for any pre-conditioning strategies we pseudotyped a simian immunodeficiency virus (SIVagm)-based vector with the Sendai virus F and HN proteins (F/HN-SIV-GFP) and transduced mouse nasal epithelium (4x10e8 TU/mouse). At this titre transduction efficiency along the nasal septum was up to 5% when measured 10 and 30 days after transduction (n=4/time-point). The vast majority (approx. 80%) of GFP positive cells across all time-points were respiratory epithelial cells followed by neuronal and sustenticular cells (approx. 20%) in the olfactory epithelium. Interestingly, we observed two types of duration patterns of GFP expression in mouse nose. In some mice the number of GFP positive cells decreased over time (day 30: 218±17 cells/section, n=4, day 360: 19±17 cells/section, n=6). However, in other mice strong GFP expression persisted for up to 360 days (day 360: 159±81 cells/section, n=4), significantly longer than the proposed life-span of airway epithelial cells. In order to support the idea of F/HN-SIV integration into nasal respiratory stem/progenitor cells we artificially induced cell division after SIV vector transduction by damaging the nasal tissue with the detergent (polidocanol), which strips of the epithelium within a few hours, while retaining the basal cell layer. The epithelium completely regenerates over the course of a few days. Seven and 28 days after the vector transduction (4x10e8 TU/mouse, n=3) the nasal tissue was perfused with 2% polidocanol (10 µl/mouse) and gene expression was analyzed 4 weeks after the last detergent treatment. GFP expression was detectable in various cell types of the regenerating epithelium including ciliated and basal cells. Importantly, GFP-expressing cells were more clustered after polidocanol treatment, possibly indicating origination from a common progenitor. These data suggest that, F/HN-SIV integration may have occurred into differentiated airway epithelial cells as well as into stem/progenitor cells involved in the regeneration of the airway epithelium. Introduction: Hypersecretion of mucus is a major cause of airway obstruction in cystic fibrosis, asthma and chronic bronchitis. Though mucus is a complex, non-homogeneous mixture of secretions, one group of its constituents, the mucous glycoproteins or mucins, contributes greatly to the obstruction associated with airway diseases. Currently, there are a limited number of therapeutic agents available for controlling mucin overproduction and hypersecretion. The objective of this study is to develop a sensitive, cost-effective technique for reliably quantifying total mucin concentration in a robust mucin-secreting cell line for use in high throughput screening of small molecule compounds with potential to inhibit mucin secretion.

Methods: An enzyme-linked lectin assay (ELLA) was adapted using a lectin from Helix Pomatia (HPA), which binds specifically to N-acetylgalactosamine residues. Bovine sub-maxillary mucin was used for test development and later, as a mucin standard. The test sample was "sandwiched" between HPA pre-coated onto 96-well microtiter plates and biotinylated HPA lectin. The plate was developed with a fluorescent HRP substrate, and measured at 590 nm using a standard fluorescence plate reader. Using the ELLA, a number of carcinoma (HM3, A431, LS174, NCI-H292) and immortalized lung epithelial (16HBE, CFSME) cell lines were tested to determine their responsiveness to various secretagogues. Preliminary experiments indicated that the NCI-H292 cells, derived from a human mucoepidermoid lung carcinoma, were most suitable for adaptation to high throughput studies. NCI-H292 cells were seeded into 96-plates at a density sufficient to produce confluence in approximately 48 hours. Then, individual wells were incubated for 1 hour in either control or secretagogue-containing media. Supernatants were collected and mucin concentration of individual wells was determined using the ELLA.

Results: Optimization of the ELLA provided a sensitive calibration curve from 150 ng/mL to 2.5 mg/mL. Mucin secretory ratios (SR) were calculated by comparing mucin release from test and control wells. Significant (p ≤ 0.05) secretory ratios were obtained following incubation with UTP at 400 µM (3.17 ± 0.64), VIP at 300nM (3.60± 1.41), Carbachol at 2.4 mM (2.17± 1.2), Interleukin-1β at 5 ng/ml (2.74± 0.47) and Heparin-binding EGF-like growth factor at 90 ng/mL (2.49± 0.53); all data expressed as SR ± SEM, n = 6. Histamine at 54 µM did not result in significant mucin secretion.

Conclusion: A relatively simple and economical assay has been developed for high throughput screening of small molecules for detecting their effects on mucin secretion. Cystic Fibrosis (CF) is the most common lethal genetic disease in North America, where it occurs with a frequency of one in ~2200 live births. It is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene, the most common being the deletion of phenylalanine 508 (delF508). The delF508 mutation causes a folding defect that inhibits maturation and trafficking to the plasma membrane, however the mutant still functions as a chloride ion channel when delivered to the plasma membrane, and partial rescue (15-20%) may be sufficient to alleviate disease symptoms. Thus CFTR is an attractive target for the development of pharmaceutical therapies.

We recently described a novel, cell-based assay for protein trafficking which directly monitors expression of delF508 CFTR on the surface of baby hamster kidney (BHK) cells and is suitable for high throughput screening of small molecule libraries. In this study we have extended the approach to natural products by studying 720 extracts derived from marine sponges that were collected from the coasts of Indonesia and Papua New Guinea. Sponges are a particularly rich source of novel bioactives and secondary metabolites. Iterative use of our high throughput assay and chemical deconvolution of active fractions yielded four alkaloids JMCG1102, JMCG1103, JMCG1104 and JMCG1105. These compounds caused a detectable increase in delF508 CFTR trafficking to the plasma membrane within 6 h, and after exposure to 10 pM for 24 h all four increased surface delF508 CFTR expression 10-30% and enhanced cAMP-stimulated halide conductance. We tested a series of analogues of the alkaloids and found three others JMCG1106, JMCG1107 and JMCG1108 that also caused trafficking correction and functional ion channels in the plasma membrane. These results were confirmed in functional studies of the human bronchial epithelial cell-line CFBE stably overexpressing delF508 CFTR. These compounds increased surface delF508 CFTR by 15-80% within 6 h when treated with 1nM. These results identify a promising group of delF508 CFTR trafficking correctors that are now being actively pursued as potential pharmacotherapeutics for cystic fibrosis along with other hits from compound library screening campaigns. Work supported as part of the BREATHE program jointly funded by the CCFF and CIHR, and grants from Cystic Fibrosis Foundation Therapeutics, (CFFT).

Kim Chiaw, P. 1,2 ; Bear, C.E. 2 Background/Rationale: Biosynthetic misprocessing of the major CF mutant ∆F508-CFTR has been proposed to result from the disruption in domain-domain interactions (Du et al. Nature Struct and Mol Biol, 2005) that may lead to the exposure of di-arginine (RXR) retention/retrieval motifs. Simultaneous site directed mutagenesis of all four RXR motifs to RXK or KXR motifs overcame retention and promoted cell surface expression of ∆F508-CFTR (Chang et al. Mol Cell, 1999) . We hypothesize that it is possible to compete inhibitory interactions with RXR motifs exposed in ∆F508-CFTR by delivering synthetically derived CFTR peptides bearing one or a combination of RXR motifs (CF-RXR), each conjugated to cell penetrating peptides (CPP) to permit intracellular delivery.

Results: Confocal immunofluorescence microscopy confirms that Alexa-488-tagged CPP-CF-RXR peptides can be taken up rapidly (within 10 min) by BHK cells stably expressing recombinant human ∆F508-CFTR. Fluorophore-conjugated peptides appear to be concentrated in vesicular compartments, consistent with the current hypothesis that CPP-peptides enter cells into endosomal vesicles via macropinocytosis from which efflux into the cytosol occurs (Wadia et al. Nature Med, 2004) . We determined that treatment of cell cultures with CPP-CF-RXR peptide at a final concentration [1 µM] for 90 minutes is effective in inducing cell surface functional expression of human ∆F508-CFTR in BHK cells as observed by a 2.3-fold increase in cyclic AMP-mediated iodide efflux over control ∆F508-CFTR (n=6; p < 0.01). This rescue is specific to the CF-RXR peptide as controls (CPP alone or peptides bearing a KXK motif in place of RXR) show no functional rescue of ∆F508-CFTR (n=6; p > 0.05). Preliminary cyclic AMPmediated iodide efflux results also indicate that ∆F508-CFTR functional rescue by CPP-CF-RXR peptide is as effective as temperature (27 o C) rescue. Although we hypothesize that the peptide acts to facilitate trafficking of the major mutant to the cell surface, immunoblotting studies fail to show a clear conversion of the ER-form (Band B) to the fully processed, Band C form of the protein. Currently, we are developing more sensitive assays of cell surface expression as well as evaluating the possible efficacy of this peptide in potentiating function of ∆F508-CFTR.

Discussion: Future efforts include investigating the relative efficacy of peptides corresponding to the four different RXR motifs in rescuing ∆F508-CFTR trafficking. The promising candidate described above is currently being evaluated with respect to functional rescue of the major mutant in primary human bronchial epithelial cells obtained from patients homozygous for ∆F508-CFTR (Joseph Zabner and Phil Karp, University of Iowa) and in Cftr-∆F508 mice. These studies may provide the basis for future pre-clinical studies of peptide based therapies of Cystic Fibrosis (supported by CCFF (BREATHE Program Grant) and CIHR -(Strategic Training Program in the Structural Biology of Membrane Proteins Linked to Disease)). Several promising methods of promoting ∆F508 CFTR escape from ER degradation increase CFTR chloride channel activity at the cell surface. However, increasing data from a number of laboratories indicate that in both epithelial and non-epithelial cell lines expressing ∆F508 CFTR, the rescued chloride channel is unstable at the cell surface compared to wild-type CFTR. Small molecule correctors have therefore become a widely used approach to enhance ∆F508 CFTR function at the cell surface. In most cases, the exact mechanism of these promising compounds has not been established. In the present study, we use stably transfected, polarized CFBE41o-cells to elucidate the effect of two of these compounds, CFcor-325 and Corr-4a, on the surface stability of wild type and low temperature rescued ∆F508 CFTR. Specifically, we cultured cells expressing ∆F508 CFTR at 27C for 48 hours to establish a large pool of surface-expressed ∆F508 CFTR; we then monitored CFTR internalization rates and half-life with and without small molecule treatment using surface biotinylation-based assays, performed as described previously (Jurkuvenaite et al. J. Biol. Chem. 281:3329, 2006) . As a control, internalization and half-life of transferrin receptor was also followed. Our results indicate that both CFcor-325 and Corr-4a slow ∆F508 CFTR endocytosis from ~30%/2.5 min to ~5%/2.5 min, and this is similar to wild type CFTR internalization rates. Neither wild type CFTR nor transferrin receptor internalization rates were affected by corrector treatment, suggesting that the effects of the correctors were specific for ∆F508 CFTR. In the surface half-life studies, both correctors doubled ∆F508 CFTR halflife from 2 to 4 hours, but they did not correct the half-life to wild type levels (~8 h). Again, the correctors had no effect on either wild type CFTR or transferrin receptor surface half-lives. Our results suggest that small molecule correctors may not only rescue CFTR from ERAD during biogenesis, but can contribute to the stabilization of ∆F508 CFTR at the cell surface, and this effect seems to be ∆F508 CFTR-specific. The ∆F508 mutation in CFTR causes defects in chloride channel gating and cellular processing. Cell cultures models suggest the need for both a potentiator and corrector to restore ∆F508-CFTR channel function and processing, respectively. Previously, we identified several classes of potentiators, including 'phenylglycines', which probably bind to CFTR at the cell surface and restore normal channel gating (Pedemonte et al., Mol. Pharm. 2005; 67:1797 -1807 . We also identified several classes of correctors, including 'bisaminomethylbithiazoles', which may bind to ∆F508-CFTR in the ER and partially restore its folding and plasma membrane targeting (Pedemonte, et al., J. Clin. Invest. 2005; 115:2564 -2571 . These compounds were identified from screening a diverse collection of 150,000 small molecules; the most active potentiator was phenylglycine PG01 {2-[(2-1H-indol-3-ylacetyl)-methyl-amino]-N-(4-isopropylphenyl)-2-phenyl-acetamide} and the most potent corrector was the bisaminomethylbithiazole Corr-4a {N- (2-[5-chloro-2-methoxyphenylamino]-4'-methyl-4,5'-bithiazol-2'yl) benzamide}. Here we synthesized and characterized a hybrid molecule (Corr-4a-linker-PG01) containing both corrector and potentiator fragments. Based on SAR data of potentiators and correctors, a hybrid molecule was designed that incorporated an enzymatically hydrolysable linker to generate and deliver the potentiator and corrector-linker fragments. Using synthetic organic chemistry we synthesized a hybrid molecule containing a PG01-OH moiety and a Corr-4a-linker-COOH moiety, linked with an ethylene glycol spacer through an ester bond. The potentiator and corrector fragments (after cleavage) had low micromolar potency or better. However, the intact hybrid molecule was inactive, likely because of its large size (990 daltons) and hence poor cell penetration. As proof-of-principle evidence that the hybrid molecule can be hydrolyzed to the active fragments, treatment with carbonic anhydrase or rodent stool specimens gave the appropriate potentiator and corrector-linker fragments identified by LCMS. Smallmolecule corrector-potentiator hybrids have potential utility as single drug therapy for CF caused by the ∆F508 mutation. Supported by CFF and NIH. CFTR is a chloride channel gated by ATP binding and hydrolysis at its two nucleotide binding domains (NBD1 and NBD2). The two NBDs may dimerize in a head-to-tail configuration, as observed in other ABC transporters, forming two ATP binding pockets (ABP1 and ABP2), with the ATP molecules buried at the dimer interface. ABP2, formed by the Walker A and B motifs of NBD2 and the signature sequence of NBD1, was proposed as the site critical for the ATP-dependent opening of the CFTR channel, while ABP1 (consisting of the Walker A and B motifs of NBD1 and the signature sequence of NBD2) may contribute to the stabilization of the open channel conformation. G551D, the third most common CF-associated mutation, is characterized by a very low open probability, and patients carrying this mutation present a severe phenotype. This mutation is located in the signature sequence of NBD1 (and thus in ABP2). We have recently characterized this mutant channel and found that its low activity is ATP-independent. This behavior corroborates with the idea that the occupancy of ABP2 by ATP is crucial for catalyzing the opening of the channel. Interestingly, we now found that a high affinity ATP analog, N 6 -(2-phenylethyl)-ATP (P-ATP), increases G551D currents primarily by increasing the open time of the channel. This effect of P-ATP is reduced when P-ATP was applied together with ATP, suggesting a competition between ATP and P-ATP for a common binding site. Introducing the mutation Y1219G (located at ABP2), which reduces by more than 50-fold the ATP-binding affinity in wild-type channels, does not alter the effect of P-ATP in the G551D/Y1219G mutant. In contrast, when we introduce the mutation W401G (located at ABP1) in the G551D background, the effect of P-ATP is reduced remarkably, suggesting that ABP1 is the binding site where P-ATP binds to increase the activity of G551D channels. These new results further confirm the idea that nucleotide binding at ABP1 could stabilize the open channel conformation. Since the competition experiments show that ATP and P-ATP are readily exchanged in the binding site, we conclude that the ATP molecule is not occluded in ABP1, at least under the G551D background. Our observation that P-ATP enhances the G551D activity by binding at ABP1 implicates that ABP1 can potentially be a target for drugs to bind and increase the channel activity.

Supported by CFF grant to S. Bompadre and NIH grants to T.-C. Hwang (HL53445R01, DK55835R01) and S. Bompadre (DK75408K01).

Van Goor, F.; Hadida, S.; Grootenhuis, P. Vertex Pharmaceuticals, San Diego, CA, USA Several CF-causing mutations, including ∆F508, G551D, and R117H, cause gating defects in CFTR that decrease the open probability (Po) of the channel, resulting in decreased Cl-secretion across epithelia of multiple organs, including the lung. A therapeutic strategy for the treatment of CF is to develop pharmacological potentiators that increase Po leading to increased fluid transport in affected tissues. VX-770 was identified through cell-based screening and chemistry optimization as a potent potentiator of CFTR, including ∆F508-, G551D-, and R117H-CFTR. In single-channel recordings of CFTR expressed in Fisher rat thyroid cells, VX-770 increased the Po of ∆F508-CFTR from 0.11 ± 0.05 to 0.45 ± 0.06. The Po of G551D increased from 0.08 ± 0.03 -0.52 ± 0.19. These levels are similar to the Po of wild-type CFTR, indicating that VX-770 restores normal gating at the single channel level. To assess the potency and efficacy of VX-770 in native airway cells, Cl-secretion was monitored in Ussing chamber studies using human bronchial epithelia (HBE) cultured from the bronchi of CF and non-CF donors. In the absence of the potentiator VX-770, CFTR-mediated Clsecretion in ∆F508/∆F508or G551D/∆F508-HBE reached a maximum amplitude of 2.6 ± 0.5 and 2.7 ± 0.2 uA/cm2, respectively. This is ~5% of that observed in wild-type-HBE (56 ± 6 uA/cm2) and consistent with the low level of residual CFTR activity reported in some individuals with severe CF disease. In ∆F508/∆F508-HBE, VX-770 caused a 2-fold increase (peak response; 5.5 ± 0.5 uA/cm2) in CFTR mediated Cl-secretion with an EC50 of 33 ± 8 nM, whereas in G551D/∆F508-HBE, Cl-secretion increased 14fold (peak response; 36 ± 3 uA/cm2) with an EC50 of 600 ± 200 nM. The improved efficacy observed in G551D/∆F508-HBE is consistent with a higher cell surface density of G551D-CFTR compared to ∆F508-CFTR, which has defective trafficking to the cell membrane. VX-770 was found to be highly selective for CFTR and to have good oral bioavailability in multiple species with a half-life of 6 -14 hours. The biological and pharmacokinetic data support the clinical development of VX-770 for the treatment of cystic fibrosis. We would like to acknowledge the CFFT for their support and Dr. Joseph Pilewski for providing CF HBE. The PARI eFlow ® rapid nebulizer is a member of the "soft mist" inhaler class for the administration of aerosolized antibiotics that eliminates the need for a compressor.

Objectives: To compare the pharmacokinetics and safety of tobramycin inhalation solution (TOBI ® ) delivered via the PARI eFlow ® rapid electronic nebulizer vs the PARI LC ® Plus jet nebulizer with compressor.

Methods: We compared tobramycin levels in sputum and serum, incidence of bronchospasm and frequency of adverse events (AE) of 300 mg TOBI, administered twice-daily for 2 weeks from these two delivery systems in a randomized, crossover study in 25 cystic fibrosis patients. Blood and sputum samples were collected over a dosing interval after the first and last dose on each device (days 1 and 15). Tobramycin serum and sputum area under the concentration-time curves (AUC) were derived and the therapeutic ratio (mean sputum AUC / mean serum AUC) calculated. Pulmonary function tests were performed before and 30 min after nebulization on days 1 and 15, and AE were recorded.

Results: The total nebulization time was reduced by half for the PARI eFlow rapid vs PARI LC Plus (7.3 ± 1.7 min vs 17.7 ± 3.8 min [p<0.0001] on day 1, and 8.0 ± 1.8 vs 16.3 ± 2.7 [p<0.0001] on day 15, respectively). Tobramycin predose serum levels >2 µg/ml or Cmax >12 µg/ml, predefined as a potential for increased risk of systemic toxicity, were not exceeded in any patient during use of either nebulizer. As tabulated below, mean systemic exposure to tobramycin from the PARI eFlow rapid on day 15 was slightly lower, by 14%, whereas mean sputum exposure was higher, by 80%, compared with the PARI LC Plus. Consequently, the therapeutic ratio was nearly 2-fold higher for the PARI eFlow rapid due mainly to higher sputum concentrations. Adverse events occurred in 16 patients using the PARI LC Plus nebulizer (primarily headache and abdominal pain) and in 19 patients using the PARI eFlow rapid nebulizer (primarily headache, cough, dyspnea). On day 15, no clinically relevant bronchospasm (defined as ≥20% decrease in FEV 1 ) was observed for either treatment. The mean percentage change in FEV 1 at 30 mins from predose on day 15 was -2.76% for the LC Plus and -1.27% for the eFlow rapid.

Conclusions: TOBI delivered via the PARI eFlow rapid electronic nebulizer required shorter delivery times with similar safety and a higher therapeutic ratio (sputum/serum exposure) compared with the PARI LC Plus jet nebulizer.

Sponsored by Novartis Pharma AG which acknowledges support from PARI GmbH INTRODUCTION Alpha-1 antitrypsin (AAT) is a protein protecting lung tissues by inhibition of neutrophil elastase. The function of neutrophil elastase is to digest bacteria and other foreign objects in the lungs. In a person deficient of AAT, the neutrophil elastase acts uncontrolled, destroying healthy tissue. The result of the damage is emphysema, which progresses if not treated. Patients show a specific impaired breathing pattern of short inhalation followed by prolonged exhalation. Current therapy consists of weekly i.v. infusion of AAT of about 60-90 mg/kg body weight. By this method only about 2% of the administered dose is estimated to reach the lungs. Since AAT can currently be derived only from human blood serum by an expensive purification process, the worldwide supplies are limited and do not allow to treat all patients. An improvement of the delivery efficiency is highly desirable. Inhaled treatment would offer a targeted therapy by delivering AAT directly to the lungs and allow for a better use of the limited drug. The eFlow ® , a novel electronic nebulizer, is more efficient than conventional nebulizers. It operates by means of a perforated vibrating membrane, applying no or only little stress to the nebulized substances. Hence it is well suited for the pulmonary delivery of proteins. It has been shown previously, that a 2% AAT solution can be nebulized by eFlow ® without a loss of biological activity. We investigated two eFlow ® configurations comparing the in-vitro delivered dose for different breathing patterns.

METHOD A highly purified 2% AAT preparation (Kamada Ltd, Rehovot, Israel) was aerosolized with the eFlow ® electronic nebulizer. Initial studies were conducted using the 30L configuration of eFlow ® . Delivery performance is compared to a specially designed 30XL configuration, with increased aerosol chamber volume. Aerosol delivery efficiency was determined by breath simulation using an emphysema breathing pattern (tidal volume 450 ml, 17 breaths/min, inh:exh ratio = 1:2.5) generated by a PARI Compas™ breath simulator. As reference, a standardized regular breathing pattern (tidal volume 500 ml, 15 breaths/min, inh:exh = 1:1) was also investigated. The Respirable Fraction (drug in droplets < 5 and 3.3 µm) was determined using the Andersen Cascade Impactor at 28.3 L/min. Samples were analysed for AAT activity by an elastase inhibition assay.

When the eFlow ® 30L configuration was investigated for the standardized breathing pattern, a delivered dose (DD) of 70 ±3% of the loaded dose (92mg) was found. This value was reduced to 50 ±1% when the emphysema pattern was applied. With the new eFlow ® 30XL configuration an improved DD of 81 ±3% was obtained. Furthermore, when the emphysema pattern was applied, the DD did not drop strongly, but remained at values above 68%. With 83% of aerosol droplets in the respirable size range (<5µm) and 42% <3.3 µm, a high % of AAT can be expected to reach both, the central and peripheral lungs of patients.

CONCLUSION Inhaled AAT has the potential to significantly improve the efficiency of AAT replacement therapy. A customized eFlow ® electronic nebulizer produced delivered doses > 60% even when breathing patterns of patients with impaired breathing capabilities were mimicked. The formulation is optimized for rapid aerosol administration through use of higher drug concentrations and is taste-masked. This study was conducted to evaluate the aerosol performance of novel higher dosed levofloxacin inhalation solutions in a customized eFlow ® electronic nebulizer.

METHODS Two concentrations of MP-376 (50 mg/ml and 30 mg/ml of levofloxacin) were nebulized at fill volumes of 2 and 4 ml using the eFlow ® 35L nebulizer configuration. The in-vitro Delivered Dose (DD) was determined by breath simulation using a standardized breathing pattern (tidal volume 500 ml, 15 breaths/min, inh:exh ratio = 1:1) generated by a PARI Compas™ breath simulator. The geometric droplet size distribution was determined by Laser Diffraction (LD). The aerodynamic droplet size distribution was determined using the Andersen Cascade Impactor (ACI). The invitro Respirable Dose (RD) was calculated by multiplying the Respirable Fraction (RF, %droplets < 5µm) derived from the cascade impaction experiment and the DD derived from breath simulation. Nebulization time was determined by the electronic shut-off of the eFlow ® control unit. All tests were conducted for three devices in duplicate, each (n=6).

The eFlow ® electronic nebulizer delivered MP-376 at both concentrations and fill volumes with equal efficiency, obtaining in-vitro DDs of around 60±5 %. The average nebulization times were between 3 and 4 minutes for the 2 ml fill volume and between 7 and 8 minutes for the 4 ml fill volume and were also independent of the concentration in the investigated range. Determination of the Mass Median Diameter (MMD) by LD resulted in values between 3.5 and 4.1 µm. Values of the Mass Median Aerodynamic Diameter (MMAD) obtained by cascade impaction were similar (between 3.7 and 4.3 µm). The average Respirable Fraction was in the range between 70% and 75% for both experimental methods.

CONCLUSION The eFlow ® electronic nebulizer can be customized to deliver high doses of MP-376. The short nebulization times achieved are patient friendly and should support high patient compliance. MP-376, with the potential for once daily administration, provides significant advantages over currently avail-able aerosol antibiotics. Clinical evaluation of MP-376 delivered by the eFlow ® nebulizer in CF patients is in progress.

LaVange, L. 1 ; Engels, J. 2 ; Accurso, F.J. 3 1. University of North Carolina, Chapel Hill, NC, USA; 2. Inspire Pharmaceuticals, Inc., Durham, NC, USA; 3 . University of Colorado, Denver, CO, USA Introduction: Percent change from baseline in a continuous outcome variable, such as lung function, is a useful descriptive measure in therapeutic clinical trials. While often favored by clinicians as the primary efficacy measure, the use of percent change as the basis for statistical comparisons raises a number of issues. Defined as 100*(Post-Pre)/Pre, where Pre and Post represent baseline and follow-up values, respectively, percent change is a ratio of random variables and as such, does not follow a normal distribution. Analyzing percent change with standard methods can result in inefficiences and may be inappropriate without a suitable transformation (e.g., logarithmic). Furthermore, the treatment effect estimated on the transformed scale is difficult to interpret. The purpose of this abstract is to describe an alternative method for analyzing percent change and illustrate its utility in CF clinical trials.

Methods: Probability plots are generated to compare the distributions of percent change in FEV 1 from baseline to study endpoint among treatment groups. The plots are similar to Kaplan-Meier curves, with percent change on the horizontal axis in lieu of survival time. The percentage of patients on the vertical axis depicts the cumulative percentage of patients who had a percent change at least as great as the corresponding value on the horizontal axis. A log-rank test statistic provides an overall test for any difference in distributions among treatment groups. The methods are illustrated using data from a 28-day, phase 2, multicenter, randomized, double-blind, placebo-controlled clinical trial of denufosol in CF patients with a screening FEV 1 ≥75% predicted.

Results: A probability plot of denufosol (active doses combined, n=68) vs. placebo (n=21) is shown below. More denufosol patients (43%) experienced improvement compared to placebo (33%). The log rank test, adjusted for study site, showed that the overall distributions of percent change in FEV 1 between denufosol and placebo were significantly different (p-value=0.029).

Conclusions: The proposed methods provide a convenient means for assessing differences in percent change and may be useful in CF clinical trials. Comparisons of both a descriptive and inferential nature can be made with minimal assumptions, thereby avoiding the pitfalls in analyzing percent change with standard techniques. The methods are easy to implement with existing software packages (e.g., SAS).

Trial supported by Inspire Pharmaceuticals, Inc. and CF Foundation Therapeutics, Inc.

Chen, X. 1,2 ; Davis, P. 1 Compacted DNA nanoparticles formulated with PEGylated polylysine and plasmid DNA transfect airway cells in vivo at high efficiency, so they have great potential for gene therapy. The mechanism by which DNA nanoparticles enter the cell and get expressed is not well understood. We previously showed that rhodamine labeled DNA nanoparticles enter primary human tracheal epithelial (HTE) cells within 15 min, and accumulate in the nucleus in 1 hr, where the transgene can be expressed. DNA nanoparticles do not enter via clathrin-mediated endocytosis (CME), which leads to the degradation of internalized material in lysosomes. Further characterization of the kinetics and trafficking pathway may allow us to improve the formulation of the nanoparticles and facilitate its gene delivery. As they do in HTE cells, rhodamine labeled DNA nanoparticles enter HeLa and 16HBEo-, an immortalized human bronchial epithelial cell line, within 15 min, and by 1-4 hr accumulate in the nucleus, especially the nucleolus, where they colocalize with nucleolin. Expression of the GFP reporter gene in the nanoparticles was observed at 18 hr, which confirms the functionality of the labeled DNA nanoparticles. Cellular entry is energy dependent, as little or no intracellular rhodamine was observed at 4°C by 4 hr. We also applied rhodamine labeled nanoparticles and biotin-conjugated transferrin, a marker for CME, simultaneously. Little colocalization was observed during a 15 min to 4 hr time course, so the nanoparticles do not follow this pathway.

We then found a direct and strong relationship of cell surface nucleolin and the ability of the cells to take up and express DNA nanoparticles. Nucleolin directly binds to DNA nanoparticles with KD=25.9 nM. Manipulations of cell surface nucleolin in HeLa cells affect transfection of DNA nanoparticles with a positive correlation. We therefore costained nucleolin in both HeLa and 16HBEo-cells treated with the rhodamine labeled DNA nanoparticle. The rhodamine fluorescence colocalizes with nucleolin both in the cytoplasm at early time points and in the nucleus/nucleolus at 1 hr and 4 hr, which further supports that nucleolin might be associated with the nanoparticles during their trafficking. To confirm this association, we performed uptake experiments with MS-3, a mouse monoclonal antibody against nucleolin, alone or with rhodamine labeled nanoparticles over a 15 min to 4 hr time course. The MS-3 antibody enters HeLa cells at similar kinetics to the nanoparticles. At 30 min, we observed substantial amount of cytoplasmic staining, while nuclear staining tends to increase by time until 4 hr. It has little or no colocalization with transferrin in the cytoplasm, which suggests that this antibody also enters the cell via a pathway other than CME. In contrast, it showed extensive colocalization with DNA nanoparticles in both cytoplasm and nucleus at all time points, which suggests that cell surface nucleolin and DNA nanoparticles are associated or at least in close vicinity during their trafficking inside the cell. Therefore, we suggest that DNA nanoparticles enter the cell by binding to cell surface nucleolin and enter the cell via the same pathway as nucleolin antibody.

Chen, X. 1,2 ; Davis, P. 1 DNA nanoparticles are non-viral gene delivery vectors in clinical trial for treating genetic disorders including Cystic Fibrosis. We previously discovered that cell surface nucleolin serves as a receptor for the DNA nanoparticles, and is important for their gene delivery efficiency. As nucleolin has no signal sequence or membrane-spanning domain, it is not clear how nucleolin are expressed on the outer surface of plasma membrane or what signals increase its surface delivery. We initially observed that the transfection efficiency of DNA nanoparticles in HeLa cell decreases by 72.2% following 24hr serum-free medium treatment, which reduces cell surface nucleolin by 35.7% as determined by cell surface biotinylation and streptavidin bead pulldown. Since serum depletion inhibits cell proliferation and affects cell cycle progression, we examined the level of cell surface nucleolin at different stages of cell cycle. HeLa cells were arrested at S phase by high concentration of thymidine, and allowed to progress synchronously through cell cycle after removing this block. Surprisingly, we observed substantial increase of cell surface nucleolin at the onset of M phase, about 8 hours after release of thymidine block. It has been reported that nucleolin is phosphorylated by a cell cycle dependent kinase (Cdk) cdc2, and has eight consecutive Cdk phosphorylation sites. Therefore it is appealing to speculate that phosphorylation by cdc2 might increase the targeting of nucleolin to the cell surface.

We then determined the potential export signal in nucleolin by serial deletions. Deletion of C-terminal 584 aa of nucleolin, including the C-terminal glycine/arginine rich (GAR) domain and the four RNA recognition motifs (RRM) does not affect its arrival at the cell surface. The N-terminal 123 aa of nucleolin is sufficient to target a GFP protein to the cell surface. In contrast, when we further delete the 8 Cdk phosphorylation sites from aa 69 to 123, cell surface expression of nucleolin is significantly diminished. Furthermore, nucleolin lacking the N-terminal 68 aa is not present on the cell surface. Therefore, phosphorylation on the Cdk sites may serve as a signal to enhance the cell surface expression of nucleolin. There has been recent interest in dry powder inhaled mannitol as a therapeutic agent in patients with cystic fibrosis (CF). It is has been shown to increase mucociliary clearance (MCC) by airway rehydration. Whilst there has been one short term clinical trial of mannitol in adult subjects [1] , to date no studies have been conducted looking at its potential as a therapy in children with CF. It could be argued that children may have the potential to benefit most from a therapeutic agent that acts relatively proximally in the CF pathogenic pathway. The aim of this study was to determine acute tolerability of inhaled mannitol in children with CF.

We recruited 39 children (aged 8 to 18 years) with CF. Inclusion criteria were either rhDNase treatment or an FEV 1 > 40% and < 70% of mean predicted normal value. 39 bronchial provocation challenges with incrementally increasing doses (5, 10, 20, 40, 80, 160, 160mg) of dry powder mannitol were carried out. Subjects were pre-treated with bronchodilator 15 minutes prior to the challenge (400 mcg of salbutamol or 1 mg of terbutaline). FEV 1 was measured following each dose increase up to a maximum cumulative dose of 475mg. Oxygen saturation monitoring was carried out throughout. A positive challenge was defined by a drop in FEV 1 of >15% from baseline. These subjects received bronchodilator treatment and had spirometry repeated at 15 minute intervals until FEV 1 returned to within 5% of baseline. Those children with a negative challenge had spirometry repeated 15 minutes post-completion of the challenge Results Mean baseline FEV 1 was 68% predicted (45-94, SD 11.7). 9/39 subjects (23%) had a positive challenge. Mean PD 15 (dose of mannitol required to cause a 15% reduction in FEV 1 ) was 262mg (411-107, SD 104). The mean time to complete the challenge was 26 minutes (17-37, SD 5.0) for negative challenges and 27 minutes (15-40, SD 8.0) to PD 15 for those subjects with a positive challenge. We found no association between a positive challenge and age, sex, weight, height, baseline FEV 1 , Pseudomonas aeruginosa colonization, bronchodilator reversibility, previous Aspergillus fumigatus sen-sitization, atopy or corticosteroid use. There was a non-significant trend for lower FEF 25-75 (means 0.99 versus 1.40; p=0.099) and higher prevalence of Aspergillus fumigatus cultured in sputum at baseline (5/9 versus 7/30 children; p=0.066) amongst those children who went on to have a positive challenge. There was no significant drop in oxygen saturation in either group. Although cough was common during the challenge, other adverse events were infrequent.

We find that 23% of children with CF could not tolerate inhaled mannitol as compared with 12% of adult subjects as reported in the previous study [1] . We could not identify factors predictive of a positive mannitol challenge in these patients. The most common cause of cystic fibrosis (CF) is the deletion of phenylalanine 508 (∆F508) in the CF transmembrane conductance regulator (CFTR) chloride channel [1] . A major problem with ∆F508 CFTR is that the protein is defective in folding so that little mature protein is delivered to the cell surface [2] . Expression of ∆F508 CFTR in the presence of small molecules known as correctors or pharmacological chaperones can increase the level of mature protein [3] [4] [5] [6] . Unfortunately, the efficiency of correctorinduced maturation of ∆F508 CFTR is low and other approaches are needed to increase the therapeutic value of correctors. We postulated that expression of ∆F508 CFTR in the presence of multiple correctors that bound to different sites in the protein may have an additive effect on maturation. In support of this mechanism, we found that expression of P-glycoprotein processing mutants (CFTR's sister protein) in the presence of two compounds that bind to different sites (rhodamine B and Hoechst 33342) had an additive effect on maturation. Therefore we tested whether expression of ∆F508 CFTR in the presence of combinations of three different classes of corrector molecules would increase its maturation efficiency. It was found that the combination of the quinazoline VRT-325 together with the thiazole corr-2b or bisaminomethylbithiazole corr-4a doubled the steady-state maturation efficiency of ∆F508 CFTR (about 40% of total CFTR was mature protein) compared to expression in the presence of a single compound. The additive effect of the correctors on ∆F508 CFTR maturation suggests that they may directly interact at different sites of the protein. The use of multiple correctors has the potential to increase the therapeutic value of pharmacological chaperones. Pharmacol. 70, 297-302. The CFTR protein is expressed at the surface of the airway epithelium, where it plays a critical role in maintaining airway hydration by secretion of chloride. For gene therapy of CF lung disease to be successful, it is critical that the endogenous transgene be expressed in the correct cell type and at levels sufficient to restore normal function. The expression of high levels of CFTR has resulted in CFTR mediated chloride secretion in a wide range of experimental systems. However, the cell type or types that need to be corrected and the level of CFTR expression required to ameliorate the pulmonary manifestations of CF are not yet clear. Because CFTR has been localized to the ciliated cells of human airway epithelial cells, we have been investigating the use of ciliated cell-specific promoters to improve the efficiency and safety of gene therapy for CF and other airway diseases. In previous studies a fragment of the human FOXJ1 promoter was shown to target transgene expression specifically to ciliated cells. However, expression of CFTR from this promoter in transgenic CF mice did not significantly improve the CF phenotype, as measured by the nasal potential difference (PD) technique (Ostrowski et al, 2007) . This may be due to the inability of this promoter to correct the olfactory epithelium, which comprises approximately 50% of the mouse nasal cavity. In CF mice, the olfactory epithelium exhibits sodium hyperabsorption and an absence of CFTR mediated chloride secretion, similar to the respiratory epithelium (Grubb et al, 2007) . In this work, we used a fragment of the ciliated cell-specific promoter FOXJ1 to drive CFTR expression specifically in human ciliated cells. Replication deficient adenovirus expressing either EGFP or CFTR from the FOXJ1 promoter were used to transduce well-differentiated cultures of human CF cells following treatment with C10 to disrupt tight-junctions. Cultures were studied 48 hours after treatment, and those treated with Ad-FOXJ1/EGFP showed strong expression of EGFP that was dependant on ciliated cell differentiation. RNA analysis demonstrated strong expression of CFTR from the FOXJ1 promoter, and Western blotting demonstrated levels of protein that were much greater than the level in normal airway cells (>100-fold). Immuno-localization of CFTR with specific antibodies resulted in a strong signal at the apical membrane of ciliated cells. Cultures treated with Ad-FOXJ1/CFTR demonstrated a significant increase in forskolin-stimulated short circuit current (Isc; 5.0 +/-0.8 µA/cm2, mean +/-sem, n=9), that was approximately 8-fold greater than the response in Ad-FOXJ1/EGFP treated cultures (0.6 +/-0.2 µA/cm2, n=9). The increase in Isc was blocked by INH172, an inhibitor of CFTR mediated secretion. Under these conditions, Ad-FOXJ1/CFTR restored approximately 30% of the forskolin response of normal human airway cells (15.1 +/-0.3 µA/cm2, n=3). Because patients expressing low levels of normal CFTR mRNA (5-20%) have mild disease symptoms, these studies demonstrate that the incorporation of the ciliated cell-specific FOXJ1 promoter into gene therapy vectors may be useful for treatment of CF.

Supported by NHLBI RO1 HL70199 and the CFF. There is a compelling need for safe and effective AI therapy for CF lung disease. Low-dose methotrexate (MTX) has been used to treat inflammatory diseases and its use has been reported in CF. Encouraging results previously described regarding MTX in CF prompted this study. The study objective was to determine if MTX safely reduces inflammation in the airways of CF patients.

Methods: This was a single-center, open label study of MTX in stable CF patients with mild to moderate lung disease. Baseline levels of neutrophils, free elastase, pro-inflammatory cytokines, and bacteria were determined from an induced sputum specimen. Subjects received 15 mg/m2/week of oral MTX (single dose). After 12 weeks of treatment, a second induced sputum specimen was obtained for the same inflammatory indices. Within subject comparisons (end of treatment vs baseline) were performed for the following primary endpoints: total white cell and neutrophil counts, percent neutrophils, and concentrations of free elastase, IL-8, IL-6, TNF-α, and IL-1β. Sputum quantitative microbiology was obtained at baseline and end of therapy. Routine laboratories and spirometry were performed monthly. Pharmacokinetic testing was performed on the final day of treatment.

Results: Thirteen subjects were screened with 8 started on MTX. Six completed the protocol and 2 withdrew early secondary to adverse events (gastrointestinal [GI] complaints and pulmonary exacerbation). Five of 6 subjects completing the protocol had declines in pulmonary function: mean change in FVC was -0.257 L (range -0.47 to -0.09); mean change in FEV1 was -0.15 L (range -0.29 to -0.06); change in FEF25-75 was 0.01 L (range -0.25 to 0.22). Mean change in weight was -2.7 pounds (range -12 to 7.5). Mean change in free elastase was 62.4 µg/ml (range -68.8 to 406.4; sd 177.0). Mean change in IL-8 was 121511 pg/ml (range -122248 to 238400; sd 121511). Similar results were obtained for the other cytokines. Analysis of cytology specimens is pending. ESR and CRP did not show significant changes. Two subjects completing the protocol had significant GI side effects, including 1 requiring admission for severe abdominal discomfort. Quantitative microbiology specimens revealed an increase in colony counts in 2 subjects, decreases in 3, and mixed results in the 6th. Other safety laboratories were not remarkable. Pharmacokinetic studies are pending. Complete statistical analysis and safety assessment are also underway.

Conclusions: The small sample size of this study precludes definitive conclusions regarding the safety or efficacy of MTX. However, the data suggest that 1) induced sputum inflammatory indices can be used to assess an AI in clinically stable patients, and 2) MTX therapy may not be beneficial, and may be difficult to tolerate as along-term AI therapy for CF lung disease. Additional analyses of cell counts and microbiology from this study are ongoing, and may provide additional insight into the effect of MTX in the CF lung.

Sponsored by the CF Foundation , a cellular component of many gram negative bacteria (e.g. Pseudomonas aeruginosa), is a common airway stimulus. Dampening the inflammatory response in CF reduces the progressive decline in lung function, so anti-inflammatory agents have become both a cornerstone of CF clinical care and a focus of therapeutics research development. Reactive oxygen species [ROS] can play a role in proinflammatory signaling, including the activation of nuclear factor κB [NFκB] . Research has demonstrated increased oxidative stress in CF epithelial cells, potentially highlighting one of the mechanisms responsible for the exuberant inflammation observed. We have previously shown that CDDO, a novel anti-inflammatory agent, significantly limits NFκB activation in CFTR deficient cell culture models at nanomolar concentrations. Using CF mice, we have also shown that intra-tracheal CDDO limits neutrophil accumulation and the concentrations of proinflammatory cytokines and chemokines in bronchoalveolar lavage [BAL] fluid in response to LPS. The synthetic triterpenoids have been shown to activate the nrf2 transcription factor, thereby inducing several genes involved in redox balance. We now present data demonstrating that CDDO may inhibit inflammation in models of CF pulmonary disease by reducing the oxidative stress within cells.

Methods: B6.129S6-Cftr tm2Mrc CF mice received 25µl of 10µM CDDO in PBS with 3%DMSO or vehicle (control) daily. Drug was administered intratracheally with an atomizer (PennCentury) for three days before stimulation. In separate experiments, all mice received 1µg LPS or free Pseudomonas aeruginosa intratracheally. Mice were sacrificed by CO2 asphyxiation and cardiac puncture 20 hours after stimulation and BAL, serum and lung tissue were obtained. The lungs were incubated in lysis buffer containing protease inhibitors and immediately frozen. Upon thawing, the tissue was homogenized and whole lung protein extracted. One mg of lung protein from each experimental animal was separated by two dimensional gel electrophoresis. Gels were then imaged and analyzed with PDQuest to identify protein spots that were differentially expressed in drugtreated mice compared with controls. Spots of interest were excised from the gels, trypsin digested and analyzed by liquid chromatography/tandem mass spectroscopy.

Results: Over 120 proteins with greater than 2-fold differences in expression were identified. We categorized proteins appearing in repeated analyses and identified several affecting intracellular redox regulation; including glutathione S-transferase mu1 and 2, peroxiredoxin 5 and 6, enolase 1 (alpha non-neuron), contrapsin, alpha-1 antiproteinase and Cu/Zn superoxide dismutase. These proteins were expressed at significantly greater concentration in the lungs of mice treated with CDDO rather than vehicle.

Conclusion: CDDO exhibits anti-inflammatory effects in mouse models of CF pulmonary disease and one potential mechanism for this effect may be the upregulation of reducing proteins to combat oxidative stress. Methods: LPC (15 uL of 0.1%, 0.3% or 1.0% (w/v) in PBS) was administered (trans-orally via a cannula) to trachea of C57L/B6 mice. In sheep (1 of 5 completed) we targeted delivery of 4.5 mL of 0.3% LPC dissolved in PBS to the main bronchus of the right lung at branch 9 in a 3 month old sheep (bronchoscope, via tracheostomy). One hour later 30 uL (mice) or 0.45 mL (sheep) of a LV-LacZ vector (3.12x109 TU/ml) was administered in the same manner. One week post-exposure lungs were inflation-fixed in situ, removed, stained using X-gal, sectioned, and counterstained with Saf-O or H/E.

Results: No LacZ gene expression (blue cells) was found in lungs of control (PBS-treated) mice or in the left (untreated side) of sheep lung. Blue cells were found in scattered punctuate groups, or in lines of stained cells in mice and sheep. Mice given the highest LPC dose (1.0%) had extensive gene transfer in larynx, trachea, carina, and in large, middle and small airways of most lobes of the lung, reaching 79% airway perimeter cell transduction in one animal. With decreasing LPC dose the distribution and number of blues cells was reduced, but transduced epithelial cells including nonciliated columnar cells, ciliated cells, and basal cells were seen in all cases.

In sheep, blue cells appeared in the right main bronchus between branch 6 and branch 15, and in the branch 9 airway, and with highest expression found near branch 8 and 9. Cross-sections revealed that primarily ciliated columnar cells and basal cells (and no goblet cells or macrophages) were transduced.

Conclusions: LV gene transfer into mouse or sheep lung can be enhanced by pretreatment with LPC. In mice this early data suggests an LPC dose-dependency. For sheep, we await further results to confirm the encouraging finding in this first study; however, it does confirm that gene transfer into the airways of large animals with a lung size similar to humans can also be achieved using LPC pre-treatment and a VSV-G pseudotyped lentivirus. The transduction of both ciliated and basal lung epithelial cells in-vivo in both models is an encouraging funding for the development and understanding of our continuing efforts to produce life-long airway gene transfer suitable for a CF airway gene therapy.

Supported by: NH&MRC, USA CFF, SA Channel 7 CRF, Philanthropic donations. Increased airway Na+ absorption is a characteristic abnormality in cystic fibrosis (CF), and is thought to play a key role in the pathogenesis of CF lung disease. We have previously demonstrated that mimicking Na+ hyperabsorption by overexpression of βENaC in mouse airways results in airway surface liquid (ASL) volume depletion and reduced mucus clearance causing a spontaneous CF-like lung disease with high pulmonary mortality, and airway mucus plugging, mucous cell metaplasia and chronic airway inflammation in surviving βENaC-transgenic (βENaC-Tg) mice (Mall et al., Nature Med.10:487, 2004 ). In the present study, we used βENaC-Tg mice to test if inhibition of increased airway Na+ absorption by the classic ENaC blocker amiloride has therapeutic effects on CF-like lung disease in vivo. Specifically, we determined the effects of 'early' and 'late' ENaC blocker intervention on mortality, airway mucus obstruction and inflammation by starting intrapulmonary amiloride therapy in βENaC-Tg mice either at birth, i.e. prior to the onset of lung disease, or after mucus obstruction and inflammation had established. To achieve this goal, newborn, 5 day, or 4 week old βENaC-Tg mice and wild-type littermate controls were treated by intranasal instillation of amiloride (10 mmol/l; 1µl/g body weight, tid) or vehicle (H2O) alone for a period of 14 days. Initial deposition studies showed that this treatment protocol resulted in pulmonary amiloride concentrations sufficient for inhibition of ENaC. During amiloride therapy, growth and survival were monitored, and any loss in body volume due to diuretic side effects was substituted by subcutaneous injection of isotonic saline (NaCl 0.9%). After the 14 day treatment period, mice were euthanized, bronchoalveolar lavage (BAL) was performed to determine inflammatory cell counts, and lungs were processed for histology and morphometry to quantitate airway mucus obstruction and mucous cell metaplasia. We show that amiloride significantly reduced pulmonary mortality by~70% (P < 0.001), when therapy was started in newborn βENaC-Tg mice. Further, early amiloride treatment significantly reduced airway mucus content (P < 0.05), mucous cell numbers (P < 0.05), and BAL eosinophils (P < 0.001) compared to vehicle treated βENaC-Tg mice. In contrast, amiloride therapy had no benefits on airway mucus obstruction, mucous cell metaplasia or inflammation, if treatment was started in 5 day, or 4 week old βENaC-Tg mice with established lung disease. Taken together, our results are consistent with previous human studies where amiloride lacked therapeutic benefits in CF patients with established lung disease, and demonstrate for the first time that early inhibition of Na+ hyperabsorption is an effective therapy for CF-like lung disease in vivo. These results warrant an evaluation of more potent and longer acting amiloride derivatives in chronic lung disease in mice, and a clinical evaluation of preventive ENaC blocker therapy in human trials. Supported by CFF (MALL04G0) and EC (MEXT-2004-013666 Objective: Hypertonic saline aerosol delivered intranasally is currently being studied to enhance mucociliary clearance. There is evidence of patients' reluctance to use concentrations above 3% due to potential discomfort. Our study was performed to determine the short term tolerance of 3.5% and 7% hypertonic saline versus normal saline delivered intranasally via a nebulizer/compressor system (PARI SinuStarTM, PARI Respiratory Equipment, Midlothian, VA).

Methods: Using the SinuStarTM nasal aerosol delivery system, we administered 3 concentrations of saline solution, (0.9%, 3.5%, 7%) to 18 healthy, adult volunteers for 5 minutes each. A washout period of 5 minutes between treatments allowed volunteers to wipe their nose and cleanse their mouth with water. A 6 question self-administered questionnaire was completed following each treatment using a 9 point scale (1=high tolerability; 9=low tolerability). Burning sensation, cold, cough, throat irritation, runny nose, and overall comfort were measured. The order of treatments was randomized and volunteers were blinded to the concentration.

Data: For all measurements 3.5% and 7% were equally well tolerated compared to normal saline with burning, cough, and throat irritation measuring 2.2 or below. No variable measured more than 3.8. 17 out of 18 volunteers stated they would continue this treatment on a regular basis.

Conclusions: Our study indicates that during the time of treatment nasal aerosol delivery of hypertonic saline up to 7% is well tolerated in healthy adults. There appears to be no issue of discomfort associated with hypertonic saline that would prevent nasal aerosol treatment compliance. Our overall research goal is to discover new drugs for clinical treatment of cystic fibrosis (CF) that will correct the biochemical defect in the predominant CF mutation, the ∆F508 form of the cystic fibrosis transmembrane conductance regulator (CFTR) protein, which accounts for over 67% of all CF cases. This mutation leads to a misfolded protein which is rapidly degraded as well as changes in its function and half-life at the cell surface. However, it has been proposed that only a fraction of the normal surface expression level is needed to provide a clinically significant impact on the disease. Thus, from a pharmacological standpoint, strategies that can partially but effectively correct these defects would be expected to have a clear clinical benefit to these CF patients, and these strategies may also be applicable to other CFTR mutations. We previously demonstrated that several members of a particular class of related drugs, the anthracyclines, anthraquinones, and anthracenediones, significantly increased CFTR cell surface expression and function in cell culture. In particular, we demonstrated that a non-cytotoxic concentration of doxorubicin (Dox), a model anthracycline drug, significantly increased total cellular and membrane-associated CFTR protein levels and CFTR-associated chloride currents in human colon cancer T84 cells, and also caused a two-fold increase in ∆F508 CFTR-associated chloride current in a canine MDCK-C7-derived cell line that expresses a stably transfected copy of human ∆F508 CFTR. Our previous studies also demonstrated that Dox is able to impart structural integrity to ∆F508 CFTR, increasing its half-life and decreasing its proteolytic sensitivity, which is indicative of efficient folding. Additionally, two other structurally related drugs, i.e., the anthracycline, daunorubicin, and the anthraquinone, mitoxantrone, were shown to have similar effects on ∆F508 CFTR expression. In the current work, we have extended these effects to two other related aza-anthracenedione molecules, BBR2778 (Pixantrone) which is a cancer chemotherapy drug with lower non-target toxicity than Dox and which is in Phase III clinical trials, and a structurally related analog, BBR2828, which is essentially non-cytotoxic with a 100-fold lower cytotoxic potency than BBR2778. Both compounds increased CFTR-associated chloride currents to a similar extent as Dox in CFBE human airway epithelial cells expressing ∆F508 CFTR, as measured by Ussing chamber experiments. The ability of the non-cytotoxic analog BBR2828 to do so is particularly important, since it indicates that correction of the ∆F508 CFTR defect is not directly tied to the toxicity of this class of compounds per se but rather is a result of other structural features. Thus, it is likely that this and/or other non-toxic analogs in this chemical class can be developed that have potential as clinically useful agents for treatment of CF in patients. Supported by a Cystic Fibrosis Foundation (CFF) grant to JWH and RM, and by the CFF-supported Dartmouth CF program. Cystic Fibrosis (CF) is caused by mutations in cystic fibrosis transmembrane conductance regulator (CFTR). CFTR is not only an epithelial chloride channel, but it also regulates other transporter and ion channels, including epithelial sodium channels (ENaC) and aquaporin water channels. In patients with CF, absent or dysfunctional CFTR results in abnormal electrolyte and fluid content on the epithelial surface. Treatments intended to normalize ion transport in CF airways through non-CFTR dependent pathways represent attractive approaches to alleviate the underlying pathologic defect. Calcium (Ca2+)-activated Cl-channels have been proposed as rescue channel for the cyclic AMP-dependent CFTR Cl-channels, offering a target for CF pharmacotherapy. Increases in cytosolic Ca2+ concentration activate epithelial Cl-channels, but inhibit epithelial Na+ channels, which is also beneficial in correcting the hyperabsorption of Na+ in CF. Herein we developed a High Throughput Screening (HTS) assay for screening compound libraries with the intention of finding compounds which can increase intracellular Ca2+. In the pilot primary screening, we screened a MSSP library which contains 2000 known bioactive compounds and natural products. 7 compounds were classified as the real hits after the secondary screening validation with the hit percentage 0.35%. We have found that the EC50 of these 7 compounds are less than 100uM in dose-response studies and none of those compounds were toxic to the cells. We have further discovered that those compounds stimulate Cl-secretion through activating Ca2+ dependent Cl-channels in CF and non-CF human airway epithelial cells in short-circuit current experiments. In summary, our data suggest that modulation of intracellular Ca2+ is a target for CF therapy. The compounds identified by our study will provide possible therapeutic leads in the treatment of CF. Non-viral gene delivery particles with various synthetic polymer coatings have been developed for cystic fibrosis (CF). Intracellular trafficking and in vivo tissue distribution of these particles must be carefully monitored to guide rational design of efficient delivery system. Fluorescent semiconductor quantum dots (QDs) allow sensitive, long-term and multi-target imaging in cellular environment, and so are a promising tool in tracing gene delivery materials in vitro and in vivo. However, application has been limited by the lack of efficient and reproducible techniques of QD bioconjugation.

Here, we describe a method of labeling DNA nanoparticles with tunable QDs that may enable us to study gene delivery in vitro and in vivo. Highly fluorescent ZnS-coated CdSe QDs were first synthesized and encapsulated in amine-containing phospholipid micelles. The non-viral vector system we used is based on a polymer backbone consisting of 30 lysines with a cysteine on the N-terminus (CK 30 ) which is conjugated to polyethylene glycol (PEG). This polymer vector condenses DNA plasmids into stable nanoparticles that efficiently transfect airway epithelial cells in vivo. To introduce the aminecontaining QDs, we utilized a hetero-bifunctional PEG with an NHS ester at one end, which was first reacted with the amine groups of the QDs and a maleimide group at the other end which was subsequently reacted with the thiol group in CK 30 . Excess reagents were removed from the final conjugates QD-PEG-CK 30 by filtration. Conjugation was first confirmed by 0.5% agarose gel electrophoresis. Free QDs were positively charged and migrated toward the cathode. PEGylated QDs migrated significantly slower than free QDs, and QD-PEG-CK 30 conjugates migrated at still a different rate. To further demonstrate conjugation, we biotin-labeled the CK 30 and assessed the binding of QD-PEG-CK 30 -biotin to avidin coated agarose beads by monitoring the yield of fluorescent beads. QDs, PEGylated QDs, and physical mixture of QDs and CK 30 -biotin were also incubated with the agarose beads as controls. Fluorescent beads were detected only when QD-PEG-CK 30 -biotin was present. To directly evaluate the ability of QD-PEG-CK 30 to bind double stranded DNA, we also immobilized double stranded DNA fragments to agarose beads. QD-PEG-CK 30 , but not unconjugated QD bound to the immobilized DNA fragments as determined by the yield of fluorescent beads.

QD-PEG-CK 30 was then used to compact luciferase reporter plasmids. Electron microscopy images of the compacted DNA showed that QDs were integrated into some DNA nanoparticles. We transfected HeLa cells with QD-labeled and unlabeled DNA nanoparticles. Luciferase assay and toxicity assay indicated that QD-labeled DNA nanoparticles transfected HeLa cells, though with less reporter gene expression than unlabeled nanoparticles and more toxicity than uncompacted QD.

In conclusion, we have developed a method to conjugate QD to DNA nanoparticles. Improvement of the quality and biocompatibility of the QD may be required for further in vitro and in vivo studies. This system is also versatile. Other less toxic optical makers can be readily tested.

This study was supported by DK58318 and the CFF. In cystic fibrosis (CF) the Epithelial Sodium Channel (ENaC) hyperactivity plays a role in the pathogenesis of chronic lung disease. The missing ENaC regulation by the CF Transmembrane Conductance Regulator (CFTR) causes increased absorption of sodium ions and fluid across airway epithelia leading to the depletion of the perciliary liquid layer and to the consequent inhibition of mucus clearance. We developed a HIV-based lentiviral (LV) vector containing a siRNA cassette to efficiently knockdown the expression and activity of ENaC in human respiratory cells. Background: Sildenafil has been implicated in the relocation of cystic fibrosis transmembrane conductance regulator (CFTR) protein. The effect was observed in vitro and in the presence of doses roughly 300 times larger than those commonly used for treating erectile dysfunction.

Aim: To evaluate in vivo therapeutic efficiency of clinical doses of sildenafil and vardenafil, two approved type V phosphodiesterase inhibitors, for correcting chloride transport defect in ∆F508 mice.

Methods: We measured transepithelial potential difference in vivo across the nasal mucosa as a readout for sodium and chloride conductance. The effect of a single intraperitoneal injection of sildenafil (0.7 mg/kg) or vardenafil (0.14 mg/kg) was investigated in DF508/DF508 and normal homozygous mice.

Results: In DF508/DF508 mice, chloride conductance, evaluated by perfusing the nasal mucosa with a chloride-free solution in the presence of amiloride and with forskolin, was corrected 1 h after sildenafil administration. A more prolonged effect, persisting at least for 24 h, was observed with vardenafil. The forskolin response was increased after sildenafil and vardenafil in both normal and DF508 mutant animals. No effect on sodium conductance was detected in any group of animals.

Conclusion: Our results provide preclinical evidence of effectiveness of both drugs for correcting chloride transport defects in the CF.

Acknowledgments: This work was supported by grants from the French CF Association, Vaincre la Mucoviscidose and by an educational grant from Pfizer Belgium. There is a strong interest in developing small molecules able to correct the phenotypic effects of cystic fibrosis (CF) mutations. Many mutations (e.g. ∆F508) impair the function of CFTR protein, by altering the protein targeting to the plasma membrane and/or by causing an abnormally low open channel probability. Drug-like organic compounds may restore the correct membrane localization ("correctors") or increase channel activity ("potentiators") of mutant CFTR. In the last years, various research teams have identified molecules with activity as correctors (corr-4a, VRT-325, miglustat, curcumin) or as potentiators (tetrahydrobenzothiophenes, phenylglycines, sulfonamides, VRT-532, 1,4-dihydropyridines). However, these compounds have been tested using different assays and the results are sometimes controversial, with marked differences in declared efficacy and potency of compounds.

We have tested a panel of correctors and potentiators to directly compare their effects under the same experimental conditions. For this purpose, we have used the functional assay based on the halide-sensitive EYFP-H148Q/I152L to measure ∆F508-CFTR activity in FRT and A549 cells. The assay for potentiators consisted in stimulation with the test compound (0.02 -60 µM) plus forskolin (on cells previously incubated at 27°C for 24 hours). The assay for correctors consisted in 24 hours incubation of ∆F508 cells with test compounds and then determination of CFTR activity in the presence of forskolin plus genistein (50 µM).

Our results indicate that all potentiators are active in our assay with a comparable maximal effect but with values of potency that vary significantly among compounds. The potency order measured in FRT cells was: DHP-194 = PG-01 (Ka~50 nM) > SF-01 (Ka = 140 nM) > DHP-229 = DHP-226 (Ka~180 nM) > act-2b (Ka = 250 nM) > felodipine (Ka = 900 nM) > VRT-532 (Ka = 2.6 µM) > genistein (Ka = 18.5 µM). A similar order of potency was found also in A549 cells expressing ∆F508.

The activity of correctors showed a more marked dependence on cell line. While the potency was comparable between FRT and A549 cells, the maximal effects showed clear differences. In FRT cells, corr-4a, corr-4b and corr-3c generated a maximal effect that was 1.5 -2-fold higher than that obtained by incubating the cells at low temperature. Conversely, VRT-325 and VRT-640 were less effective (maximal correction equivalent to 50-75 % of low temperature rescue). In A549 cells, all compounds were instead less effective than low temperature, with VRT-325 being the molecule eliciting the highest activity (50-60 % of low temperature).

Our results indicate the 1,4-dihydropyridine DHP-194 and the phenylglycine PG-01 among the most potent activators of the mutant CFTR channel. The similar behavior of potentiators in two different cell lines is consistent with the assumption that all potentiators act with a similar mechanism, by interacting with the CFTR protein itself. In contrast, the cell line dependence of correctors suggests that they act with indirect mechanisms, possibly by interacting with proteins involved in CFTR biosynthesis and trafficking.

Supported by CFFT and Telethon-Italy. We also thank CFFT and RFUMS for providing chemical compounds. Lung damage in cystic fibrosis (CF) patients is determined by mucus accumulation, Pseudomonas aeruginosa infection and chronic inflammation. Extracellular GSH is a scavenger of free radicals produced by neutrophils in inflamed tissues.

Glutathione transferases (GST) are a superfamily of dimeric proteins which conjugate glutathione to a wide range of substrates including oxidants and are involved the synthesis of leukotriens.

Clinical beneficial effects have been reported in CF patients following treatment with the macrolide azythromicin (AZM); anti-inflammatory properties have been proposed as possible mechanism.

The aim of this study is to investigate the regulation of the GSTT1 and GSTM1 activity and expression by AZM.

Reductions of about 25% and 40% on GST enzymatic activity were detected in IB3-1 and 2CFSMEo-cells respectively. GSTs mRNA expression in CF airway epithelial cell lines was analysed by quantitative PCR (qPCR). The level of GSTT1 and GSTM1 basal expression in CF cells IB3-1 was significantly higher than in isogenic non-CF cells C38. We found statistically significant decreases of GSTT1 and GSTM1 mRNA of about 30% and 25% respectively in IB3-1 cells after treatment with AZM for 24 hours, restoring the levels observed in C38 cells. In 2CFSMEo-cells after exposure to AZM we observed 50% and 45% reductions in GSTT1 and GSTM1 mRNA respectively. The macrolide JM, known to lack clinical anti-inflammatory properties, had no significant effects on GSTT1 and GSTM1 mRNA expression in all cell lines. Furthermore, AZM did not alter the mRNA expression levels of GSTP1, a glutathione-S-transferase not differentially expressed in CF and isogenic non-CF cells. Decreased expression of 50% and 85% of GSTT1 protein has been detected by immunoblotting in IB3-1 and 2CFSMEo-cells, respectively, following treatment with AZM. In the same conditions we found a drastic reduction of protein level of GSTM1 in both CF cell lines.

Finally, GSTs activity and the expression of GSTT1 and GSTM1 proteins in CF cells, were reduced approximately to the same level detected after treatment with interleukin 10 (IL-10), an anti-inflammatory cytokine, shedding light on a possible correlation between GSTs inhibition and antiinflammatory properties of AZM.

The effects of AZM described in this study suggest that downregulation of GSTT1 and GSTM1 expression may result in increased availability of intracellular GSH making CF cells less susceptible to oxidative stress induced by chronic inflammation. Inhibition of GSTT1 and GSTM1 might provide a therapeutic approach for limiting the effects of inflammation critical for lung damage in CF patients.

This study is supported by Italian CF Research Foundation; Comitato di Vicenza-Associazione Veneta per la lotta contro la Fibrosi Cistica; Azienda Ospedaliera Verona, Italy.

Tradtrantip, L.; Padmawar, P.; Yangthara, B.; Verkman, A.

Activation of CFTR chloride conductance by GPCRs involves cAMP elevation and PKA-mediated CFTR phosphorylation. We developed a 'pathway screen' in which CFTR-mediated iodide influx is used as a read-out of GPCR activation. The cell-based fluorescence assay utilizes multiply transfected epithelial cells expressing wildtype CFTR, YFP-H148Q/I152L and a specified GPCR. We recently used this assay to identify a new class of nanomolar-affinity, 5-aryl-4-benzoyl-3hydroxy-1-(2-arylethyl)-2H-pyrrol-2-one vasopressin-2 receptor antagonists (Yangthara et al., Mol. Pharm. 2007, in press ). Additional screening of 100,000 diverse small molecules yielded 4 novel chemical classes of inhibitors of CFTR activation. The potential molecular targets of pathway inhibitors include the GPCR, Gs or Gi proteins, adenylyl cyclase, phosphodiesterase, PKA and CFTR. A series of target identification studies was done to classify the new pathway inhibitors, which involved the use of agonists acting at different sites in the activation pathway, and specific site-of-action assays. The pathway screen yielded 3 new small-molecule CFTR inhibitors that are unrelated to thiazolidinone and glycine hydrazide inhibitors. One interesting class of pathway inhibitors, thiophenecarboxylates, represent the first small-molecule phosphodiesterase activators, which strongly reduce cAMP and cGMP concentration in many cell types. The best thiophenecarboxylate greatly reduced intestinal fluid secretion in closed loop mouse models of cholera and Travelers' diarrhea, and slowed cyst growth in a model of polycystic kidney disease. Other pathway inhibitors, which are potential effectors of G-proteins and PKA, are under evaluation. The GPCR-linked CFTR pathway screen developed here is useful for high-throughput parallel identification of small-molecule inhibitors of multiple targets in the CFTR activation pathway. Potential uses of the inhibitors identified here include therapy of secretory diarrheas, polycystic kidney disease, and cyclic nucleotide-dependent tumor growth, as well as pharmacological creation of CF-phenotype in ex vivo human tissues and animal models. Supported by CFF and NIH. Inhibitors of intestinal CaCCs are predicted to have anti-secretory effects in certain diarrheas, and activators of airway and intestinal CaCCs are of potential use in cystic fibrosis therapy (activation of 'alternative' chloride channels). The purpose of this study was to identify small-molecule CaCC inhibitors and activators that target CaCCs directly, rather than ubiquitous upstream processes such as calcium signaling. We screened a collection of 100,000 chemically diverse small molecules using a novel high-throughput screening assay involving lentiviral introduction of a YFP-based halide sensor in CaCC-natively expressing human epithelial cells. CaCC inhibitors were identified from iodide influx following CaCC simulation by carbachol/ATP. We identified five classes of CaCC inhibitors with micromolar potency, including tetrahydro-cyclopentaquinolines, and 3-aryl-5-(trifluoromethyl)-pyrazoles each of which was unrelated to known transport modulators. Two classes of compounds inhibited calcium-activated halide flux following stimulation by multiple types of agonists, including thapsigargin and calcium ionophores, and by patch-clamp analysis appear to target CaCC directly. Structure-activity analysis of 137 analogs of 'hits' yielded compounds with improved potency, which have been resynthesized and characterized for use in assays of antidiarrheal effiacy in rodent models of viral and drug-induced secretory diarrheas. Screening for CaCCselective activators that act in a sustained manner (non-transiently) was accomplished using a similar cell-based fluorescence assay, but instead testing for increased halide influx. Several classes of putative CaCC activators with micromolar-potency were identified in screening of 100,000 small molecules, whose mechanism-of-action and specificity are under investigation. Small-molecule modulators of CaCC function that target CaCCs directly have potential clinical applications, and may be useful in defining the physiological roles and molecular identity of CaCCs. Supported by CFF and NIH. Introduction: Inhaled hypertonic saline 〈HS〉improves lung function and decreases pulmonary exacerbations in older children and adults with CF. Initiating therapeutic interventions in the youngest patients, particularly those that target the underlying CF defect of airway surface liquid volume depletion, has potential to preserve lung function and improve prognosis. Subbarao et al performed baseline, post-albuterol and post-HS lung function testing in infants using the raised volume rapid thoracoabdominal compression technique 〈RVRTC〉and demonstrated no significant drop in lung function 〈 Pediatric Pulmonology, 2007〉. However, performing three sets of RVRTC maneuvers under the same sedation could prove difficult. Before conducting a therapeutic trial of HS in this population, a simplified protocol must be possible at multiple centers. We sought to evaluate a simplified approach as well as to analyze changes in lung function and clinical findings after acute administration of HS.

Methods: In this ongoing study, clinically stable children with CF between the ages of 4 months and 4 years inhale 2.5 mg of albuterol prior to sedation with chloral hydrate. RVRTC and plethysmography are then performed before and after inhalation of 5 mL of 3‰ HS. FVC, forced expiratory volume in 0.5 seconds 〈FEV 0.5 〉, FEF 25-75 , FRC, RV/TLC ratio, respiratory rate, oxygen saturation, and chest exam findings are recorded. Predefined stopping criteria include a 20% drop in FEV 0.5 or in oxygen saturation to below 90‰.

Results: Six subjects 〈mean age 1.6 ± 0.9 years〉 have completed the protocol with 3‰ HS. Comparison of post-albuterol lung function to that obtained 15 minutes after 3‰ HS revealed no changes in mean FVC 〈561 vs 562 mL; P=0.93〉, mean FEV 0.5 〈426 vs 427 mL; P=0.89〉, mean FEF 25-75 〈801 vs 817 mL/sec; P=0.59〉, FRC 〈measured after each inhaled therapy in 3 of 6 subjects; 260 vs 260 mL; P=1.0〉, or RV/TLC 〈measured after each inhaled therapy in 3 of 6 subjects; 0.35 vs 0.34; P=0.77〉. Respiratory rate, oxygen saturation and chest exam were unchanged.

Conclusions: Results from this study demonstrate that a two-step protocol may used to evaluate the safety of HS. Based on these findings, acute administration of 3‰ HS is safe in children ages 4 months to 4 years with CF. Despite the known improvement in mucociliary clearance, preliminary findings demonstrate a lack of an immediate response in lung volume measures. Given the demonstrated benefits in older children and adults, a multicenter therapeutic trial of HS is warranted.

Supported by the Cystic Fibrosis Foundation. Methods and Results: We have characterized AAV serotypes 1-9 in addition to twenty novel vectors isolated from human or macaque tissues to transduce the murine airway epithelium in vivo. Vectors [1E+11 genome copies (GC)/mouse] expressing α-1-antitrypsin (AAT) and βgalactosidase (β-gal) were co-instilled into the mouse lung or nose. Transgene expression levels were monitored by assaying AAT concentration in serum as well as the number and cell-types positive for (β-gal) expression in lung and nasal airways. Of all vectors tested AAV5 and AAV6 were the two most efficient vectors in conducting airways. When these AAV vectors (2E+11 GC) were subsequently evaluated on human ciliated airway epithelial cultures (HAEC), in contrast to our findings in mouse airways, AAV5 failed to transduce HAEC, whereas AAV6 resulted in~10% of the HAEC expressing transgene. Since AAV6 was the most efficient vector in mouse and human airway epithelium we performed structure-function analyses of the AAV6 vector capsid and found two atypical capsid residues that were unique in otherwise conserved positions (F129, K531). To generate a potentially fitter vector, residue F129 was mutated to its conserved state. Residue K531 was found to confer lung tropism and was left unchanged. The resulting vector, AAV6.2, transduced mouse lung and nasal airway with greater efficiency than all AAV vectors tested. The increased transduction efficiency of this vector was also observed (~20%) in HAEC derived from six different human subjects. To continue our preclinical studies in a more relevant model, AAV2/6.2 expressing EGFP was tested in ciliated cultures derived from macaque airways and showed 40-50% of cells expressing EGFP one week after inoculation with 2E+11 GC. Confocal microscopy revealed that AAV6.2 targeted a significant number of ciliated cells: the airway cell-types that likely require CFTR expression in CF patients. AAV6.2 expressing rhesus α-fetoprotein (rhAFP) was then inoculated in the nasal airways of a rhesus macaque and transduction evaluated by monitoring concentration of rhAFP in the nasal lavage fluid. We have found that RhAFP expression remained high (~300 ng/mg) and stable for at least 72 days.

Conclusion: The enhanced transduction efficiency of AAV2/6.2 vector in human and macaque airway cultures and its ability to stably transduce the nasal airway of a rhesus macaque in vivo demonstrates that AAV2/6.2 is a good candidate gene transfer agent for the efficient expression of CFTR in human CF airway epithelium.

Submitted for presentation at the American Society of Gene Therapy. Supported by GSK, CFF, CFPO1, P30, MTCC. Pharmacological correction of ∆F508 CFTR cellular processing is a potential therapeutic strategy for cystic fibrosis. Recent high-throughput screening has identified synthetic small molecules, such as bisaminomethylbithiazoles (corr-4), which partially restore chloride permeability in ∆F508 mutant cells. The purpose of this study was to examine the utility of natural compounds (Chinese medicinal herbs) as ∆F508 correctors. A herbal compound fraction library was constructed from 500 herbs most frequently used in Traditional Chinese Medicine (TCM) that are believed to contain therapeutic compounds for a broad spectrum of human diseases including lung disease. For construction of the TCM fraction library, crude herbal extracts were first prepared by 95% ethanol extraction on Soxhlet reflux apparatus followed by automated fractionation by preparative HPLC. Eighty fractions were collected from each of the 500 herbs. Each fraction contained 1 to 17 (average 10.4) individual compounds as determined by analytical HPLC. Collected fractions were dried and 1 milligram of the material was dissolved in 200 µl DMSO to generate 5 mg/ml solutions in 96-well plates. Each 96-well plate contained 80 fractions from one herb. High-throughout screening was done using the FRT cell-based fluorescence assay developed previously (J. Clin. Invest. 115:2564 -71, 2005 . Of 16,000 fractions screened, 87 active fractions from 12 herbs were identified, with 26 positive fractions verified in secondary screening. The positive fraction did not increase halide transport in control non-transfected cells, and halide transport in ∆F508-corrected cells was fully abolished by CFTRinh-172. We have fractionated some of the most active fractions by preparative HPLC to identify which compound(s) conferred activity. For example, in one fraction there were 14 single compounds, 2 of which conferred corrector activity with IC50s < 5 µM and efficacy comparable to that of low temperature rescue. These results demonstrate the feasibility of ∆F508-CFTR corrector discovery from natural compounds. Further fractionation, characterization and structure determination are in progress. The unexpectedly high 'hit'rate for the natural compounds suggests their further exploration in CF therapy. Cystic fibrosis (CF) is the most common genetic disease affecting the Caucasian population, with an incidence of approximately one in three thousand births. CF transpires as a result from a mutation in the cystic fibrosis transmembrane conductance regulator protein (CFTR), which regulates ion transport across epithelial membranes. Subsequently, patients afflicted with CF have an abnormally excessive incidence of chronic lung infection, with organisms such as Pseudomonas aeruginosa. Because cystic fibrosis is characterized by chronic bacterial infections, excessive neutrophil recruitment to the lungs, and a coinciding increase in pro-inflammatory cytokine production and nuclear factor-kappa B (NF-κB) activation, we hypothesized that exogenous addition of the NF-κB inhibitor IκBα might ameliorate this phenotype. We cloned the human IκBα gene as well as a mutated IκBα gene into plasmids with chicken-beta actin hybrid promoters. We then tested the new plasmids, pAAV2.CB-hIκBα and pAAV2.CB-hIκBαM, in vitro in the presence and absence of Pseudomonas aeruginosa infection in the IB3-1 and S9 cell lines. Both plasmids produce IκBα at high levels as shown by enzyme linked immunosorbant assays (ELISAs). We also show that pAAV2.CB-hIκBα transfected IB3-1 cells, after infection with Pseudomonas aeruginosa, express significantly reduced levels of Interleukin (IL)-1β (2 fold, p<0.0001), IL-8 (13 fold, p<0.0001), and TNF (2 fold, p=0.0007) as well as NF-κB activation (2 fold, p<0.0001) compared to P. aeruginosa-infected IB3-1 controls as determined by human cytokine and NF-κB phosphoprotein Bio-Plex assays; cytokine expression and NF-κB activation levels in infected pAAV2.CB-hIκBα transfected IB3-1 cells were between levels found in infected IB3-1 and S-9 cells, excluding IL-8 levels which were below S-9 levels of expression. Flavonoids are among the most potent CFTR modulators known. Equol [7hydroxy-3-(4'-hydroxyphenyl)-chroman] is a product of intestinal metabolism of dietary isoflavones such as daidzein, and has been of recent interest in studies of cancer, cardiovascular risk, and neurologic disease. Equol is metabolically stable, and 49% circulates in the free (non-protein bound) form, which is considerably greater than the proportion of free daidzein (19%). Structural differences such as modification of ring C (e.g. saturation at C-2/3, and absence of carbonyl group at C-4) and an absent hydroxyl at position C-5 distinguish equol from compounds previously reported to modulate CFTR activity. In prior work by our center, we showed that equol activates wt and ∆F508 CFTR in membrane patches excised from BHK cells and in CFBE41o-monolayers studied in Ussing chambers. Activation by equol occurred only after R-domain phosphorylation in wt and ∆F508 CFTR constructs, but was independent of Rdomain phosphorylation in ∆R CFTR, indicating activity may be related to dimerization of the NBDs or other domain-domain interaction. Because a molecule that alters NBD interactions might have effects on the aberrant processing of ∆F508 CFTR, we screened equol and other flavonoids by preincubating compounds with CFBE41o-cells expressing ∆F508 CFTR for 18 hours, and then tested for rescue of CFTR dependent Cl-channel activity after exchange of media solution. We found that preincubation with equol (50-100 µM) induced rescue of short-circuit current compared to vehicle treated cells (12 vs. 4 µA/cm2, p<0.001, n=8). We then evaluated 24 hour preincubation with equol (50uM) for biochemical evidence of CFTR processing correction in ∆F508 CFBE41o-cells grown in polarizing conditions. Immunoprecipitation and in vitro phosphorylation demonstrated minimal formation of Band C compared to vehicle treated cells, but adaptation of a more sensitive avidin label/biotinylation assay specific for surface localized CFTR revealed clear evidence that equol preincubation led to CFTR at the plasma membrane. Next, we examined equol in HeLa cells stably transduced with ∆F508 CFTR. Preincubation of equol led to dose-dependent increases in halide transport measured by fluorescence-based halide efflux (SPQ) after stimulation by genistein (0.50 and 0.95 fluorescence slope (∆%/sec) with equol 5 and 50 µM respectively, compared to 0.2 ∆%/sec in cells pretreated with vehicle alone; p<0.005, n=120-180 cells/condition). Immunohistochemical staining of ∆F508 HeLa cells for CFTR with 24-1 C-terminus antibody showed rescue of surface localized protein with equol (50 µM) preincubation for 16 hours, while vehicle treated cells showed only perinuclear staining. In summary, we show functional, biochemical, and immunohistochemical evidence that the naturally occurring flavonoid equol corrects the ∆F508 processing defect in two model systems. A naturally occurring agent that both activates and corrects ∆F508 CFTR deserves further exploration as a potential CF therapeutic, and may lead to new insights regarding domain-domain interactions that influence the activation and biogenesis of the mutant channel. Supported by NIH and CFF. mote ∆F508 CFTR maturation has been identified. Although several small molecule agents have been described that overcome ∆F508 CFTR processing defects in specific cellular models, few studies have directly compared the activity of temperature corrected and chemically corrected ∆F508 CFTR in polarized cell systems. In the present study, we examined chemical and temperature corrected activity of ∆F508 CFTR. Correctors included all members of the CFFT Modulator Library (C1, C2, C3, and C4; Rosalind Franklin University, Chicago, IL). In a screen using ∆F508 CFBE41omonolayers, maximal corrector activity across the two model systems exhibited a rank order of C4 >> C3 > C2 = C1, using forskolin (20 µM) and genistein (50 µM) stimulated Isc as a sensitive test for ∆F508 CFTR activity at the plasma membrane. No change in Isc was observed in matched control (parental) cells lacking ∆F508 CFTR expression. Based on these results, we further defined the activity of C4 in ∆F508 CFBE41oand FRT model systems, including dose/response and time dependence for peak Isc rescue. In CFBE41o-cells, exposure to 2 µM C4 for 8 hrs produced maximal reproducible correction of ∆F508 CFTR processing, with loss of activity following prolonged or high concentration exposure; in FRT cells, peak effects were seen at 24 hours. ∆F508 CFTR activity following small molecule treatment qualitatively mirrored temperature correction (27°C growth for 24-48 hrs) in both cell types. Maximal currents produced by stimulation with forskolin (20 uM) and genistein (50 µM) in ∆F508 FRT monolayers following C4 pretreatment were 83% of that produced by temperature correction (p<0.001). Forskolin was responsible for 57% of maximal current in FRT ∆F508 cells following chemical correction and 43% of maximal currents following temperature correction. In ∆F508 CFBE41o-cells, maximal currents following chemical correction were 22% of that produced by temperature correction (p<0.001). Forskolin was responsible for 10.8% of maximal currents in CFBE41o -∆F508 cells following chemical correction and 15.6% of maximal currents following temperature correction. These results illustrate dose and time response with a small molecule corrector in two polarizing epithelial model systems, and provide reassurance that observations based upon ∆F508 CFTR following low temperature incubation are relevant to functional analysis after chemical correction. Similarity in activation properties between chemical and low temperature correction suggest it is unlikely that the two maneuvers result in ∆F508 CFTR with significantly different structural properties. The studies also indicate fundamental differences in ∆F508 CFTR behavior in FRT compared to CFBE41ocells, and emphasize the importance of identifying an agent that can restore cAMP dependent regulation to bronchial epithelial cell types. Supported by the NIH, CFF and CFFT. PARTICIPANTS: N=38 patients (21 randomised to AZM and 17 to placebo) who had successfully completed a course of intravenous antipseudomonal antibiotics immediately before the trial (mean age: 23.7 years; mean FEV1: 62% of predicted).

MEASUREMENTS AND RESULTS: After treatment (mean dose of 21.2 mg/kg body weight once a week) pulmonary function declined in both groups compared to baseline (i.e. after cessation of IV antibiotics). The azithromycin group had signifcantly better results regarding the mean changes in serum CRP (AZM: +0.9 mg/l, placebo: +21.6 mg/l, p=0.019), lipopolysaccharide binding protein in serum, LBP (AZM: +0.9 µg/ml, placebo: +7.0 µg/ml, p=0.015), serum interleukin-8 (AZM: -3.1 pg/ml, placebo: +2.9 pg/ml, p=0.001) and alginate in sputum (AZM: +85 µg/ml, placebo: +353 µg/ml, p=0.048). Quality of life (German version of the CFQ) showed significantly better results after AZM in adolescents and adults. Azithromycin was well tolerated with no increase in treatment-related adverse events.

CONCLUSION: Once-weekly azithromycin ameliorated inflammatory reactions and improved quality of life. A decline of pulmonary function after cessation of intravenous antibiotics could not be prevented, however.

This study has been sponsored Pfizer Gmbh, Germany This open-label, multicenter study was conducted in the USA and Australia to evaluate the clinical responsiveness of a patient-reported outcome measure, the CFQ-R Respiratory scale, by determining the minimal clinically important difference (MCID) following a 28-day course of tobramycin inhalation solution (TIS). CF patients (N=84 [≥6 to <18 yrs, n=56; ≥18 yrs, n=28]) with Pseudomonas aeruginosa and clinical symptoms predictive of a pulmonary exacerbation (increased cough, increased sputum /chest congestion, decreased exercise tolerance or decreased appetite) were enrolled. Three efficacy measures were included: 1)Change in forced expiratory volume in 1 second (FEV1) from baseline (day 0) to end of treatment (day 28) or end of study (day 42); 2)One question about change in respiratory function (days 28 and 42; Global Rating of Change Questionnaire, Respiratory Domain, GRCQ RD;0 =no change;7 =maximal improvement or worsening);and 3)Change in CFQ R-Respiratory Scale (day 28, 42).

Average change from baseline FEV1 (mean [standard deviation, SD]) was 5.1% (16.5) at day 28 and 4.9% (16.0) at day 42. Based on GRCQ-RD at day 28, 22 patients (28%) reported no change in respiratory symptoms (score <1.1), 32 (41%) a minimal change (≥1.1 to <3.1), 16 (20%) a moderate change (≥3.1 to <5.1), and 9 (11%) a large change (≥5.1). Mean (SD) change from baseline CFQ-R Respiratory was 5.7 (19.2) at day 28 and 5.9 (20.7) at day 42. At day 28, change in CFQ-R was moderately correlated with change in FEV1 and with GRCQ-RD; the correlation was stronger for patients with baseline FEV1 <75% of predicted FEV1 values (see Table) .

Mean change from baseline on the CFQ-R Respiratory scale was 10.0 at day 28 for patients with GRCQ-RD scores indicating minimal change in symptoms (≥1.1 to <3.1; n=30); this provided an estimate of the MCID for the CFQ-R Respiratory Scale for those in exacerbation. This MCID value was consistent with estimates from distribution-based methods( 1 ⁄2 SD and standard error of measurement).

The CFQ-R was responsive to changes in pulmonary symptoms in patients in mild exacerbation following TIS treatment; the MCID in this population was 10.0 points. Responses on the CFQ-R-Respiratory scale were moderately correlated with changes in FEV1.

Funded by Gilead Sciences, Inc. This was an open-label, multicenter study conducted in the USA. We determined the minimal clinically important difference (MCID) for the CFQ-R, Respiratory Scale following a 28-day course of tobramycin inhalation solution (TIS) in patients with CF and chronic PA infection (N=140, 14 children [<13 yrs]). Patients had received ≥3 courses of TIS (mean = 5.4) within the previous year, however their respiratory symptoms were stable at study entry, with forced expiratory volume in 1 second (FEV1) between 25% to 75% of predicted values. Efficacy measures included: 1) Percent change in FEV1 (L) from baseline (day 0) to treatment end (day 28); 2) A single question about change in respiratory function (day 28; Global Rating of Change Questionnaire; Respiratory Domain, GRCQ-RD; 0 =no change, 7 =maximal improvement or worsening); 3) Change in CFQ-R-Respiratory Scale (day 0 to 28); and 4) Change in log10 PA colony-forming unit (CFU) density in sputum (day 0 to 28). The MCID for CFQ-R was estimated using three methods: 1) change in CFQ-R (day 0 to 28) for the patient subset with minimal change in respiratory function, as determined by the GRCQ-RD at day 28; 2) CFQ-R standard error of measurement (SEM) from a validation sample), and 3) 1 ⁄2 standard deviation (SD) of the CFQ-R Respiratory Scale scores.

At day 28, change from baseline FEV1 (mean [SD]) was 1.0% (11.3), change from baseline CFQ-R was -0.6 (12.6), and change in log10 PA CFUs was -0.4 (1.5) . Based on the GRCQ-RD at day 28, 37 patients (41%) reported no change in respiratory symptoms (score <1.1), 40 (44%) a minimal change (≥1.1 to <3.1), 10 (11%) a moderate change (≥3.1 to <5.1), and 4 (4%) a large change (≥5.1). Pearson r-values for the correlation of change in the CFQ-R-Respiratory scale and change in FEV1, log10 CFUs and GRCQ-RD were 0.1, 0.08, and 0.46, respectively. Estimates of the MCID for CFQ-R-RD ranged from 4.6 to 5.9 for adults/adolescent patients (Table) .

In patients with CF who had no immediate need for antipseudomonal therapy at study entry, the CFQ-R-Respiratory scale appeared responsive to changes in patient disease perception following 28 days of TIS treatment; the MCID for CFQ-R was approximately 5 points for the adolescent/adult patient population.

Funded by Gilead Sciences, Inc. Ex vivo chloride secretion measurements (Intestinal current measurement, ICM) in CF patients have been established over the past 15 years to study the CFTR-basic defect in more functional detail. Modified Micro-Ussingchambers are used to registrate the transepithelial short-circuit current (Isc) in freshly obtained human rectal suction biopsies as a measure of ion transport after stimulation with secretagogues. Hereby, the CFTR Clchannel, its amount of residual function in CF and alternative Cl-channels can be investigated by a standardised protocol.In the course of the development of CFTR pharmacotherapeutics as well as agents activating alternative Cl-channels, ICM may function as an useful outcome parameter in preclinical and clinical trials. It is easy to perform repeatedly at all patients ages and comprises the safety advantages of an ex vivo method which is relevant especially for early study phases. Aim of this study was to describe reference values and quantify the intraindividual variability of different ICM parameters.

Methods: A total of n=574 rectal biopsies from n=212 individuals; with pancreatic insufficient (PI)-CF (n=22; mean age 14.5 years), pancreatic sufficient (PS)-CF (n=9; 10.8 years), excluded CF by ICM diagnostics (n=169; 12.8 years) and healthy control (n=12; 26.3 years) were included into analysis. For calculation of intraindividual variability, 2-4 biopsies/patient were compared with respect to basal tissue resistance (Rt basal), basal open circuit potential difference (PD basal), basal short circuit current (Isc basal) and the Isc responses to stimulation with carbachol (10-4 mol/l, serosal), 8-bromocyclic monophosphate (cAMP) (10-3 mol/l, mucosal+serosal) + forskoline (10-5 mol/l, serosal) and histamine (5x10-4 mol/l, serosal).Results:We determined ICM reference values for the groups of PI-CF (Isc basal 23.4 ± 18.1 µA/cm 2 , ∆Isc carbachol 0.4 ± 3.6 µA/cm 2 ,∆Isc cAMP/forskoline 2.0 ± 2.9 µA/cm 2 ,∆Isc histamine -3.1 ± 4.6 µA/cm 2 );PS-CF (Isc basal 45.6 ± 31.5 µA/cm 2 ,∆Isc carbachol 3.4 ± 6.8 µA/cm 2 ,∆Isc cAMP/forskoline 8.0 ± 10.4 µA/cm 2 ,∆Isc histamine 6.3 ± 10.8 µA/cm 2 ),and healthy control (Isc basal 39.0 ± 26.1 µA/cm 2 , ∆Isc carbachol 27.3 ± 7.1 µA/cm 2 , ∆Isc cAMP/forskoline 15.2 ± 10.6 µA/cm 2 , ∆Isc histamine 27.4 ± 15.2 µA/cm 2 ). For the total cohort, mean coefficients of variation were: Rt basal 29%, PD basal 48%, Isc basal 49%, ∆Isc carbachol 59%, ∆Isc cAMP/forskoline 64%, ∆Isc histamine 80%. Conclusion:This first comprehensive analysis of the intraindividual variability of ICM basal tissue and Cl-secretion parameters provides the basis for the method as an useful outcome measure for future clinical trials aiming to rescue the CFTR basic defect. Possible effects of pharmacological therapeutics in CF relevant human epithelia have to be adequately interpreted with respect to subject variability and laboratories reference data. Ex vivo Cl-secretion measurements have the potential of being an essential step in the evaluation process of CFTR-correcting/potentiating agents on their way from laboratory screening to the application in human CF tissue without any risk of toxicity. center, placebo-controlled, double-blinded pilot study we assessed safety and tolerability of 2.5 mg/d Moli1901 versus placebo (normal saline) administered by inhalation (PARI LC Plus) once daily for 28 days. Patients included were ≥16 years of age in phase I and ≥12 years of age in phase II, with a FEV 1 >60% predicted and stable lung disease. Overall, 12 subjects received Moli1901 and 6 placebo. Exclusion criteria included ABPA, B. cepacia infection and severe liver disease. The study involved 7 clinic visits over a period of 8 weeks to assess adverse events, spirometry, pulse oximetry and quality of life. A total of 111 adverse events (AEs) were observed in 16 subjects (101 AE in 12 subjects receiving Moli1901), with only 1 (productive cough) in the Moli1901 group being of severe intensity. The most frequent AEs related to the study medication were (productive) cough (43x) and dry throat/throat irritation (14x), and most of these resolved within 1 hour after inhalation. In the Moli1901 group no significant AE, defined as a decline of FEV 1 ≥20% from baseline accompanied by symptoms, a decrease in oxygen saturation to <90% or a fall of 8% from baseline requiring therapeutic intervention, or a change in safety parameters judged to be clinically significant was observed. This trial was not primarily designed to show efficacy; however, the median change in FEV 1 from day 1 to day 56 was -3% in the placebo group, and +2% in the Moli1901 group, and this difference was significant (Wilcoxon test, p=0.0217). Similarly, there was a significant difference between the median change in FEF 25-75% from day 1 to day 56 in the placebo group as compared to the Moli1901 group (-10% vs. +1%, p=0.029). No significant changes were observed for the other study days or for FVC and pulse oximetry. Moli was well tolerated in this trial, with the observed AEs generally being mild and of short duration. These encouraging explorative results are currently being further evaluated in explorative and confirmatory trials. 

Introduction: Results from published data elucidate that microbes found in the upper respiratory tract are similar or the same as those found in the lower airways of CF patients. Inhaled, aerosolized drug delivery to the lower respiratory tract is an established treatment route. However, drug delivery systems capable of depositing drug to the paranasal cavities are not yet established and require evidence of deposition and efficacy. PARI developed the VibrENT™ paranasal drug delivery system to enable the aerosol and drug to penetrate into the nose and sinuses.

Objectives: This study was conducted to demonstrate that the PARI VibrENT™ pulsating drug delivery system is capable of ventilating the human paranasal sinuses of 3 healthy volunteers.

Methods: 81mKr-gas was continuously ventilated through the nasal tract of three healthy non-smokers in front of a single-head gamma camera (DIACAM, Siemens, Germany), using the PARI VibENT™ pulsating drug delivery system. The nebulizer was coupled to the right nostril and a flow resistor to an output tube was inserted into the left nostril. During ventilation with the Krgas (about 10 sec) the subject closed their soft palate to transmit the pulsation and to prevent penetration into the lower respiratory tract. The gas supply of the VibrENT™ was directly taken from the 81mKr-gas generator output channel. Kr-gas ventilation imaging was performed with and without pulsation. Serial images were recorded with anterior and lateral views. Additionally, MRI (magnetic resonance imaging) lateral slices of the subjects' head were recorded. The gamma camera images were superimposed to the MRI images by adjusting the spatial resolution.

With no pulsation from the VibrENT™ no ventilation of sinus cavity was visualized by gamma camera images and radioactivity was detected in the nose only. When pulsation was added the maxillary sinuses can be visualized in the gamma camera images of all volunteers.

Conclusions: Without pulsation no ventilation was observed. Gas penetration to the paranasal sinuses can be demonstrated using the pulsating action of the PARI VibrENT™, potentially enabling drug delivery via aerosols. This confirms results of in vitro studies using a cast model. 81mKr-gas ventilation of the nasal cavities during 10 sec breath holding in front of a planar gamma camera head (anterior) using the PARI VibrENT without (w/o, left image) and with (w, right image) the pulsation system. The delivery and the exhaust tubing of the Kr-gas are shown together with the outline of the head, obtained from MRI pictures. Introduction: Pediatric patients with CF were previously studied in clinical trial studies of denufosol, a novel selective P2Y 2 agonist that enhances ciliary beat frequency and activates chloride secretion to hydrate the airways in the lung. Pediatric patients are often discouraged from participation in clinical trials until later stages of drug development.

Aims: In order to evaluate the safety experience with denufosol in this population, we have retrospectively examined integrated data for pediatric CF patients with mild to moderate pulmonary disease that participated in three Phase 2 studies. Demographic and baseline characteristics in addition to safety and tolerability results for 115 CF patients aged 5-18 years old are reported.

Methods: Three Phase 2, multicenter, randomized, double-blind, placebo-controlled, parallel group studies were conducted. Patients were randomized to receive either denufosol (20, 40 or 60 mg) or placebo (normal saline) TID for 28 days by inhalation. Only Study 1 included denufosol 40mg, while all 3 studies included denufosol 20mg and 60mg. The FEV 1 predicted normal required to be eligible was >75% (Study 1); 60%-90%, inclusive (Study 2); >60% (Study 3). All three studies included a one-week pre-randomization period during which reproducibility of FEV 1 (L) ±12% was required in order to be randomized to double-blind study medication. Patients were allowed to use bronchodilators, dornase alfa and corticosteroids in Studies 1, 2 and 3. Patients were allowed to use oral antibiotics including macrolides and inhaled tobramycin solution in Studies 2 and 3.

Results: A total of 115 CF patients 5-18 years old were randomized and dosed in three 28-day studies. Eighty-one received denufosol (active doses combined) and 34 received placebo. Demographics were similar for all treatment groups -denufosol pediatric patients had a mean (SD) age of 11.8 (±3.70) years old compared to placebo pediatric patients who had a mean (SD) age of 11.9 (±4.05) years. Denufosoltreated patients were 59% male and placebo-treated patients were 53% male. The mean (SD) percent predicted FEV 1 at baseline was similar between treatment groups [89.9% (±15.13) and 92.0% (±14.28) for denufosol and placebo, respectively]. The overall incidence of treatment emergent adverse events (AE) was similar between treatment groups (84% denufosol, 82% placebo). The most common AE was cough, reported by 48% and 47% of patients that received denufosol and placebo, respectively. Seven percent of denufosol and 6% of placebo patients prematurely discontinued from the study due to AEs. There were no differences in compliance with administration of study drug (96% in patients given denufosol and 100% in patients given placebo).

Conclusion: Doses of denufosol up to 60mg given TID for 28 days were well tolerated in pediatric CF patients 5-18 years old. These data demonstrate that inclusion of patients 5-18 years old is feasible regardless of administration of denufosol or placebo. A longer term Phase 3 study of denufosol in CF patients >5 years old in North America is currently ongoing.

Acknowledgements: This research was funded by Inspire and the Cystic Fibrosis Foundation. 

Objective: Pulmonary delivery of anti-infectives provides the potential to attain PK-PD indices exceeding those which can be achieved with systemic dosing. MP-376 is a novel formulation of levofloxacin that enables delivery of high concentrations over a short period, and provides taste masking. The objective of this study was to: i) determine the safety of aerosol doses of MP-376 and, ii) determine pharmacokinetics of levofloxacin in serum, urine, and sputum following aerosol doses of MP-376 using the PARI eFlow nebulizer in normal healthy volunteers (NHV) and patients with cystic fibrosis (CF).

Methods: NHV and patients with stable CF were enrolled in a single within-subject ascending dose study of 3 dose levels (loaded doses of 78, 175, and 260 mg) of inhalational MP-376 (levofloxacin solution for inhalation) or placebo. Study participants were monitored for safety and changes in pulmonary function. Serum, urine, and sputum (CF patients only) samples were collected at various times following the dose and assayed for levofloxacin concentration using HPLC. In addition, each participant received a single IV dose of levofloxacin to determine the systemic bioavailability following aerosol MP-376. Noncompartmental and compartmental methods were used to determine serum, sputum and urinary pharmacokinetic parameters. Pharmacokinetic deconvolution methods were used to determine the amount of the dose remaining in the lung over time.

Results: This study is ongoing. Dosing in 8 NHVs (6 active MP-376, 2 placebo) has been completed. There were no serious adverse effects, and no significant changes in pulmonary function were noted. Preliminary pharmacokinetic data from 4 NHVs show a proportional increase in serum levofloxacin concentrations with increasing aerosol MP-376 dose; serum AUCs were 3.1, 1.9, 5.1, and 8.9 mg-h/L for the IV dose, low, mid, and high aerosol doses, respectively. Deconvolution of the serum levofloxacin concentrations from aerosol MP-376 and IV dosing shows absorption of drug from lung over time.

Conclusion: Preliminary results show that MP-376 is well tolerated following aerosol administration in normal volunteers. Serum concentration data following aerosol dosing suggests that absorption of levofloxacin into the systemic circulation is a major route of elimination from the lung. Studies in CF patients are in progress and will be presented. Methods: A 10 mg/kg aerosol dose of MP-376 was given using a microspray aerosol generation device. When placed just above the tracheal bifurcation, and activated, it delivers a bolus aerosol dose to the lung. A 10 mg/kg intravenous (IV) dose of LVX was given as a slow bolus into the lateral tail vein. Plasma LVX was determined in all rats up to sacrifice at 1, 2, 3, or 6 hours, at which time rats were humanely euthanized and bronchialalveolar lavage (BAL) performed. Fluids were analyzed for LVX concentration using an HPLC/MS method.

Results: The plasma LVX versus time profiles of both IV and aerosol doses were best described by a two compartment model. The plasma AUC after an IV dose of levofloxacin and an aerosol dose MP-376 were similar (3.79 mg᭹hr/L vs. 3.72 mg᭹hr/L, respectively) suggesting near 100% bioavailability from the lung. After an aerosol dose, the mean residence time (MRT) was prolonged when compared to the intravenous dose (0.88 vs. 0.79 hours). This delay in absorption was associated with an increase in BAL LVX AUC0-6h in BAL (1.6 mg᭹hr/L vs. 8.3 mg᭹hr/L for IV vs. aerosol dosing, respectively).

Conclusion: These data show that levofloxacin is highly available to plasma following a single microspray aerosol dose of MP-376. The aerosol dose does produce a slightly longer mean residence time in plasma, suggesting delayed, but complete absorption from the lung; this delay in absorption was associated with increase in BAL AUCs. These data suggest that high concentrations of LVX can be attained in lung fluids following an aerosol dose of MP-376.

AUC: area under the time-concentration curve; CL: clearance; Cmax: maximum concentration; F: bioavailability; MRT: mean residence time. Methods: Animals were assigned to one of five exposure groups for all studies. For the 28-day toxicology studies, animals were exposed for up to four hours daily for twenty-eight consecutive days, followed by a 28-day recovery period. Respiratory function and cardiovascular safety were conducted on the first and last day of the 28-day study in dogs. Rats were exposed to 5, 10, or 20 mg/kg/day, and dogs were exposed to 5, 10, or 15 mg/kg/day. At the conclusion of the study, necropsy was performed and all respiratory tissues were harvested, weighed, and underwent gross and microscopic examination. A separate dog respiratory safety study was conducted in which dogs were exposed to 5, 10, or 20 mg/kg.

Results: Four week repeat dose inhalational exposure of MP-376 at target doses up to 15 mg/kg/day in dogs and 20 mg/kg/day in rats was not associated with any test article-related changes in any respiratory tissues. Aerosol administration of MP-376 at doses of up to 20 mg/kg in dogs was found to have no acute effects on minute volume, tidal volume, respiratory rate, or ECG.

Conclusion: Local effects due to the inhalational administration of levofloxacin formulated as MP-376 were not observed in either rats or dogs. These data suggest that the risk of respiratory toxicity from nebulized doses of MP-376 is low. Methods: Bacteria were grown overnight in Mueller-Hinton Broth (MHB) at 37°C and then sub-cultured into fresh MHB and allowed to reach log phase (4 hours). Female Swiss mice were infected under anesthesia by intratracheal instillation of 0.05 ml of a 2 x 106 CFU/mL bacterial suspension. IP doses were selected to provide LVX exposures (as AUC) comparable to that obtained with systemic dosing regimens. For the P. aeruginosa infection, treatment was initiated by either the IP or aerosol route 2 hours post-infection. Mice were euthanized 1 and 4 hours after the start of treatment, lungs harvested, and bacterial counts in lung determined.

Results: LVX plasma pharmacokinetic profiles were nearly identical following intraperitoneal or aerosol dosing. The geometric mean log CFU/lung pair (SD) for the P. aeruginosa are shown in the table below:

In both the K. pneumoniae and P. aeruginosa lung infection studies, aerosol administration was more effective than systemic administration; for K. pneumoniae, the extent of bacterial killing at 24 hrs was ca. 1.4 log CFU greater with aerosol MP-376 than with systemic LVX.

Conclusion: Aerosol administration of MP-376 produces a greater extent of bacterial killing than systemic dosing of LVX in mouse models of pneumonia due to K. pneumoniae and P. aeruginosa. The majority of morbidity and mortality in cystic fibrosis patients is caused by chronic and persistent lung infections especially with Pseudomonas aeruginosa. Since galactosyl ceramide had been previously shown to be involved in Pseudomonas internalization, ceramide levels in the plasma of CF patients were assessed and compared to healthy volunteers using HPLC followed by Mass Spectrometry. The results demonstrated that CF patients displayed significantly lower levels of several ceramide species. Also, Cftr-knockout mice displayed diminished ceramide levels in CF related organs (lung, pancreas, and ileum) and plasma compared to wildtype controls. Treatment with a semi-synthetic retinoid (fenretinide), which was previously reported to induce ceramide in neuroblastoma cell lines, was able to increase ceramide concentrations in CF related organs in Cftr-knockout mice to the levels of wildtype mice. Treatment also dramatically improved the ability of Cftr-knockout mice to control Pseudomonas infection. Following infection with Pseudomonas-impregnated agar beads, fenretinide treated Cftr-knockout mice were able to clear bacterial infection as efficiently as wildtype mice. Overall, these findings not only documented a novel deficiency of ceramide in CF patients but also demonstrated a pharmacological means to correct this defect in Cftr-knockout mice. Our data provides a strong rationale for clinical intervention that may benefit cystic fibrosis patients suffering from CF lung disease. 

Recent reports show that adult bone marrow-derived stem cells can localize to and acquire phenotypic and functional markers of lung epithelium. These findings raise the novel possibility of stem cell therapy for multitude of lung diseases. However, only small numbers of adult marrowderived stem cells localize to lung and it is not clear whether these cells will be clinically useful. We investigated whether mesenchymal stem cells (MSCs) obtained from human cord blood might have increased potential to participate in structural lung remodeling. Cord blood was obtained from normal deliveries at the University of Vermont. Mononuclear cells were isolated and plastic adherent cells were expanded and characterized as MSCs according to International Society for Stem Cells Research (ISSCR) criteria. Following systemic (2x106 cells/mouse by tail vein) administration to sublethally irradiated (1.4 Gy) immunodeficient (NOD/SCID) mice, lungs harvested 1 day, 2 weeks, 1 or 3 months later demonstrated small numbers of human b2-microglobulin positive cells in the airway epithelium at all time points. Small number of cells was found also stain positive for pancytokeratin (pan-CK) and rarely, we identified cells of b2-microglobulin+/pan-CK+ in the airway epithelium of these mice after 2 weeks. We are currently characterizing the phenotype of these cells with CCSP and CFTR but these data suggest that CB-MSCs may be a potential alternative source of stem cells for use in lung remodeling. High-throughput screening (HTS) and other drug development programs have identified CFTR activators and potentiators that require secondary evaluation and mechanistic confirmation. We recently evaluated the CFFT modulator library (Rosalind Franklin University, Chicago, IL) and found some, but not all, CFTR potentiators induced potent phosphorylation of the regulatory-domain (R-D), conventionally viewed as the first step in CFTR activation. To assess whether this observation has functional significance regarding ion channel activation by these agents, we evaluated CFTR potentiators in CFBE41o-and Fisher rat thyroid (FRT) cells stabily transduced with ∆F508 CFTR. Cells were studied in modified Ussing chambers under control conditions and after correction of ∆F508 CFTR misprocessing using either low temperature (27οC x 48 hrs) or pre-incubation with the chemical corrector C4. Total currents were determined following potentiator (at reported EC 50 ), forskolin (2 µM), and genistein (50 µM) stimulation. In temperature corrected ∆F508 CFBE41ocells, potentiator 1 (P1, CFpot-532), an agent that does not induce R-D phosphyorylation, caused modest activation when acutely administered (5 µM) (15.5 vs 1.1 µA/cm 2 in vehicle treated cells, P < 0.005). Importantly, this agent potentiated forskolin mediated short-circuit current (ISC: 9.8 vs 1.5 µA/cm 2 , P < .05). Forskolin accounted for 43% (vs. 4% with vehicle) of total current, demonstrating potent rescue of cAMP dependent CFTR activity, an effect not previously reported in this cell type (P<0.005). In contrast, two CFTR potentiators that induce R-D phosphorylation, P8 (UCcf-029, a benzoflavone intermediate, 2 µM) and P10 (UCcf-152, an isoxazole, 50 µM) induced modest direct activation of CFTR (5.3 and 6.1 µA/cm 2 respectively, P=0.10 and P<0.01), but no evidence of forskolin potentiation (P8: 2.3 µA/cm 2 , 10% of total stimulated current P=NS; P10: 0.4 µA/cm 2 , 1%, P=NS). In ∆F508 FRT cells grown at low temperate, P1 elicited strong potentiation of forskolin mediated currents (121.7 vs. 46.7 µA/cm 2 with vehicle, P <0.01), and modest direct CFTR activation (8.7 vs. 0.51 µA/cm 2 , P < 0.05). P8 and P10 were limited to minimal (not significant) activation and no evidence of forskolin potentiation. In CFBE41oand FRT cells pretreated with C4 (to chemically correct ∆F508 CFTR processing), P1 again potentiated forskolin mediated current (101.3 vs. 43.2 µA/cm 2 with vehicle, P<0.001 in the FRT model) but did not directly activate I SC . Findings with P8 and P10 were otherwise as seen with low temperature corrected cells. In summary, in ∆F508 CFTR polarized epithelia, P1 both activates and potentiates CFTR activity (with potentiation being the predominant effect, a unique observation in CFBE41ocells), while P8 and P10 confer activation of CFTR without forskolin potentiation. Given that P8 and P10 induce R-D phosphorylation while P1 does not, our findings suggest that agents that do not confer phosphorylation of the R-D may be better suited to rescue the endogenous cAMP mediated component of I SC . Screening of potential therapeutic agents for effects on R-D phosphorylation may help predict utility at restoring ∆F508 chloride channel activity.

Supported by NIH and CFF. Activation of CFTR is conventionally viewed as a two step process: PKA-regulated phosphorylation of multiple sites within the regulatory domain (R-domain), followed by ATP dependent gating mediated by binding sites at the interface of the two nucleotide binding domains (NBDs). CFTR 'potentiators', small organic molecules that overcome mutant CFTR gating defects at the cell surface, have been proposed as therapies for CF. Although a number of these agents are advancing to the clinical testing phase, their mechanism(s) of action are not well understood. As a step towards better characterizing CFTR potentiators available through the CFFT modulator library or other resources, we are developing standardized biochemical and functional assays to evaluate the R-domain during CFTR activation in living cells. We have previously described a gel-shift method by which phosphorylation of isolated R-domain (residues 635-836) can be monitored. Using this method, we have confirmed that one potentiator, P1 (CFpot-532), does not induce phosphorylation of the R-domain (4% of forskolin response, n=7, P = NS). Unexpectedly, two potentiators, P8 (UCcf-029, a benzoflavone intermediate) and P10 (UCcf-152, an isoxazole), robustly confer phosphorylation of the R-domain (P8: 32% of forskolin response, n=8, P = 0.002, P10: 37% of forskolin response, n=8, P = 0.004). We found the phosphorylation conferred by either P8 or P10 could be inhibited with the PKA inhibitor H89 (10 µM). Maximal stimulation of phosphorylation occurs within 2 minutes, indicating time dependence similar to forskolin. Importantly, neither P8 nor P10 increased total cellular cAMP, a finding confirmed by a number of other laboratories. The results may therefore implicate compartmental inhibition of CFTR-associated phosphotases (eg. PP2A) or phosphodiestereases (eg. PDE4) as an underlying mechanism by which isoxazole or certain flavone-derivative compounds stimulate CFTR. To further test this hypothesis, P1, P8 and P10 are being examined for effects on ∆R-CFTR chloride channel activity. These studies provide a means by which novel CFTR potentiators can be biochemically categorized based on R-domain phosphorylation, a measure of the first step of CFTR activation. Compounds working through distinct mechanisms may have particular relevance to certain CFTR mutations, and could provide synergy in the clinical setting. Supported by NIH and CFF. Mucociliary clearance (MCC) is an innate defense mechanism that protects the lungs from bacteria and viruses. MCC requires maintenance of a thin layer of airway surface liquid (ASL) to eliminate inhaled particles. The ASL volume is tightly regulated by a balance of ion and water transport across the airway epithelia. In CF, the loss of CFTR Clsecretion coupled with unregulated Na + absorption via ENaC results in ASL volume depletion and impaired MCC. Although significant progress has been made in the identification of the basic disease mechanism in CF, therapeutic approaches that address abnormal CFTR biogenesis are not currently available.

The extracellular nucleotides ATP and UTP are important mediators of ASL volume and MCC. In the airways, secreted ATP acts on the G-protein coupled P2Y 2 receptors to fine-tune MCC via the regulation of apical ion transport, ciliary beating, and mucin secretion. We have previously demonstrated that ATP controls ASL volume by inhibiting absorption through ENaC and increasing secretion through apical membrane chloride channels. P2Y 2 receptor agonists are good candidates to treat CF. However, the rapid hydrolysis of ATP and UTP on the airway surface of CF patients limits the effectiveness of this approach. Consequently, Inspire Pharmaceuticals has developed di-nucleotide molecules which retain the ability to activate P2Y 2 receptors, but are more resistant to hydrolysis by ectonucleotidases. Denufosol, Inspire's lead compound for CF treatment, potently activates P2Y 2 receptors to stimulate Cl -/water secretion, ciliary beating, and mucin release in epithelial tissues. However, the stability and potency of denufosol has not been determined using human bronchial epithelial (HBE) cells.

In primary cultures of HBEs, denufosol was significantly more stable than UTP. Under thin-film conditions, the initial hydrolysis rate of a therapeutic concentration of denufosol (1 mM) was 0.08 nmol × min -1 × cm -2 with a half-life of approximately 1 hour. By comparison, denufosol is more than 10 times more stable than UTP on the mucosal surface of HBEs. Importantly, the increased stability of denufosol translated into an increase in efficacy for this compound in vitro. Equipotent concentrations of denufosol and UTP significantly increased ASL height. However, denufosol produced maximum ASL height increases that were both greater (77.6% versus 33.2% maximum increase over basal ASL for denufosol and UTP, respectively) and longer-lasting (72.2% versus 16.6% increase over basal ASL following a 60 minute application of denufosol and UTP, respectively) compared to UTP. Furthermore, the ASL height increases are specific for denufosol as the addition of apyrase blunts the response to UTP, but not denufosol.

Our data demonstrate that denufosol is more stable than UTP on the mucosal surface of human airway epithelia, which results in larger and more sustained increases in ASL volume. Recent phase II clinical trials show that administration of denufosol over 28 days was well tolerated and associated with improved lung function in mild CF patients. Taken together, these data suggest that denufosol is a promising candidate molecule for CF therapeutics. Mucociliary clearance (MCC) is the primary airway host defense against chronic exposure to infectious and noxious agents. MCC is dependent upon ciliary beating and the volume and composition of the airway surface liquid (ASL). ASL volume is regulated via isotonic fluid transport which is dominated by Na + absorption in the superficial epithelium. In CF, unregulated Na + absorption through ENaC drives ASL volume depletion and a subsequent decline in MCC. The accumulation of mucus in the airways of CF patients supports persistent and life-threatening bacterial infection.

Increasing airway surface hydration represents a promising therapeutic approach for treating CF. In vivo, aerosolized osmotic agents such as hypertonic saline (HS) improve MCC and lung function in CF patients. However, the benefits derived from HS treatments are transient, as NaCl is relatively rapidly absorbed by airway epithelia. Previously, Donaldson et al. (NEJM 2006) tested the hypothesis that blocking Na + absorption with the ENaC inhibitor amiloride would increase the efficacy and extend the benefit of HS. Surprisingly, amiloride blunted the response to HS, which resulted from a previously unrecognized property of amiloride to inhibit aquaporin-mediated water transport. While potent inhibitors of ENaC would likely be of great benefit to the CF treatment milieu, the compounds then available were not adequate for this purpose.

Parion Sciences has developed novel 2-substituted pyrazinoylguanidine compounds that selectively inhibit ENaC and are >100-fold more potent than amiloride. In the present study, we evaluated the ability of two compounds, 552-02 and 680, to increase ASL volume under thin film conditions. In primary cultures of human bronchial epithelial cells (HBEs), 552-02 alone produced a small, but significant increase in ASL height. In CF cultures exposed to phasic motion that simulates the shear stress generated by normal tidal breathing, 552-02 alone increased ASL heights by 46.7%. Additionally, we tested the effects of 552-02 pre-addition in combination with 4% HS. As expected, HS alone produced a rapid and substantial increase in ASL height (671.7% by 10 minutes) which declined to 50% bỹ 1 hour post-treatment. Pre-treatment of cultures with 552-02 prior to HS was both more potent (867.8% by 10 minutes) and longer-lasting (50% initial response at ~4 hours) than HS alone. On normal HBEs cultures, the 680 compound alone increased the basal ASL height by 107.7% following a 2 hour exposure. Similar to 552-02, 680 in combination with HS likewise extended the longevity of the HS effect (t1/2 ~4 h versus ~1 h for 680 + HS and HS, respectively).

By blocking Na + absorption, the Parion compounds alone increased ASL volume in both normal and CF HBEs. Strikingly, when used in combination with HS, the Parion compounds enhance and sustain the increase in ASL volume associated with HS alone. Our data demonstrate that the Parion compounds alone are sufficient to increase basal ASL volume. Furthermore, these data provide a proof-of-concept that combinational therapies utilizing osmotic agents and compounds that regulate ion transport will provide therapeutic benefits to CF patients. Methods: A total of 24 patients (with FEV 1 ≥40% of predicted) received 500 mg of Arikace by inhalation for 14 days. Drug was administered using the PARI LC Star nebulizer. Laboratory parameters, adverse events and pulmonary function tests were collected for all study subjects in order to determine safety and tolerability of Arikace™. Sputum samples were collected to determine changes in bacterial density, and amikacin pharmacokinetic parameters were assessed at selected time points in urine, serum and sputum specimens. Change in FEV 1 , FEV 1 % predicted, FEF 25-75 % and FVC relative to baseline and change in log 10 CFU on days 7 and 14 were assessed.

Results: On Day 7, 14 and 21, the observed change for FEF 25-75 % was 0.49 (p < 0.001), 0.42 (p = 0.02) and 0.34 L/sec (p = 0.04), respectively. On Day 7 and 14, the observed change for FEV 1 was 0.24 (p = 0.002) and 0.13 L (p = 0.10), respectively, and was 7.49 (p <0.001) and 4.38 L/sec (p = 0.03) for FEV 1 % predicted. Significant relationships (p ≤ 0.05) between log 10 CFU and serum AUC:MIC ratio, and between changes in log 10 CFU and FEV 1 , FEV 1 % predicted and FVC were identified. Treatment was safe and well tolerated with the most frequent adverse events being dyspnea and headache of mild to moderate severity.

Conclusion: Inhaled Arikace™ 50 mg/mL was well tolerated and in select patients improved pulmonary function. Together these clinically relevant changes from baseline likely represent drug effect and warrant further development of liposomal amikacin for inhalation in patients with CF.

Allergic Bronchopulmonary Aspergillosis (ABPA) is a disease caused by hypersensitivity to Aspergillus Fumigatus (AF). The prevalence of ABPA is approximately 1%-15% of patients with CF and can contribute to worsening of their pulmonary disease. The treatment of ABPA consists of high-dose oral corticosteroids for many months and may include also antifungal antibiotic such as itraconazole. The clinical effectiveness of corticosteroids is usually shown by an improvement in clinical symptoms and radiological parameters, as well as a reduction in total IgE. The prolonged period of systemic steroids may cause significant side effects such as cushingoid appearance, hypertension, glucose intolerance and more. Over the past 5 years, we have used pulse high-dose methylprednisolone in 5 patients with CF. Method:All 5 patients were diagnosed as suffering from ABPA by the standard criteria as the may have mucoid impaction or central bronchiectasis on chest radiography, an elevated IgE (>417 IU/mL), the presence of specific IgE anti AF and an elevated eosinophil count. The patient may have a positive skin prick test to AF allergen. They were treated once a month with IV pulse high-dose methylprednisolone (10mg/kg/a day) for 3 days once a month for several months until the total IgE decreased to normal values. Three patients were treated also with Itraconazole. In 2 patients Itraconazole was discontinues due to side effects. Results: We treated 5 patients age 10-35 years old (3M/2F). Four patient are pancreatic insufficient and 1 pancreatic sufficient. IgE was 622+/-371 (119-1213) at the time of diagnosis and decrease to 55+/-64 (57-89). FEV1 increased from 65+/-16 to 78+/-15. The patients also gained weight and improved their chest x-ray finding. Side effect were minor and mainly during the treatment days and resolved 1-2 days after each treatment and included malaise during the infusions, glucose intolerance on infusion days in one case, and hypertension in another case. Conclusion: IV pulse steroid treatment should be considered in ABPA as it was found effective with fewer side effects that should be expected from prolonged oral corticosteroid treatment. 

Lung pathology in individuals with Cystic Fibrosis (CF) is linked to sodium hyper-absorption. The defective regulation of the epithelial sodium channel ENaC is thought to be a major contributor to reduced Airway Surface Liquid (ASL) volume and impaired mucociliary clearance of the airways. Thus, strategies designed to inhibit ENaC function may result in clinical benefit. RNA interference (RNAi), mediated by short, double-stranded RNA molecules, can be used to target complementary RNA sequences for degradation via the RNA Induced Silencing Complex (RISC). We investigated the possibility of using RNAi to reduce expression of ENaC in the mouse lung, as proof of principle for a strategy to reduce sodium hyperabsorption in the CF lung. Potent siRNA molecules capable of efficient knockdown of ENaC alpha, beta and gamma subunits were identified in an optimised cell culture system utilising mouse kidney M1 cells (50,000 cells per 24-well; 20pmol siRNA complexed with Lipofectamine 2000; 48 hour harvest) in which ENaC expression was quantified using real-time RT-PCR. Approximately 80% knockdown of each ENaC subunit was observed with the most potent siRNA molecules in this system. Subsequently, siRNA molecules were delivered to the lungs of female BALB/c mice via hydrodynamic tail vein injection (40 µg naked siRNA, n=5-6). After 24 hours, lungs were harvested, RNA extracted and ENaC mRNA measured using real-time TaqMan PCR. Whereas, ENaC subunit mRNA levels in mice treated with a negative control siRNA were not different from untreated mice (p < 0.05), delivery of ENaC alpha and beta-specific siRNA molecules resulted in a reduction to 56.5 ± 12.3 % or 33.3 ± 4.7 % of the expression levels observed in untreated mice, respectively (p < 0.05, Mann-Whitney U). Interestingly, despite efficient knockdown in cell culture, in vivo treatment with the most potent ENaC gamma-specific siRNA molecule led to no reduction in ENaC gamma expression compared with untreated controls (p > 0.05). These data show proof of principle that ENaC expression can be reduced in the lung using siRNA. Further work is needed to assess the functional consequences of inhibiting ENaC in the lung. 

Gene therapy for Cystic Fibrosis lung disease will likely require longterm transgene expression; however, in the lung, many gene transfer agents have resulted in only transient gene expression. Previously we have shown that the choice of enhancer/promoter elements has a strong influence on the duration of reporter gene expression following delivery of non-viral vectors to the mouse lung (Gill et al., 2001 , Gene Ther, 8, 1539 . However, it is also possible that other elements of plasmid vector design play an important role. Using a clinically relevant mouse lung aerosol model, we have studied the effect of varying the plasmid CpG-dinucleotide content on the duration of expression from plasmids that had an identical enhancer/promoter sequence. Firstly we constructed a CpG-free plasmid vector containing a synthetic CpG-free luciferase gene under the transcriptional control of the human CMV enhancer/EF1α promoter. Secondly, we constructed a similar vector containing an identical enhancer and promoter but with a small number of CpG motifs, in the luciferase gene (97 CpGs) and in the backbone sequence (52 CpGs). We then investigated the level and duration of transgene expression following aerosol delivery of these plasmids complexed with the Genzyme lipid GL67 to the lungs of BALB/c mice (2.5 mg/ml pDNA, 6mM GL67, 10ml, Pari LC+ Nebuliser). Total lung extracts were assayed for luciferase activity at days 1, 2, 7, 14 & 28 (n=6 per time-point). The CpGcontaining plasmid initially directed approximately 2-fold higher levels of reporter gene expression than the CpG-free plasmid (p<0.05 both days 1 and 2 post-delivery). However, while expression from the CpG-containing plasmid subsequently fell slowly to background levels, expression from the CpG-free plasmid increased 2-fold to day 7, and remained at the peak levels observed with the CpG-containing plasmid until the end of the experiment at day 28 (CpG-free plasmid 4-fold, 16-fold and 50-fold higher expression at days 7, 14 and 28 post-delivery respectively; p<0.05 for each). These data suggest that the ability of a promoter/enhancer sequence to direct long-term expression in the mouse lung is dependent by the sequence of the plasmid backbone. One explanation for this observation is that a specific, but as yet unidentified, sequence exists in the CpG containing backbone that triggers transcriptional silencing. Alternatively, the well-described host inflammatory response to CpG-containing DNA, may be involved and the CpG content of the vector could be responsible for these differences. We are currently evaluating the potency and duration of effect of CFTR expressing, CpGfree, plasmids in late-stage pre-clinical studies prior to the initiation of a further round of clinical studies of non-viral gene transfer in CF patients. Background: Novel approaches to CF therapy by improving/restoring CFTR chloride channel function using gene therapeutic tools or small molecules are currently undergoing phase I and phase II clinical trials. There remains a great need for accurate and practical methods of measuring CFTR function in vivo to act as primary outcome measures of efficacy. We propose that an in vivo assay measuring CFTR function in sweat glands offers several potential advantages over the nasal potential difference test.

Objective: To develop a reliable test of sweat gland function that is capable of measuring the range of CFTR function in vivo.

Methods: We are performing repeated measures of sweat gland function in healthy controls (n=10), obligate heterozygotes (n=10), pancreatic sufficient (CFPS, n=10) and insufficient CF patients (CFPI, n=10). Sweat secretion is stimulated by iontophoresis (30µA/cm2) with 1% pilocarpine. To elucidate the most sensitive and discriminatory parameter of sweat gland function within this study group we are performing simultaneous measurements of: a) the transglandular potential difference of a stimulated skin area using an ECG electrode (ESPD) and b) single sweat gland potential differences in single sweat glands (SPD). Using a Wescor® collector cup we also measure: c) sweat secretion rate and (d) sweat chloride concentration (sweat Cl-).

Results: We report interim results of our ongoing study (Table) . Sweat rate was not different between the groups, but a gender difference was observed (mean±SD; 12 male: 1.6 ± 0.7µl/min/cm2; 15 female: 0.9 ± 0.5µl/min/cm2, p<0.001) as previously described. Our preliminary data show that ESPD as well as sweat Cl-allow good discrimination among Healthy controls, CFPS and CFPI (p<0.01 and p<0.05 respectively). SPD show similar trends to ESPD but with greater overlap (differences not significant).

Conclusion: These encouraging preliminary results, particularly the ESPD method and sweat Cl-following stimulation with a lower iontophoretic current, justify further efforts to complete enrolment of subjects and to further refine technical challenges such as minimizing the effect liquid junction potentials across the ECG electrode. Results will also be compared to the classical sweat test using higher iontophoresis current and NPD measurements in these patients.

This study is supported by the CCFF and Genome Canada. INO-4995, a prodrug, has been demonstrated to inhibit nasal potential difference (PD)in human CF nasal airway epithelia and CF mice and this effect is more potent with repeated dosing. As such it is being developed as potential therapeutic for cystic fibrosis. INO-4995 cell entry is facilitated by its hydrophobic propionoxy(methyl)ester protecting groups which can be hydrolyzed by intracellular carboxyesterases after the prodrug enters the cell. Once the protecting groups are removed, the drug (INO-4913) with the ether-linked octyl group is expected to be more slowly metabolized. However, the precise kinetics of the uptake of INO-4995, its conversion to its active metabolite, and Methods. Confluent cultures of Hela or T84 cells were incubated for varying periods of time ranging from 10 min to 3 hrs with medium containing [ 3 H] INO-4995 (2 x 10 6 cpm/ml) and 5µM INO-4995 cold carrier to ascertain the rate of uptake. In pulse chase experiments after cultures were incubated for 2 hours, they were washed and incubated with media without radiolabel for varying periods of time. After the indicated times, the media was removed and the cells were washed, harvested, extracted and the aqueous and organic extracts subjected to SAX and reversed phase HPLC respectively.

Results INO-4995, a prodrug, has been demonstrated to inhibit nasal PD in human CF nasal airway epithelia and CF mice and this effect is more potent with repeated dosing. INO-4995 cell entry is facilitated by its hydrophobic propionoxy(methyl)ester protecting groups which can be hydrolyzed by intracellular carboxyesterases after the prodrug enters the cell. Once the protecting groups are removed, the drug (INO-4913) with the ether-linked octyl group is expected to be more slowly metabolized. However, it is not known whether the mucus present in CF airways could impede its access to airway epithelia or whether INO-4995 would cross the epithelial layer to enter the bloodstream. Here, in order to address these issues we measured its uptake into human tracheal epithelial sheets in the presence and absence of mucus and compared its transepithelial permeation to mannitol.

We studied the uptake and transepithelial fluxes of [ 3 H]INO-4995 in primary cultures of human tracheal epithelium with transepithelial resistance of from 500 to 1000 ohms.cm 2 . The mucosal surface of some tissues was rinsed three times with PBS to remove mucus for comparison with control cells which were not rinsed. ELISA for human airway mucins performed on the rinsings confirmed that three rinses sufficed to remove the mucosal mucous blanket. The development of long-term, safe and efficacious gene therapies for lung disease will require efficient gene delivery to stem cells which produce the differentiated progeny of affected epithelial tissues. Recombinant lentiviral vectors are being considered for gene therapy applications for the lung because they can efficiently transduce a wide variety of stem and progenitor cell types. However, it is of concern that strong constitutive viral promoters have produced oncogenesis in X-SCID patient recipients of retroviral vectors in part due to the disregulation of proto-oncogenes by the strong viral promoter/enhancer located near the integration site. We propose that regulation of transgene expression to the tissue or cell type of interest will enhance the safety of gene therapy for respiratory disease. In this study we have investigated the potential for respiratory specific transgene expression from integrated lentiproviruses in cell lines in vitro and in mice in vivo. The lentiviral vectors constructed for this study contained regulatory elements predicted to produce lung specific transgene expression: the surfactant protein C promoter (SPC) for alveolar epithelial type II cell (AECII) expression, the Clara cell 10kd protein (CC10) for Clara cell expression in the airway, and the Jaagskiete sheep retrovirus (JSRV) promoter for expression in both cell types. Transgene expression from the SPC and CC10 vectors was restricted to AECII and Clara cell lines respectively, while expression from the JSRV vector was observed in multiple respiratory and non-respiratory cell types. Following intra-tracheal delivery of lentivector supernatant to mice, transgene expression was observed in AECII from the SPC lentivector (N=8 mice), and in Clara cells from the CC10 promoted lentivector (N=6 mice). Transgene expression was not detected in non-respiratory tissues following intravenous delivery of CC10 and SPC lentiviral vectors to neonatal murine recipients (N=4 and N=7 respectively). In summary, incorporation of genomic regulatory elements from the SPC and CC10 genes resulted in respiratory specific transgene expression in vitro and in vivo. These vectors will provide a useful tool for the study of lung biology and the development of gene therapies for lung disorders such as CF. Cystic fibrosis (CF) is associated with chronic pulmonary inflammation and progressive lung dysfunction, associated with the formation of oxidants derived from neutrophil myeloperoxidase (MPO). Indeed, respiratory tract levels of MPO are extensive, and have been found to correlate with decreases in respiratory parameters or disease severity in CF. MPO-derived oxidants thus may play a role in perpetuating the progressive lung dysfunction associated with CF. Based upon these premises, we reasoned that MPO could represent a novel therapeutic target for the treatment of inflammatory diseases such as CF. Herein we report that substituted urea and thiourea compounds are potent inhibitors of MPO. Utilizing tetramethylbenzidine (TMB) as a classical heme peroxidase substrate, we screened a large number of structurally varied urea and thiourea derivatives. Alkyl, cycloalkyl, and aromatic ureas and thioureas all proved to be effective MPO inhibitors, however, with a high degree of variability in potency. Of those compounds examined, substituted phenylthioureas proved to be the most potent MPO inhibitors. Various ortho-, meta-, and para-substituted phenylthioureas (PTUs) were then assessed for inhibitor potency. The most potent inhibitor of MPO was 2-chlorophenylthiourea (IC50 = 700 nM, Ki = 320 nM). The potency of ortho-PTUs as MPO inhibitors showed a linear relationship with ortho sigma substituent constants (reflecting resonance, steric and inductive effects). Moreover, cyclic voltametry experiments with ortho-substituted PTU compounds revealed a linear relationship between inhibitor potency and thiourea oxidation potential. This correlation is indicative of a substituent effect on the relative ease of sulfur oxidation within the thiourea moiety. This finding suggests that the redox-active sulfur plays a crucial role in MPO inhibition, perhaps via a direct interaction with the heme of MPO. Thioureas proved to also inhibit the chlorinating activity of MPO, whereas sustituted ureas were not effective in this regard. Both ureas and thioureas were also effective inhibitors of dityrosine formation by MPO. These findings will help facilitate the rational design and optimization of MPO inhibitors with even greater potency, which could serve as novel therapeutic agents for the treatment of CF. Gene therapy is the transfer of a normal copy of the CFTR gene and should, in theory, provide a long-term cure for CF. The development of prenatal screening programs for cystic fibrosis has provided an opportunity to identify CF patients in utero, and may provide an opportunity to treat the disease before birth, prior to the onset of disease symptoms. The long-term goal of our study is to investigate the therapeutic potential of delivering the normal CFTR gene to epithelial stem cells in fetal CF mice by in utero administration of recombinant lentiviral vectors carrying the normal CFTR gene. This would provide us with the unique opportunity to "correct" the chloride channel defect in the epithelial stem cells, which give rise to the epithelial cells in the tissues affected by CF at a very early stage. The aim of the current study is to optimize the in utero gene delivery of recombinant lentiviral vectors, carrying an EGFP reporter gene to epithelial stem cells in fetal mice. Mice have been used for these studies, as mouse strains with a variety of CF mutations are available. In all studies, the uterus was exteriorized following a mid-line laparotomy, and the injections were made directly into the amniotic fluid of individual fetuses. In the first studies, the survival of pregnant females and pups following saline injection at E16.5 (n=2) and E17.5 (n=3) was evaluated. All five females survived the procedure and an average of 54% (26/48) of the pups present at the time of injection were born alive. This experiment demonstrated that the surgical procedures were reasonably safe.

In the next experiments, we injected 5.0ul lentiviral supernatant (~4.5x107iu/pup) into the amniotic fluid of each pup and evaluated the effect of the gestational age pups on transduction efficiency. The pregnant mice were allowed to carry the fetuses to term. At P0, P7 and P28, 2-3 pups of each litter were sacrificed and skin, trachea, lung, stomach and intestines were harvested for analysis of gene transfer and eGFP transgene expression. All but one pup (11/12) from the E12.5 time point aborted prior to parturition. From the E14.4 and E16.5 injections, 6/12 and 7/7 pups were born naturally, respectively. At E14.5 and E16.5, proviral sequences were observed in the skin, trachea and stomach, although the level of transduction varied among individual fetuses and tissues. Immunofluorescence analysis on the skin of one mouse with a high proviral copy number demonstrated that some of the EGFP positive cells were also positive for vimentin or F4/80, markers of mesenchymal cells and macrophages, respectively but negative for pancytokeratin.

Our preliminary studies suggest that in utero gene therapy with lentiviral gene delivery may be safe, however ongoing studies will provide more in depth analysis on safety and feasibility of delivering the normal CFTR cDNA to epithelial stem cells in the tissues affected by CF in utero. Future studies will investigate the therapeutic potential of delivering a lentiviral vector containing the CFTR gene to fetal pups in a murine model of CF.

Supported BACKGROUND & OBJECTIVES: Cystic fibrosis is characterized by progressive loss of lung function resulting from chronic bacterial infection and concomitant airway inflammation. While a number of bacteria are known pathogens in CF, mounting evidence suggests that the lungs of CF patients harbor numerous bacterial species, and that interactions within this ecosystem may affect outcomes. However, the composition of this community is not well characterized. We therefore initiated work to profile the bacterial diversity of the lungs of CF patients with end-stage disease by 16S rRNA gene sequencing.

METHODS: Genomic DNA was purified from snap-frozen explanted lung tissue of 21 randomly selected CF patients undergoing transplantation in Toronto between 1998 and 2006. Bacterial 16S rRNA genes were amplified from each sample using broad-spectrum PCR primers, and PCR products used to generate 16S rDNA libraries for each patient. 16S rDNA profiles for each patient were generated by automated DNA sequencing of up to 96 randomly selected clones per library, and tentative species assignments made by comparison to known 16S rRNA gene sequences using phylogenetically-based analyses.

RESULTS: Sequence analysis of 1514 16S rDNA sequences from this preliminary set of 21 patients (avg. = 72 sequences/patient; range = 36-96) demonstrates a range of complexity of the microflora in CF lungs, with some patients showing a single dominant organism and others showing multiple co-existing species. In total, over 60 species have been identified to date in this set of patients. These include species identified previously by clinical microbiology, including Pseudomonas aeruginosa (detected in 20/21 patients), and Burkholderia species (detected in 4/5 patients known clinically to be infected). Numerous organisms not previously reported in the lung were also identified, including several from the taxonomic order Burkholderiales. Interestingly, while organisms commonly associated with CF were detected in patients transplanted throughout the year, a cluster of four plant-associated species demonstrated striking seasonal variation in rates of detection.

CONCLUSIONS: The lungs of CF lung transplant patients harbor a diverse bacterial community. Sequencing of 16S rRNA genes is a powerful method to profile this diversity. Correlation of these profiles with clinical information may help identify new patterns of infection and important interactions between species in this community. 

Introduction and aims: Pulmonary infection, primarily with Pseudomonas aeruginosa, is the leading cause of morbidity and mortality in Cystic Fibrosis (CF) patients. We have demonstrated that anaerobic bacteria, belonging primarily to the genera Prevotella, Veillonella, Propionibacterium and Actinomyces, can be cultured from the sputum of CF patients. The aim of this study was to determine the antibiotic susceptibility of these anaerobic isolates grown planktonically and as biofilms.

Methods: The planktonic susceptibility of anaerobic isolates to antibiotics used in the treatment of CF (meropenem, piperacillin/tazobactam) as well as to antibiotics traditionally used to treat anaerobic infection (clindamycin, metronidazole and ampicillin) was examined using E-test® strips. Selected anaerobic isolates were grown as biofilms for 4 or 24 hours in 96 well trays prior to challenge with antibiotic. Bacterial biofilm formation was determined using crystal violet.

Results: The planktonic susceptibility of 39 anaerobic isolates (14 Prevotella, 5 Veillonella, 5 Propionibacterium, 8 Actinomyces and 7 additional isolates) comprising 9 different genera was determined. Although all of the anaerobic isolates examined were sensitive to meropenem when grown planktonically, high levels of resistance to the anti-anaerobic drugs, metronidazole and clindamycin were observed. Examining susceptibility by genera, the Veillonella were the most resistant to piperacillin/tazobactam and the Propionibacterium the most resistant to metronidazole. Furthermore, several isolates, including 4 of the 14 Prevotella isolates examined, were found to be resistant to multiple antibiotics. The susceptibility of anaerobic biofilms to antibiotics was found to be isolate and antibiotic dependent. Generally antibiotic treatment of 4 hour old biofilms was often able to eradicate biomass and prevent biofilm formation; however, antibiotic treatment after 24 hours of biofilm formation with up to 100x the MIC of an antibiotic often failed to eradicate the biofilm.

Conclusion: These results suggest that alternative antibiotic treatment regimes may be necessary to treat CF pulmonary infection if anaerobes are present. These results also suggest, based on in vitro susceptibility, that meropenem and not metronidazole or clindamycin would be the most effective antibiotic in treating any anaerobes present in CF pulmonary infection. Early eradication protocols for the first appearance of Pseudomonas aeruginosa (Psa) have become standard practice in many Cystic Fibrosis (CF) clinics. The details of such protocols often vary significantly between clinics, making it important to compare efficacy from one approach to another. However, comparison of reported results is currently very difficult. This is not only because of variations in completeness of reporting, but especially because calculation of average time to regrowth cannot by definition include patients who remain free of Psa regrowth at the time the "average" is calculated (nor those in whom a re-appearance of Psa is found to be due to a different strain). Such results would be better expressed using "Time to Event" (TTE) statistics (analogous to "Life Tables") and different protocols can be better compared using this methodology.

We report for the first time, results from our Psa early aggressive eradication protocol for the period 1995-2007, expressed using TTE statistical analysis. Children are seen in clinic on average 4 times yearly for full assessment, and at each clinic visit sputum or pharyngeal cultures obtained by the physiotherapist. Standard treatment for "first growth" Pseudomonas aeruginosa consists of 2 weeks of intravenous antibiotics (piperacillin 600 mg/kg/day plus tobramycin 12 mg/kg/day) followed by 3 weeks of oral ciprofloxacin (20-30 mg/kg/day) and inhaled colymycin (100 mg bid) for 6 months. A total of 145 treatment courses have been completed for the period 1995-2007 with a 90% clearance rate, defined as 3 or more consecutive negative cultures for Pseudomonas aeruginosa over 6 months.

Since the initiation of this protocol, the percentage of patients colonized with Psa in our CF pediatric clinic has declined from 44% in 1995, down to 15% in 2007. Subsequent regrowth of Psa occurred in 1/3 of patients. Treatment courses may then be repeated a second, third, or fourth time. To determine whether these repeat Psa isolates are newly acquired or identical to the previous isolate, RAPD typing of these isolates is routinely assessed. A total of 44 isolates have been RAPD tested and have identified the same strain as the previous isolate in 16 (36%) and a different strain in 28 (64%)treatment courses. The average time to regrowth for those who had a recurrence of Psa following this treatment protocol was 24 months, but this does not include those patients who have had no further growth of Psa for periods up to 10 years. To reflect the results for all patients, including those who have not as yet had further regrowth of Psa, we have constructed TTE curves. These measures can be used to compare results using our protocol with results from other approaches to clearing Pseudomonas aeruginosa in CF.

Retsch-Bogart, G.Z. 1 Inhaled antibiotics have been part of the therapeutic armamentarium for patients with CF for decades. We studied the effect of inhaled aztreonam on respiratory symptoms in CF in a Phase 3, double blind, placebo (PL) controlled trial of AZLI, a novel formulation of aztreonam, which enrolled 164 patients with CF from 67 centers in the US, Canada, Australia, and New Zealand. All patients have completed study participation. Inclusion criteria included age ≥ 6 years, PA in sputum or throat swab, FEV1 ≥ 25% to ≤ 75% predicted, and no use of anti-pseudomonal antibiotics in the previous 14 days. Following a 14 day screening period, patients were treated for 28 days with either AZLI 75 mg or PL. Study drug was administered TID using the PARI eFlow® Electronic Nebulizer after pre-treatment with bronchodilator. Concomitant standard CF therapies were allowed, with the exception of anti-pseudomonal antibiotics, azithromycin and hypertonic saline. The primary endpoint was change from baseline to Day 28 in scores from the Cystic Fibrosis Questionnaire-Revised (CFQ-R) respiratory domain, a validated patient-reported measure of respiratory symptoms. Other efficacy measures included change in pulmonary function (FEV1, FVC, and FEF25-75), number of hospital days and number of courses of IV, oral or inhaled anti-pseudomonal antibiotics. Microbiological endpoints included change in PA bacterial density in sputum, change in susceptibility of PA to aztreonam, and emergence or disappearance of other pathogens. Safety evaluations included adverse events, airway reactivity (defined as acute decrease in FEV1 ≥ 15% at 30 minutes after dosing) and clinical chemistry and hematology. Complete efficacy and safety results from this trial will be presented.

Salam, A.P.; Orchard, C.; Wee, A.; Hodson, M.

Introduction: Oral ciprofloxacin is often used for Pseudomonas Aeruginosa (PSA) infections in patients with cystic fibrosis (CF) not requiring intravenous antibiotics. Oral chloramphenicol is commonly used at the Royal Brompton Hospital (RBH) as an alternative to ciprofloxacin, but less so in other CF centres due to concerns about aplastic anaemia and uncertainty regarding effectiveness. This study aimed to address three questions: 1) What proportion of PSA is sensitive to ciprofloxacin and/or chloramphenicol? 2) Were there any adverse haematological effects with the use of oral chloramphenicol at RBH? 3) How effective is chloramphenicol in comparison to ciprofloxacin? Study Design: We carried out a retrospective review from the RBH CF database (1985) (1986) (1987) (1988) (1989) (1990) (1991) (1992) (1993) (1994) (1995) (1996) (1997) (1998) (1999) (2000) (2001) (2002) (2003) (2004) together with analysis of patients' notes.

Methods: 1) PSA sensitivities to ciprofloxacin and chloramphenicol were recorded from sputum samples taken in late 2004.

2) All blood dyscrasias in patients who had received chloramphenicol from 1985 to 2004 were reviewed.

3) The database was searched for patients who had received a course of either oral chloramphenicol or ciprofloxacin from 2004 to 2005. 25 patients from each group were randomly selected. FEV 1 , FVC and oxygen saturations pre and post treatment completion were analysed, as were markers of symptom improvement.

Results: 1) 216 sputum samples from 216 patients grew PSA. Some were sensitive to only one antibiotic, and there were some cross-sensitivities, but in total 35.46% were sensitive to chloramphenicol and 54.25% to ciprofloxacin.

2) 648 patients received oral chloramphenicol and 755 oral ciprofloxacin from 1985-2004. Multiple courses were often given, but usually no more frequent than 3 monthly with chloramphenicol. Over 20 years, 4 cases of blood dyscrasia were recorded, none due to chloramphenicol, and all recovered.

3) In the ciprofloxacin group: the mean oxygen saturations improved by 0.25 % (p 0.672), the mean FEV 1 improved by 5.3% (p 0.126), and the mean FVC by 3.35% (p 0.202). In the chloramphenicol group: the mean oxygen saturations improved by 0.48% (p 0.118). ), the mean FEV 1 improved by 4% (p 0.438) and the mean FVC improved by 5.9% (p 0.047). 56% of patients in the chloramphenicol group and 56% of patients in the ciprofloxacin group reported symptom improvement.

A significant proportion of PSA is sensitive to chloramphenicol and in over 20 years there have been no documented cases of related aplastic anaemia. Clinical improvements with choramphenicol comparable to ciprofloxacin were also recorded. Our data shows chloramphenicol is safe and may be the oral antibiotic of choice in ciprofloxacin allergy or PSA resistance to ciprofloxacin. A randomised prospective study is needed to further evaluate the efficacy of chloramphenicol in comparison to ciprofloxacin. Pseudomonas aeruginosa lung infections are one of the main factors causing disease in CF patients. At present the generally recommended treatment is repeated courses of antibiotics. This has resulted in fewer infections and a markedly improved prognosis for the patients. But it has also severe drawbacks such as antibiotic resistance, disturbed normal flora and allergenicity. Antibiotic resistance is an emerging problem world wide and alternatives are urgently needed. Specific chicken antibodies, Anti-pseudomonas IgY, have potential both as an alternative and a complement to antibiotics (Pediatr Pulmonol 2003; 35:433) . Anti-pseudomonas IgY prevents P. aeruginosa from adhesion to epithelial cells and has affinity for flagellin in vitro. Humans do not produce anti-IgY antibodies and oral administration of IgY is generally regarded as safe. The risk that bacteria should develop resistance to IgY is extremely low. In total, 21 CF patients have received Antipseudomonas IgY for up to 12 years. Since November, 2003 the drug is given to a group of CF patients (at present 14 patients) on individual permissions granted by the Swedish MPA and reimbursed by the Swedish government. In our first study about the effect of IgY there were 2.5 positive cultures/100 months, compared to 13.7/100 months in a control group. Up to December 31, 2006, nearly five years later, the effect is maintained with 2.4/100 months for the whole study period. We are at present compiling the data from the control group during the same period. Only two siblings in the IgY-treated group have been chronically colonized (with an identical P. aeruginosa strain), which still is non-mucoid after 4.5 and 1.5 years. During the whole treatment period there have been no cultures positive for mucoid P. aeruginosa or B. cepacia and other pathogens (S. maltophilia, A. xyloxidans, atypical mycobacteria and A. fumigatus) have only appeared sporadically -possibly due to the relatively low use of antibiotics. All patients have preserved pulmonary functions and nutritional status. There have been no adverse events. Thus treatment with Anti-pseudomonas IgY has diminished the number of positive cultures and delayed the onset of chronic infection. The need for antibiotics is reduced as well as their adverse side effects. In conclusion, IgY is an important complement to antibiotics for prevention of P. aeruginosa infections in CF. Introduction and aims: We have previously shown by culture that the lungs of clinically stable Cystic Fibrosis (CF) patients are not only colonised by commonly recognised aerobic bacteria, such as Pseudomonas aeruginosa, but also by a range of potentially pathogenic anaerobic species. The aim of this study was to determine whether anaerobes are also present in the sputum of CF patients with an acute exacerbation of pulmonary infection.

Methods: Sputum samples were collected and processed using strict anaerobic bacteriological techniques, prior to commencing and at the end of antibiotic therapy, from 32 adult CF patients admitted for treatment of an acute exacerbation of pulmonary infection. Bacteria within the samples were detected by plating on selective agars, quantified by total viable count and identified by PCR and sequencing of 16S ribosomal RNA genes. As a control induced sputum samples were collected, using hypertonic saline, from 20 healthy volunteers who did not have CF and processed using similar methods.

Results: Anaerobes from a range of species including Prevotella, Veillonella, Propionibacterium, Actinomyces and Gemella were detected in high numbers (up to 6 x 10 7 cfu/g of sputum) from all patients prior to commencing antibiotic therapy with the predominant primary pathogens (P. aeruginosa or B. cepacia complex) detected in similar numbers. Anaerobic bacteria were also detected in sputum samples at the end of antibiotic treatment and for 22/32 (69%) patients they were present in lower numbers than detected before antibiotic treatment. Moreover, in 11/32 (34%) patients there was greater than a one log reduction in the total viable count of anaerobes. Similar data for detection of the predominant aerobic pathogens was also collected for 27 of these patients. P. aeruginosa or B. cepacia complex was present in lower numbers following antibiotic treatment in 22/27 (81%) patients and in 13/27 (48%) patients there was more than a one log reduction in the total viable count. Anaerobes were not detected in the induced sputum samples from 4 volunteers and in the remaining samples they were present in much smaller numbers, ranging from 10 2 to 10 5 cfu/g of induced sputum, than detected in CF sputum. The most commonly isolated bacteria from the induced sputum samples were the Actinomyces and Streptococcus, with Prevotella cultured from only 2 samples. No Veillonella and Propionibacterium species were isolated from the induced sputum samples and combined with the low frequency of isolation of Prevotella species, these results strongly suggests that the Prevotella, Veillonella and Propionibacterium species recovered from CF sputum samples are not oral contaminants.

Conclusion: These results indicate that anaerobes are present in large numbers within the CF lung during an acute exacerbation of pulmonary infection. Their presence could be of important clinical relevance to CF patients as they may contribute significantly to the inflammatory process. Airway infection and inflammation cause the majority of morbidity and mortality in cystic fibrosis (CF). The microbiology of CF is complex. Not only are the routinely identified pathogenic bacteria present, increasingly a number of other bacteria are associated with pulmonary disease. In addition, the normal microbiota likely contribute to airway disease in CF via intracellular signaling, but remains largely unstudied. The application of culture independent methods for bacterial identification in CF airway secretions provides for the first time the ability to characterize the bacterial communities in the CF airway comprehensively and efficiently. Here we present the results of a two-year study of CF airway samples to describe the bacterial communities present by ribosomal RNA sequences. We have examined 231 airway specimens from 117 subjects (82 CF, 35 controls). These results provide a unique perspective on CF airway microbiology. Bacterial communities identified in sputum were more complex than communities identified in bronchoalveolar lavage fluid (BALF). However, pulmonary status determined the level of complexity, with higher complexity observed during stable pulmonary function. We identified a collection of anaerobic bacteria that were present at high levels during pulmonary exacerbation in a subset of CF subjects. Detection rates ranged from 8-50% of CF subjects examined depending on the specific organism (Table 1 ). These organisms of interest are suspected to be involved in pulmonary exacerbation especially when cultures for standard CF pathogens are negative (~20% in our center). Preliminary data will be presented from quantitative real-time PCR experiments to track these organisms. We also will present the application of new meth-ods for community comparisons. This analysis provides the ability to look at how clinical information from individuals with CF corresponds to the microbial communities that occur in the airway. Overall, the results point to a much richer bacterial community associated with the CF airway than previously thought. The results also point to new opportunities to understand the differences within the CF population, and thus potentially improve individual patient outcomes. Cystic fibrosis (CF) patients are characterized by persistent microbial colonization of the airways and recurrent pulmonary infection. However, due to the domination of bacterial communities present in CF sputum by a small number of species (typically Pseudomonas aeruginosa), culture-independent approaches to characterize the depth of bacterial diversity in CF sputum have yielded limited information. Here we describe application of a novel microarray, the 16S rRNA PhyloChip to comprehensively describe the bacterial diversity present in temporal sputum samples of 3 CF patients who experienced periods of exacerbation (defined as hospitalization) and remission. In addition, these samples permitted assessment of bacterial community dynamism during periods of antimicrobial administration in these patients. PhyloChip analysis identified 980 bacterial taxa present across these patient samples, including a large number of known human pathogens. Antimicrobial administration caused a profound decrease in bacterial diversity in all patients examined. Despite this, a sizeable fraction of each community remained stable or proliferated during the period of antimicrobial administration, suggesting many of the bacterial taxa present were resistant to the antimicrobials administered. Taxa exhibiting this behavior included P. aeruginosa, Arcobacter cryaerophilus, Streptococcus constellatus and Burkholderia mallei amongst other pathogens. These results suggest that complex polymicrobial communities exist in CF sputum that exhibit significant antimicrobial driven dynamism and consist of multiple antimicrobial resistant pathogens. Future studies will focus on whole community gene expression to determine which virulence systems are activated in response to bacterial community perturbations and contribute to pathogenic processes in CF airways. The lungs of chronically infected Cystic fibrosis (CF) patients have been intensively studied. It is well documented that Pseudomonas aeruginosa is the predominant pathogen in CF (Høiby 1974 , Baurnfeind et al.1987 , Koch 2002 , and that P. aeruginosa accumulates in Heaps (aggregates), detected in CF sputum (Høiby 1977) and intraluminal in the lung (Baltimore et al. 1989) . The bacterial density is highest in the bronchi (Potts et al 1995) and these P. aeruginosa microcolonies are embedded in a matrix (Lam et al. 1980) . The morbidity of the infection is due to an extensive and ongoing PMN response which apparently does not eradicate the bacteria, instead leading to slow degradation of the lung. In addition, even highest deliverable doses of antibiotics fail to clear the bacteria completely, though survival has significantly increased the last 30 years due to aggressive anti P. aeruginosa therapy (Koch and Høiby 2000)

The present study was performed to investigate the significance of the aggressive therapy for the distribution of P. aeruginosa and how P. aeruginosa is organized and persists in the CF lung. The material used was: autopsies from dead short term colonized CF patients (n=12) obtained before today's aggressive antibiotic treatment , and explanted lungs from long term P. aeruginosa infected CF patients (n=3) (i.v. antibiotics every 3rd month since 1976 + inhalation of colistin daily since 1987.

Histological sections of the lungs were investigated using HE stain, Gram stain, alcian blue, antibodies against alginate and P. aeruginosa specific PNA-FISH combined with BacUni PNA-FISH and with DAPI as counter stains. Due to the high specificity of the PNA-FISH probes employed in this study, our results provide strong evidence for that before the aggressive antibiotic therapy, P. aeruginosa infected and destroyed the CF lung due to fast spreading into the respiratory zone. Today antibiotics suppress but can not eradicate the bacteria from the conductive zone, whereas the remaining respiratory zone is protected from massive biofilm infection for prolonged time. The conductive zone serves as a reservoir; here the bacteria are organized in microcolonies embedded in puss, a trait which is independent on the time course of the infection and amounts of antibiotics. These microcolonies consisted solely of P. aeruginosa. In addition, we found no bacteria adhering to the epithelial tissue. The pus consisted mainly of leukocytes, surrounding the microcolonies. A smear of DNA and dead leukocytes were detected just around the microcolonies, possible due to the Quorum sensing dependent rhamnolipid killing recently described by us (Jensen et al. 2007) .

We conclude that P. aeruginosa persists in the CF lung due to its ability to create microcolonies i.e. biofilms. Within these biofilms the bacteria are protected against antibiotics and the host defence. Respiratory viral infections, including influenza (flu) and respiratory syncytial virus (RSV) contribute to the morbidity of children with CF. Additional human respiratory viruses have been identified in children such as human metapneumovirus (hMPV) and coronavirus (cov), but these have not been examined in CF patients. In our laboratory, PCR assays have been developed to identify both traditional and newer respiratory viruses in healthy and immunocompromised children. These assays are usually performed on nasal swabs or nasal wash samples. However, oropharyngeal (OP) and sputum samples are routinely obtained in CF for bacterial culture. The goals of this trial are to determine whether these available samples can be used for viral PCR studies and to examine the epidemiology of respiratory viruses in CF. For this single center observational trial, 44 CF patients, ages 6.1 to 17.7 yrs (mean 12.8 yrs), have been recruited. Subjects provide OP and nasopharyngeal (NP) samples at quarterly clinic visits and with acute pulmonary exacerbations requiring hospitalization; sputum is also collected if patients can expectorate. NP and OP samples are pooled when both are available. Viruses assayed are flu A and B, RSV, parainfluenza (PIV) 1, 2, 3 and 4, hMPV, Cov, adenovirus (Adv), and rhinovirus (Rhv). Subjects are being followed for two respiratory viral seasons. We report here the methodology of viral PCR detection in CF sputum, as well as the results of viral detection during the first respiratory viral season in the Seattle community. The mean length of time subjects were followed was 0.84 yrs (median: 0.92 yrs, range: 0.18-1.18 yrs). Overall, 47 samples (20% of the 235 samples analyzed) were positive. Twenty-four subjects (69.1%) had at least one viral infection detected and 8 (18.2%) had two viruses at different visits. Both sputum and pooled OP/NP samples were available at 55 visits and were concordant in 80% of cases. Among the discordant cases, it was more common for sputum to be positive and OP/NP to be negative (9.1%) than the converse (3.6%). The most frequently identified viruses were Rhv (32 isolations); PIV3 and Cov were the next most prevalent with 4 and 3 isolations, respectively. In conclusion, sputum viral PCR shows promise for the diagnosis of both traditional and novel viral infections in CF. As part of ongoing clinical trials in CF, non-fermenting Gram negative bacilli (NFGNR) identified at site laboratories are often sent to the TDN Core Microbiology Laboratory for confirmation of identification. Among NFGNR from CF patients, P. aeruginosa are the most frequently isolated and also are acknowledged to be the easiest to identify. However, in a survey of 428 "P. aeruginosa" isolates sent to the laboratory in the past year, PCR identified 30 (7%) as alternate species. These included: non-aeruginosa pseudomonads, Stenotrophomonas maltophilia, Achromobacter xylosoxidans, Ralstonia spp., Alcaligenes spp., Brevidimonas spp., and Bordetella bronchiseptica. Identification of these NFGNR can be difficult; however, P. aeruginosa is an important isolate for the clinical microbiology laboratories in all CF centers to be able to identify. Many methods are available including conventional biochemical testing and rapid identification kits. In the TDN Core Laboratory we previously demonstrated the importance of combining phenotypic identification (morphology plus a short set of biochemical tests) and genotypic identification (16s rDNA sequencing of V3 and/or PCR using three primer sets) and developed an algorithm for testing. In this study, we examined all of the CF cultures performed in the clinical microbiology laboratory at CHRMC over the past two years to determine the number of isolates for which it was necessary to perform genotypic identification in NFGNR. Among this collection of 182 NFGNR from clinical cultures sent to the microbiology laboratory from the Seattle Children's Hospital CF Center, PCR identified 43 isolates as Burkholderia cepacia complex. Next most commonly identified were A. xylosoxidans (28), P. aeruginosa (23) and non-aeruginosa Pseudomonas spp. (23). Three isolates remained unidentified, even with PCR sequencing of the V3 determinant. The clinical microbiology laboratory at CHRMC has served as the core microbiology laboratory for TDN clinical trials for 8 years and thousands of CF samples are processed annually. The result is that there is increased experience with CF isolates among the laboratory personnel. However, even with experienced technologists, a significant number of NFGNR required genotypic identification in the past two years. Thus, it is important for clinical microbiology laboratories with less experience in the identification of CF NFGNR to utilize additional molecular techniques or send problematic isolates to a reference laboratory. Based on these results we propose an algorithm for the most efficient means of identifying NFGNR from CF samples using a combination of tests including phenotypic (colony morphology, biochemical testing) and genotypic identification (PCR, sequencing).

Background: The epidemiology of S. aureus has been changing in both CF and non-CF patients. In non-CF patients, methicillin-resistant S. aureus (MRSA) has traditionally caused infections in patients with risk factors such as hospitalization, surgery, and anterior nares colonization. Recently, community-acquired (CA-) MRSA infections have increased among patients without known risks. In CF, colonization of the anterior nares with methicillin-susceptible S. aureus (MSSA) is associated with transmission between patients and household members. The prevalence of respiratory tract colonization/ infection with MRSA has increased in CF, but little is known about risk factors for MRSA, including possible transmission among families, and if strains are CA-or healthcare-acquired.

Methods: This is a case-control, multicenter study to investigate potential risk factors for MRSA among children with CF 1-17 years of age. Subjects are enrolled from 3 CF centers in the New York metropolitan area that have a MRSA prevalence ranging from 11% to 21%. Cases are children with a positive respiratory tract culture for MRSA within the past 5 years. Controls (2 age and center matched per case) are negative for MRSA. The aims of the study are: [1] to determine the point prevalence of anterior nares colonization with S. aureus in children with CF and their household members; [2] to assess potential risk factors for MRSA by administering a survey to the parents of subjects inquiring about factors such as crowding, pets, and participation in contact sports and by reviewing the medical record for antibiotic use, hospitalization, and surgery; [3] to determine potential transmission of S. aureus within families and centers by assessing the molecular epidemiology of strains using pulsed field gel electrophoresis and mecA type of MRSA isolates.

Results: To date, 42 CF subjects (12 cases, 30 controls), mean age 8.9 years, and 100 household members (1 to 5 per subject) have been enrolled. Preliminary data demonstrate positive anterior nares cultures among cases, controls, and household members to be 7/12 (58%), 7/30 (23%), and 24/100 (24%), respectively. Of the 38 S. aureus isolates, 27 were MSSA and 11 were MRSA (5 cases, 3 household members of cases, and 3 household members of controls). During the past 6 months, 4 household members have been hospitalized and one has stayed overnight in a nursing home; 4 had staphylococcal infections and 2 had skin infections. Data collection and molecular analysis are ongoing.

Conclusions: We expect this study to provide insight into risk factors for respiratory tract colonization/ infection with MRSA, the role of CA-MRSA in CF, and potential strategies to prevent MRSA. Aims: 1. Revise infection control guidelines for the institution 2. Establish outpatient infection control policies consistent with those used for hospital inpatients 3. Cohort young and recently diagnosed CF patients without pseudomonas or MRSA to clinic days separate from colonized patients 4. Establish a protocol for eradication of MRSA in newly colonized patients.

Methods: Working with representatives from Hospital Epidemiology, the CF Center, inpatient units, outpatient clinics, respiratory therapy, and CF families, a uniform infection control policy for CF patients in all settings (inpatient, outpatient, home care) was drafted and revised. Arrangements were made to cohort newly diagnosed patients and children <5 years free of pseudomonas and MRSA to separate clinic days from colonized patients.

Review of the 41 MRSA positive patients in our pediatric population, for whom data was available, showed median age of acquisition of 8 years (range 1 mo to 18 yrs). For 3 patients the first culture obtained at diagnosis or transfer from another center was MRSA positive (1 mo, 1 yr, 4 yr).

An eradication protocol for newly acquired MRSA has been developed involving treatment with oral and topical antibiotics and pHisohex or Clorox washes for 5 days, plus changing disposable respiratory care equipment at the beginning and end of the treatment period. Surveillance cultures of nose, groin, and airway or sputum are done after completion of the treatment protocol and re-treatment initiated if still positive for MRSA.

Conclusion: Using a three-tiered approach of infection control guidelines, cohorting, and MRSA eradication protocol we expect to decrease our overall rate of MRSA and protect our newly diagnosed and young CF patients from acquisition. OBJECTIVE: Studies have shown that respiratory pathogens can be transmitted within CF Centers. In a recent cross-sectional study at 7 CF Centers, we assessed the rate of bacterial shedding by CF patients during office visits and noted that 6% of patients carried respiratory flora on their hands. The current study was performed to assess the effectiveness of alcohol-based rubs on hand carriage of respiratory pathogens by patients during the course of office visits.

METHODS: Four bacterial organisms were chosen for study: Pseudomonas aeruginosa (PA), Staphylococcus aureus (SA) [including methicillin-resistant (MRSA) strains], Stenotrophomonas maltophilia (SM) and Burkholderia cepacia complex (BCC). At the beginning of clinic visits, hands of study patients were cultured using the "glove-juice" technique (GJT) [1] . Hand hygiene was then performed using an alcohol-based rub. At the end of the visit, hands were again sampled using the GJT. Samples were sent to the microbiology laboratory at Dartmouth-Hitchcock Medical Center for culture and identification of study organisms. Hand and respiratory tract isolates were compared using pulsed field gel electrophoresis (PFGE). Recovery experiments were also performed to assess the sensitivity of the GJT.

RESULTS: Samples were collected from 100 encounters (50 adult; 50 pediatric). Median encounter time was 72 min (range 60.5-170.0 min). Recovery experiments determined that the level of detection for each study organism using the GJT was ~10 2 colony forming units. Confirmation of matches between hand and respiratory isolates was determined by PFGE. The number of patients with study organisms recovered from the respiratory tract was: 38 PA, 31 SA, 21 MRSA, 8 SM, and 2 BCC. The overall hand contamination rate with a matched respiratory tract isolate at either culturing time was 14% (95% CI, 8.6-22.2%). Before use of the alcohol rub, hand contamination was 9.0% (95% CI, 4.9-16.2%). In those patients with hand contamination by respiratory pathogens prior to alcohol rub, 50.0% (95% CI, 23-77%) had negative cultures at the end of the clinic visit. However, there was strong evidence of hand contamination during the encounter, despite hand hygiene, as the overall rate at the end of visits was 8.0% (95% CI, 4.2-15.0%). There was a trend toward increased hand contamination in those suffering from pulmonary exacerbations versus clinically stable patients (28.6% [95% CI, 11.8-55%] vs. 11.6% [95% CI, 6.5-20%], p=0.10).

CONCLUSIONS: Contamination of patient hands by respiratory tract pathogens is observed in the outpatient setting in approximately 14% of patients. A single application of an alcohol-based hand rub, while safe and inexpensive, has limited antiseptic effectiveness during the full course of a typical CF clinic visit. Repeated use of alcohol-based hand rubs during outpatient encounters would likely be more effective. These data support recent recommendations in the CF Foundation consensus guidelines for infection control [2] (NTHi) is the most common initial bacterium infecting the airways in cystic fibrosis (CF). NTHi infection usually precedes Pseudomonas aeruginosa infection in many chronic lung diseases, including CF, emphysema, diffuse panbronchiolitis, and immotile cilia syndrome. It is hypothesized that NTHi acts as a gateway organism paving the way for subsequent infection with P. aeruginosa, however, the mechanism by which this occurs remains poorly understood. We used a novel co-culture model of persistent NTHi biofilm infection on airway epithelial cultures to study epithelial immune responses following NTHi biofilm infections.

Objective: We hypothesized that prolonged exposure to NTHi biofilms causes airway epithelia to become tolerant to inflammatory stimuli and show subsequent decreased innate immune responses.

Methods: NTHi were inoculated onto the apical surface of Calu3 air-liquid interface epithelia and grown in co-culture. RNA from 3 sets of paired airway epithelial cultures at 4, 24, 48 and 96 hours both with and without NTHi inoculation was isolated and hybridized on custom Affymetrix chips. Data were normalized with RMA and data biases were removed using the mixed model ANOVA method. Differentially expressed gene analyses were performed using ANOVA, and pathways analyses were performed using Gene Set Enrichment Analysis (GSEA). To test the hypothesis that prolonged NTHi infection induces tolerance, epithelial cultures were pre-treated with either 4 days of NTHi or 4 days of PBS before stimulation with PBS, IL1β (100 ng/ml) or 10 6th CFUs of live P. aeruginosa before subsequent measurement of epithelial responses.

Results: Both RMA and GSEA analyses of microarray data showed many innate immune and pro-inflammatory responses at early timepoints that largely returned to baseline after 4 days of co-culture, despite an increasing apical NTHi biofilm infection. Stimulation with IL1β or P. aeruginosa after 4 days of NTHi infection resulted in significantly decreased IL8 production compared to uninfected controls. NTHi treated cultures subsequently infected with P. aeruginosa also displayed increased tight junction integrity and resistance to cell culture death.

Conclusions: Persistent NTHi infections on airway epithelia resulted in increased tolerance to bacteria and inflammatory stimuli and increased resistance to bacterial toxicity. Airway epithelial tolerance may contribute to a failure to clear subsequent bacterial infections. These phenomena would help to explain how early infections with NTHi may promote chronic bacterial infections and P. aeruginosa acquisition in CF and other NTHi related airway diseases. The colonization of CF airways by P. aeruginosa leads to intractable and persistent lung infections that resist permanent eradication by antibiotics. The lack of efficacy of current therapies is believed to be due to the formation of P. aeruginosa antibiotic resistant biofilms in the CF airways. However, little is known about biofilm formation on human airway epithelial cells. Thus, we designed a continuous flow, live cell imaging system to grow P. aeruginosa biofilms on glass or on the apical surface of polarized human cells. CFBE41ohuman airway epithelial cells homozygous for the ∆F508 mutation (CFBE), and CFBE41o-cells complemented with wt-CFTR (CFBE+wt-CFTR) were grown as polarized monolayers at an air-liquid interface and GFP-labeled P. aeruginosa were applied to the apical surface of the cells. After 6 hours biofilms did not form on glass but large bacterial macrocolonies developed on CFBE cells. Surprisingly, biofilms also developed on CFBE+wt-CFTR cells: however, P. aeruginosa biomass on CFBE+wt-CFTR cells was significantly reduced compared to CFBE cells (0.60±0.12 µm 3 /µm 2 vs 1.58±0.20 µm 3 /µm 2 ). The minimal bactericidal concentration for Tobramycin was >10,000 µg/ml for P. aeruginosa grown on CFBE and CFBE+wt-CFTR cells. These results explain why Tobramycin given clinically, which is~1,000 µg/ml in bronchoalveolar fluid collected from CF patients, fails to eradicate the persistent infection by P. aeruginosa. P. aeruginosa also formed biofilms on human bronchial epithelial cells, which produce mucus. Indeed, biomass was similar on CFBE+wt-CFTR cells, which do not produce mucus, and human bronchial epithelial cells. To determine if human airway epithelial cells secrete factors that facilitate biofilm formation, conditioned medium was collected from CFBE and CFBE+wt-CFTR cells and applied to P. aeruginosa grown on glass. Under these conditions, biofilm formation was dramatically facilitated, with a biomass of 0.02 ± 0.01 µm 3 /µm 2 when P. aeruginosa was grown in conditioned medium collected from CFBE+wt-CFTR cells and 0.11 ± 0.02 µm 3 /µm 2 when grown in conditioned medium collected from CFBE cells. The secreted factor(s) were greater than 5 kDa, as assessed by size exclusion. Our results suggest that human airway epithelial cells secrete factor(s) that facilitate biofilm formation by P. aeruginosa, in the presence and absence of mucus, and that secretions from CF cells are more effective in promoting biofilms than secretions from non-CF cells. The ability of non-CF cells to facilitate biofilm formation appears to contradict the observation that non-CF lungs do not harbor P. aeruginosa. We propose that the innate immune response in the non-CF lung effectively eliminates P. aeruginosa from the lung, thus, biofilms do not form even thought the airway epithelial cells secrete factors that have the potential to facilitate biofilm formation. However, in CF when mucociliary clearance is compromised, P. aeruginosa accumulates in the lung and secretions by CF airway epithelial cells greatly enhances the formation of drug resistant biofilms (supported by the NIH (RO1 AI51360, RO1-HL-074175, P20-RR018787, T32-DK-07301), and the CFF (STANTO97RO, ANDERS06FO).

Ma, L. 1 ; Parsek, M.R. 2 ; Wozniak, D.J. 1 1. Microbiology and Immunology, Wake Forest University Health Science, Winstonsalem, NC, USA; 2. University of Washington, Seattle, NC, USA Bacteria in natural, industrial and clinical settings live in surface-associated communities termed biofilms, which are the source of persistent infection and resistance to antimicrobial treatment. Biofilm development is initiated by the attachment of planktonic cells to a surface, followed by formation of microcolnies, and finally disperses swimming cells from microcolnies to occupy a new surface. To maintain the community structure, bacteria in a biofilm are usually enmeshed in an extracellular polymeric matrix, which consists of nucleic acids, proteins, and polysaccharides. Exopolysaccharides (EPS) have been known as a component of the biofilm matrix for years. However, little is know about how the EPS matrix forms and develops and whether an EPS can form a matrix independently. In the present report, we use lectins that specifically detect the mannose or galactose structure in Pseudomonas aeruginosa Psl EPS. This technique allows us to visualize Psl EPS on the bacteria cell surface and in the biofilm matrix. Our results indicate that Psl EPS is likely anchored on the cell surface in a helical pattern and clearly forms a matrix, which holds bacteria in the biofilm and on the surface. In a flat multiple-layer biofilm, Psl EPS is equally distributed in the entire biofilm. However, microcolonies reveal peripheral staining of Psl EPS with minimal staining of matrix in the center of the microcolonies. Instead, this region has swimming cells indicating a biofilm development stage prior to dispersion. More strikingly, this area also has concentrated dead cells and/or extracellular DNA, which fills up the viod spaces in the lower center of microcolonies at the stage prior to dispersion. These data provides a plausible mechanism for how P. aeruginosa sacrifices a portion of cells to make void spaces and free another portion of cells for future dispersion. In addition, our data also show that Psl EPS matrix and DNA matrix are not overlapping, which suggests that these two components of the matrix work coordinately to encase bacteria in the biofilm. Finally we show that the Psl EPS matrix is present in biofilms formed by mucoid P. aeruginosa, and formation of the Psl EPS matrix is independent of the production of alginate and Pel, two EPS that also contribute to P. aeruginosa biofilm formation. Overall, our data indicates that Psl EPS functions as a primary scaffold, holding biofilm cells together in the matrix. Moreover, our data along with published literature suggest a possible model for how P. aeruginosa biofilms persist in cystic fibrosis patients. 1 1. Clinical Microbiology, Rigshospitalet, Copenhagen, Denmark; 2. BioCentrum, Technical University of Denmark, Kgs. Lyngby, Denmark; 3. Department of Paediatrics, Copenhagen Cystic Fibrosis Centre, Rigshospitalet, Copenhagen, Denmark Pseudomonas aeruginosa predominates chronic lung infections in patients with cystic fibrosis (CF). A hallmark of chronic P. aeruginosa lung infections in CF patients is the presence of mucoid P. aeruginosa biofilms surrounded by numerous polymorphonuclear leukocytes (PMNs) in the lower airways. How P. aeruginosa escapes the bactericidal PMNs is not fully understood. Several virulence factors of P. aeruginosa is controlled by Quorum Sensing (QS); a density dependent mode of inter-bacterial communication based on signal transmitter molecules. Active QS is present during chronic P. aeruginosa lung infections in CF patients and we have previously demonstrated a QS-regulated tolerance of biofilm bacteria to the antimicrobial properties of the PMNs. The precise QS-regulated effect on the PMNs is, however, unknown. In the present study, further elaboration using flow cytometry and microscopy revealed that QS-competent P. aeruginosa (PAO1 and ∆rhlI lasI complemented with C4-HSL and 3-oxo-C12-HSL) induces rapid necrosis of the PMNs. This mechanism was also observed in in vitro biofilms and in mouse lungs infected with P. aeruginosa embedded in alginate. By using HPLC fractionation the toxicity could be ascribed to a single substance. Using LC-MS and 2D NMR this substance was identified as rhamnolipid B. In accordance, concentrations of rhamnolipid exceeding 200 µg/ml were found in the toxic supernatants from wild-type P. aeruginosa (PAO1) whereas non-toxic supernatants from QS-deficient knock-out mutants (∆rhlR lasR and ∆rhlI lasI) and from ∆rhlA mutants contained low amounts of rhamnolipids approaching 0 µg/ml. Our results demonstrate the potential of the QS system to facilitate infections with P. aeruginosa by utilizing rhamnolipid to disable a major first line of defense of the host -the PMNs. Furthermore, our study emphasizes the inhibition of QS as a target for the treatment of infections with P. aeruginosa.

Carlsson, M. 1,2 ; Pettersson, A. 1 ; Andersson, C. 1 ; Wieslander, J. 4 ; Eriksson, L. 3 ; Segelmark, M. 2 ; Hellmark, T. 2 1. Dept of Microbiology, Immunology, and Glycobiology, University of Lund, LUND, Sweden; 2. Department of Nephrology, Lund University, Lund, Sweden; 3. Heart Lung Division, the Cystic Fibrosis Center, Lund University Hospital, Lund, Sweden; 4. Wieslab Analys AB, Lund, Sweden The clinical consequence of chronic Pseudomonas (P) aeruginosa colonization in cystic fibrosis (CF) varies between individuals for unknown reasons. Auto-antibodies against bactericidal/ permeability increasing protein (BPI-ANCA) are associated with poor prognosis in CF. We hypothesize that there is a correlation between the presence of BPI-ANCA, the biological properties of the colonizing bacteria and the clinical conditions of the hosts. We have compared isolates of P aeruginosa from two groups of CF patients: one with positive serum levels of BPI-ANCA and deteriorating lung disease, and one with negative BPI-ANCA levels and stable clinical conditions. Epithelial cells (A549) and isolated polymorphonuclear granulocytes (PMNs) were stimulated with the clinical isolates and cell death was analyzed with flow cytometry. Interleukin-8 (IL-8) released into the supernatant was measured by ELISA. We found that the ANCA associated strains in most cases showed a pyocyanin negative phenotype. These strains also induced less inflammatory response than the non-ANCA associated strains as shown by the number of necrotic cells and IL-8 release, yet elevated compared to control. We conclude that colonization with strains of P aeruginosa that induce a weak inflammatory response is associated with unfavourable outcome in CF. We speculate that inadequate control of pathogen proliferation through an insufficient inflammatory response results in a slowly increasing number of bacteria and accumulation of dying PMNs in the airways, contributing to the progressive lung disease seen in many CF patients. 2 1. Pulmonary Critical Care, Northwestern University, Chicago, IL, USA; 2. Microbiology/Immunology, Northwestern University, Chicago, IL, USA; 3. Pulmonary Medicine, Children's Memorial Hospital, Chicago, IL, USA; 4. Pediatrics, Children's Memorial Hospital, Chicago, IL, USA Purpose: Most Cystic Fibrosis (CF) patients are infected with Pseudomonas aeruginosa (PA) and have progressive loss of lung function. Among individual patients, however, there are marked differences in rates of lung function deterioration. Although the reasons for this variability are not completely understood, it is likely that microbiological factors play a role. Type III secretion in Pseudomonas aeruginosa (PA) isolates from non-CF patients have been associated with poorer outcome. The aim of this study was to determine whether there is an association between type III secretion properties and deterioration in lung function in CF.

Methods:

We prospectively enrolled CF children and adults over 3 years. Demographics, clinical characteristics, spirometry and a respiratory culture were obtained at the first visit. The age at time of first postive PA culture was also determined. Subsequently, spirometry and respiratory cultures were obtained and clinical characteristics recorded every 6 months until December 2006. From each sputum culture 5 individual isolates were selected for evaluation and type III secretion was evaluated for each isolate by western blot analysis. The children were subcohorted into those newly infected (1st positive PA culture) or those chronically infected (minimum 2 sputum PA (+) >1 year duration).

Results: There were 100 patients evaluated, of which 31 were adults and 69 were children. Ninety percent of our population cohort was Caucasian and 53% were female. Overall 63% of the patients were on TOBI and 72% on macrolides chronically. The mean age at time of a first PA positive culture was 11 years for the chronically infected subjects(adults and children) and 6 years for the 1st time infected subcohort. A total of 2409 sputum samples were collected for the current study. The prevalence of type III secretion properties for the overall population was 21.11%. Foe adults, 73 of 807 (9.7%) PA isolates were type III secretion positive compared to 346 of 1602 isolates (26.8%) in chronically infected children and 48 of 90 isolates(53.3%) from newly infected children. At study entry the mean FEV1 % predicted was 68 +/-24 for the whole cohort. For the adult subcohort the FEV1% predicted was 60% +/-26%, for chronically infected children it was 72% +/-21% and newly infected children is was 82%+/-22%. The average annual rate of FEV1 % change was -0.67% for adults, -2.42% for chronically infected children and +4.62 for newly infected children. Analysis on the relationship between type III secretion PA isolates and FEV1% decline is ongoing and will be presented at NACF.

Conclusion: Prevalence of type III secretion in PA isolates from CF patients decreases with age.

There is a slower decline in FEV1% in chronically PA infected adults compared to chronically PA infected children.

The relationship of type III secretion to FEV1 % change and pulmonary exacerbations will be presented at NACF. 

A limited number of bacterial species -in example Pseudomonas aeruginosa, Staphylococcus aureus, Burkholderia cepacia complex, Stenotrophomonas maltophilia -is typically found in the CF lung. These facultative anaerobic bacteria exhibit the capability to switch from aerobic to anaerobic metabolism, which enables them to survive in the hypoxic environment in the CF lung [1] . Absence of oxygen would also favour the growth of fastidious anaerobes and, indeed, sequencing of 16S rRNA resulted in the detection of strict anaerobic species in CF sputum [2] . Thus, we identified and quantified strict anaerobes in CF sputum using conventional anaerobic microbiological methods, examined the recovery rate of identical species in sputum samples from the same patients, and determined the antibiotic susceptibilities. 92 sputum samples were collected from 8 CF children and 31 CF adults (mean age 25.6 ± 8.5 yrs). Samples were incubated aerobically on Columbia agar, supplemented with 10% sheep blood, and anaerobically on Brain Heart Infusion agar and Schaedler agar supplemented with 5% mutton blood, for up to 7 days. Bacteria were identified (RapID ANA II identification system, Remel, Lenexa, KS) and cfu's were determined by dilution plate counting. Fastidious anaerobes (141 strains) were submitted to E-test ® susceptibility testing (AB Biodisk, Solna, Sweden) using the anaerobically active antibiotics ceftazidime, clindamycin, meropenem, metronidazole, and piperacillin/tazobactam.

In 75.0% of the patients, the following strict anaerobes were detected with a mean of 6.3x10 6 cfu/ml (range 2.5x10 4 to 2.5x10 7 cfu/ml): Peptostreptococcus spp., Clostridium spp., Actinomyces spp., Prevotella spp., Wolinella spp., Propionibacterium spp., Streptococcus spp., Lactobacillus spp., Gemella spp., Bacteroides spp. and Eubacterium spp., whereas P. aeruginosa, S. aureus and B. cepacia revealed 7.1x10 7 cfu/ml (range 5.0x10 5 to 3.5x10 8 cfu/ml). In 30 sputum samples contaminated with P. aeruginosa together with strict anaerobes 5.2x10 7 ± 7.8x10 7 cfu/ml P. aeruginosa were counted, whereas in 9 samples with P. aeruginosa without strict anaerobes only 1.7x10 7 ± 1.8x10 7 cfu/ml were found (p=0.03). For S. aureus, no significant difference was observed. Identical strict anaerobic species were detected in 12 out of 19 patients with two or more repeated sputum samples (63%). E-test® sensitivity testing for strict anaerobes yielded high sensitivity for meropenem (only 5.7% resistant strains), piperacillin/tazobactam (20.6%), and clindamycin (24.8%), but not ceftazidime (49.0%) or metronidazole (50.4%).

High numbers of strict anaerobes are present in the majority of CF patients. Possibly, strict anaerobes promote growth of P. aeruginosa but not S. aureus. The high persistence of identical anaerobic strains reflects chronic lung infection and may be caused by their increased resistence against standard antibiotics such as ceftazidime.

[ 

In sputum in the CF lungs bacteria such as Pseudomonas aeruginosa have to metabolize anaerobically [1] . For anaerobic energy generation, P. aeruginosa can use nitrate and arginine, but also pyruvate which is produced from glucose via anaerobic glycolysis [2] and can be metabolized to lactate and vice versa. In order to investigate if P. aeruginosa may benefit from externally produced lactate, we measured the concentration of lactate in P. aeruginosa, Staphylococcus aureus, Burkholderia cenocepacia and polymorphonuclear neutrophils in vitro and in CF sputum.

In sputum samples of 25 CF patients and in neutrophils (3 x 10 7 /ml) from healthy donors lactate concentrations were determined. In addition, P. aeruginosa (starting with 1 x 10 7 cfu/ml in tryptone soy broth), S. aureus (3 x 10 6 cfu/ml), and B. cenocepacia (8 x 10 6 cfu/ml)were grown aerobically (0 through 24 hrs) and anaerobically (1 through 4 days). L-Lactate was measured spectrophotometrically (detection limit: 0.1 mmol/L), and total lactate gaschromatographically. Aerobic and anaerobic gene expression of P. aeruginosa strain PAO1 was determined using Affymetrix® microarrays.

Lactate concentrations in CF sputum amounted to 3.0 ± 3.1 mmol/L (range 0.2 to 14.1 mmol/L). Concentrations were similar in sputum samples colonized with P. aeruginosa, S. aureus (3.3 ± 3.7 vs. 2.8 ± 2.3 mmol/L, p=0.67) and B. cenocepacia (1.9 mmol/L). Neutrophils produced 3.2 mmol/L. In all samples exclusively L-lactate was found. During in vitro experiments, P. aeruginosa did not generate any lactate at all, neither aerobically nor anaerobically. In contrast, anaerobically grown S. aureus produced up to 2.8 mmol/L lactate, and B. cenocepacia up to 10.0 mmol/L. A P. aeruginosa suspension [1x10 7 cfu/ml] spiked with 10 mmol/L L-lactate did not change its concentration, indicating that P. aeruginosa does not metabolize lactate. Similar results were obtained in our gene chip experiments: after three days of anaerobic growth, the genes encoding for the lactate dehydrogenases were downregulated (PA0927 ldhA -10.78 fold, PA2382 lldA -2.81 fold) or unchanged (PA4771 lldD 1.2 fold). In contrast, the genes encoding for pyruvate decomposition to acetyl CoA (PA3416 and PA3417, both encoding for pyruvate dehygrogenase E1 components) were upregulated by 208 and 47 fold, respectively.

We could demonstrate that P. aeruginosa does not benefit from externally produced lactate. We confirmed the important role of pyruvate metabolism for anaerobic P. aeruginosa energy generation. Whether lactate production of neutrophils, S. aureus or B. cenocepacia contributes to CF lung pathophysiology still remains to be investigated.

References: [1] 

MacLeod, D. 1 ; Barker, L. 1 ; Gurgel, J. 1 ; Kenney, T. 1 ; Burns, J. 2 ; Baker, W. 1 1. Gilead Sciences, Inc., Seattle, WA, USA; 2. University of Washington, Seattle, WA, USA Antibiotic resistance may severely limit therapeutic options in individuals with cystic fibrosis (CF) or bronchiectasis. Because of frequent antibiotic treatment courses, resistance continues to emerge, even to newer agents. Treatment with multiple antibiotics in a single aerosol formulation may be a promising approach to slow development of resistance. Fosfomycin is a phosphonic acid antibiotic that is bactericidal against both Gram positive and Gram negative organisms. Fosfomycin inhibits the first committed step in the synthesis of peptidoglycan, suggesting cross resistance to other cell wall acting antibiotics will not occur. The aminoglycoside tobramycin is one of the most commonly used antimicrobials in CF, with potent activity against Gram negative bacteria and the majority of Staphylococcus aureus isolates. A 4:1 (wt/wt) fixed combination of fosfomycin:tobramycin (GS-9310/11) was used to determine the in vitro susceptibilities of a panel of respiratory pathogens: CF Pseudomonas aeruginosa (100), S. aureus (16), Haemophilus influenzae (16), Stenotrophomonas maltophilia (17) and Burkholderia cepacia complex (20), including minimal inhibitory concentration (MIC) and time-kill experiments in the absence and presence of 2% porcine mucin. Synergy was evaluated using the checkerboard method, and spontaneous resistance mutation frequencies were determined in antibiotic-containing agar (4X, 8X and 16X MIC). In vivo drug efficacy was examined using a rat agar bead pneumonia model of either P. aeruginosa or S. aureus. All experiments compared GS-9310/11 to fosfomycin and tobramycin as single agents. GS-9310/11 had a lower MIC 90 than tobramycin for the S. aureus strains, 75% of which were methicillin resistant (MRSA). For P. aeruginosa, GS-9310/11 had a lower MIC 50 and MIC 90 than fosfomycin alone, but tobramycin was more active than either. For H. influenzae and S. maltophilia, GS-9310/11, fosfomycin and tobramycin had similar MIC 50 and MIC 90 . B. cepacia complex were resistant to all three drugs. Results in the presence of mucin were similar. Time-kill studies showed a more rapid and prolonged killing of S. aureus and P. aeruginosa by GS-9310/11 compared with either agent alone at the same drug concentrations. GS-9310/11 was bactericidal and exhibited concentration-dependent killing. Synergy studies showed no antagonism between fosfomycin and tobramycin, and the majority of P. aeruginosa and all of the S. aureus tested demonstrated indifference for the combination. At 4X MIC concentrations the mutation frequency of GS-9310/11 was at least 2-3 logs lower than tobramycin and 2-6 logs lower than fosfomycin alone for S. aureus. For P. aeruginosa the mutation frequency of GS-9310/11 was 2-3 logs lower than fosfomycin and 1-2 logs lower than tobramycin. In the rat pneumonia model, GS-9310/11 and tobramycin alone demonstrated bactericidal killing of P. aeruginosa; both were more active than fosfomycin alone. In vivo killing of S. aureus by GS-9310/11 was also demonstrated. GS-9310/11 appears to have advantages over single agents for the treatment of both Gram positive and Gram negative bacterial lung infections in CF and bronchiectasis.

Britton, L.J. 1 Antibiotic resistance is becoming a major problem in the treatment of pulmonary exacerbations in cystic fibrosis (CF). Organ-isms that are resistant to multiple antibiotics infect the airways of an estimated 25-45% of adults with CF.1 As a result of the growing resistance in CF patients, many centers have been performing synergy testing of sputum cultures in addition to the conventional culture and sensitivity testing. The purpose of our investigation was to determine if antibiotic synergy studies and the use of synergistic combinations of antibiotics improve therapeutic outcomes in cystic fibrosis patients with acute pulmonary exacerbations.

Methods: This study is a retrospective chart review of CF patients who had antibiotic synergy testing performed while hospitalized for respiratory tract infections. Eligibility criteria included CF patients hospitalized with respiratory tract infections from 1998 to 2005. Laboratory data from each hospital admission was reviewed for synergy among those antibiotics commonly used against Pseudomonas aeruginosa. A review of medical charts ascertained each patient's pulmonary function (determined by FEV1 before and after antibiotic therapy), weight Z-score, organism cultured from sputum sample, time to next hospital admission, and the antibiotic(s) actually prescribed. Primary endpoints were determined to be the change in FEV1 and the time to next admission.

Results: Four hundred seventy-five hospital admissions were analyzed. A total of 48 Cystic fibrosis patients, age birth to 36 years, were included in the study. Patients receiving antibiotic synergy experienced a significant decrease in mean time to next admission (50 days with synergy vs. 68 days without synergy, p=0.004). No significant difference was found in the change in FEV1 before or after antibiotic therapy, with an increase of 8.72% with synergy vs. 9.28% without synergy (p=0.585). Patients infected with non-mucoid P. aeruginosa experienced 60 days to the next hospital admission, while patients infected with mucoid P. aeruginosa experienced 56 days to the next hospital admission (p = 0.45). No statistically significant difference was observed between the synergy and non-synergy groups in regards to nutritional status and lung function prior to antibiotic therapy.

Speculations: Antibiotic resistance in cystic fibrosis patients increases the morbidity and mortality caused by this disorder with each exacerbation the patient experiences. Synergistic antibiotic therapies should improve patient outcomes through more efficient bacterial eradication and increased time to next hospital admission; however, we were unable to substantiate this assumption based on the results from this study. Synergistic antibiotics did not show an improvement in the therapeutic outcomes of days to next admission or change in FEV1. Background: Expectorated sputum (ES) technique is currently the most frequently used method for routine assessment of lower airway infection in patients with cystic fibrosis (CF). Induced sputum (IS) using hypertonic saline (HS) has been successfully used in CF patients unable to produce sputum spontaneously, but only limited data are available comparing the diagnostic yield of expectorated versus induced sputum in CF patients. While ultrasonic nebuliser have been used in the majority of studies, new high output jet nebulisers may offer a suitable alternative technique to induce sputum expectoration.

Aim: To assess the feasibility of sputum induction using a Pari e-flow nebulizer and to compare the diagnostic yield of IS and ES in children with CF .

Methods: This is a preliminary report of an ongoing study in routine clinical care in sputum producing children with CF. ES is being obtained before sputum induction. Subsequently, sputum induction is performed using stepwise inhalation of nebulized 5 ml of 3%, 4% and 5% hypertonic saline with an e-flow nebulizer (Pari, Starnberg, Germany). Lung function is assessed by portable spirometry before the procedure and after inhalation of each saline concentration.

Results: So far 40 CF patients (18 females) with a mean age of 13.6 years (range 8 to 18 years) and FEV1 between 38-121% predicted (mean 82%) have been included in the study. All subjects provided ES samples and all produced sputum after induction. Discrepancies in cultures between ES and IS samples were seen in 9 cases (22.5%). In 6 cases additional CF pathogens were found in IS samples, whereas in 2 cases ES yielded additional organisms compared to IS. The number of distinct pathogens was similar in the remaining patient, but different bacteria were found with the 2 techniques. In 4 cases P. aeruginosa was detected only with one of the 2 techniques (2 ES versus 2 IS).

The spectrum of side effects was similar to previous reports using other nebulizer systems, with throat irritation being the most common adverse event of the IS technique. All patients who had symptoms during the procedure became asymptomatic at the time of discharge from the clinic 20-30 minutes after the procedure. 4 patients did not finish all 3 cycles of sputum induction procedure due to symptoms of shortness of breath and/or drop in FEV1>20% from baseline (in 3 patients); all reversed after salbutamol inhalation. Vomiting occurred in one patient. 3 patients refused to complete the procedure due to unpleasant taste and/or throat irritation.

Conclusion: These preliminary results show discrepancies between expectorated sputum and induced sputum cultures in a significant proportion of CF patients. This may be explained by the previously described regional differences in lower airway infection in CF airways rather than by a higher diagnostic yield of one of the techniques. The results also demonstrate that the e-flow system is an efficient and safe device for sputum induction in sputum producing children with CF.

Seidler, M.J.; Salvenmoser, S.; Müller, F.C. Dept. Pediatrics III, University Heidelberg, Pediatric Pulmonology, Cystic Fibrosis Centre & Infectious Diseases, Heidelberg, Germany Background: The preferred growth form of bacteria is a biofilm. S. aureus, H. influenzae, and P. aeruginosa can produce an extracellular matrix (ECM) with implications in cystic fibrosis (CF) lung disease. The biofilm can protect against host defenses and antimicrobials. A. fumigatus is a frequent colonizer of the CF respiratory tract and can cause Allergic bronchopulmonary aspergillosis (ABPA). While antifungals in vitro are active against A. fumigatus, in vivo antifungal therapy is often complicated or resistance is observable.

The aim of this study was to investigate the ability of A. fumigatus to form a biofilm-like matrix in vitro on polystyrene (PS), human bronchial epithelia cells (16HBE) and human bronchial epithelia cells with F508del/F508del (CFBE41o-).

Methods: A. fumigatus ATCC #9197 was incubated in RPMI at different pH and concentrations of FBS. Temperature, production time and different flow conditions were varied on PS, 16HBE and CFBE41o-. Dry weight measurement and antifungal drug susceptibility testing was performed. Scanning electron microscopy (SEM) and confocal scanning laser microscopy (CSLM) images were analyzed.

Results: The thickest biofilm was produced on PS with RPMI (+10% FBS, pH=6.5) at 35°C for 72h slightly rocking. Biofilm dry weight on PS was 2.2 mg after 4 h and 8.3 mg after 72 h. The dry weight of produced biofilm exceeded 7.4 mg on 16HBE and 7.7 mg on CFBE41o-cells after 72h of biofilm production. There was no significant difference in dry weight increase between the cell lines and PS. Planktonic A. fumigatus was susceptible to itraconazole (0.125µg/ml), voriconazole (0.063µg/ml) and amphotericin B (1µg/ml). Aspergillus in biofilm was resistant against all drugs (>32µg/ml). The SEM pictures displayed a network of hyphal structures and matrix at 12h. Characteristic flow channels were observed at 72h. CSLM images displayed conidia and hyphal structures embedded in matrix formations. A-Alexafluor dyed polysaccharides of the cell wall and of the ECM in the biofilm. Three dimensional constructs of the CSLM pictures displayed biofilm on 16HBE and proofed viability of the cells after 48h co-incubation. Differences in biofilm production between 16HBE and CFBE41o-were not significant.

Conclusions: A. fumigatus is able to form a biofilm structure in vitro on PS, 16HBE and CFBE41o-. A biofilm-like matrix produced by A. fumigatus was evidenced by dry weight measurement, SEM, CSLM and antifungal drug resist-ance in comparison to planktonic cells. Potential clinical implications of A. fumigatus biofilm formation in vivo require further attention and investigations.

Etherington, C. 1 ; Peckham, D. 1 ; Conway, S. 1 ; Hall, M. 2 ; Denton, M. 2 1. Seacroft Hospital, Regional Adult CF Unit, Leeds, United Kingdom; 2. Microbiology Department, Leeds Teaching Hospitals, Leeds, United Kingdom Susceptibility testing results are not predictive of clinical response to antibiotic therapy in chronic Pseudomonas aeruginosa infections in cystic fibrosis. We assessed the impact of reducing the number of routine susceptibility tests performed on clinical outcome in these cases. In June 2006 we introduced a protocol of limiting susceptibility tests to P. aeruginosa isolates obtained from respiratory samples taken at the commencement of antibiotic therapy, when there was evidence of clinical failure, or routinely if not tested in the previous three months. At all other times, isolates were identified and reported as normal but P. aeruginosa isolates were not subjected to susceptibility tests. Between 1st June and 30th November 2006 P. aeruginosa, was isolated on at least one occasion from 193 patients attending our Adult CF Unit. In this six month period we reduced the number of susceptibility tests by 56% (from a projected 2,231 tests on 872 samples to an actual 972 tests on 427 samples). This resulted in projected savings of $4,777 in consumables and 170 hours (costed at $8,871) of laboratory staff time per annum, a total saving of $13,649 (£6,500) per annum. We assessed the response to intravenous antibiotic treatment between the study period in 2006 and the same period in 2005. No significant differences in median change of FEV1, FVC, CRP, white cell count, weight, or duration of intravenous antibiotics were observed. For CF units sending regular, routine sputum samples, a reduction in the number of susceptibility tests performed in cases of chronic P. aeruginosa can be carried out without impacting on clinical outcomes. We report our preliminary results for a total of 233 strains obtained from 112 patients (85 patients harbored more than 2 strains). These patients were classified in 3 groups of respiratory insufficiency according to their FEV1: severe in 31 patients (FEV1<40%), moderate in 36 patients (FEV1: 40-60%) and mild in 45 patients (FEV1>60%).

The clonal distribution was analyzed for the different strains of SA (1 to 6) isolated from patients sputum. These strains were analyzed for their antibiotic susceptibility and typed by pulsed-field electrophoresis gel (PFGE) after SmaI digestion of chromosomal DNA.

Thirty seven patients (33%) were colonized with MRSA. Nine patients were both colonized with MRSA and MSSA. Among MRSA strains, 43/61 (70%) were also resistant to more than three other antibiotic family. Strains harboring minor differences in the banding pattern (>80% similarity as assessed by the Dice coefficient) were considered clonal. Our results show that 83% of the patients were colonized with a single persistent strain during the year of follow-up. Consecutive isolates with different PFGE profiles were obtained from only 14/80 patients (17%). PFGE analysis revealed that MRSA isolated from 27 patients were grouped in 8 clusters. These results revealed a possible clonal relationship between MRSA isolated from different patients with CF.

We did not find any difference in the distribution of SA strains in our CF patients among the 3 groups of respiratory insufficiency.

The study is ongoing in our adult CF population and in Necker pediatric CF centre in Paris.

Giusti, R. 1 ; Furfaro, S. 2 1. Pediatrics, Long Island College Hospital, Brooklyn, NY, USA; 2. Research, Lumina Fund, New York, NY, USA Palivizumab(Synagis) is a humanized monoclonal antibody to RSV. The 2006 Redbook acknowledges that some patients with Cystic Fibrosis may be at increased risk of RSV infection but that there is insufficient data to determine the effectiveness of this therapy.

The objective of this study was to assess: 1)Practice patterns of CF physicians in the US and Canada 2)The severity of RSV disease in CF infants during the past RSV season.

3)If there is a standard of care concerning the use of Synagis for CF infants.

Methods A questionnaire was developed using a web-based commercial vendor. An embedded web link was distributed via an e-mail sent to all US and Canadian Pediatric CF Center Directors. Respondents clicked on a link to respond to the survey and results were automatically tabulated in real-time.

Completed responses were received from 84 center directors (73 US and 11 Canadian) for a response rate of 63% in US and 44% in Canada. Most responders (70%) have prescribed Synagis for infants with CF in the first RSV season, however only 38% routinely prescribed Synagis for all infants with CF. Only 35% expressed having had difficulty in obtaining insurance approval for this medication. Many physicians indicated that Synagis was frequently prescribed by the general pediatrician and that infants living at long distances from a CF Center may be hospitalized at local hospitals. These issues may affect the accuracy of the data and result in an underestimate of actual Synagis prescription and hospitalization rates.

There were 456 infants diagnosed with CF in the past year and 172 of these infants were reported as symptomatic. There were 74 CF infants (16%) reported as having a documented RSV infection and 13 (18%) of these infants had received Synagis. There were 31 infants (42%) with documented RSV infection who responded to outpatient management.There were no deaths but 40 patients were hospitalized and 8 of these had received Synagis. Of the 3 infants that required admission to an ICU 2 had received Synagis. Of the 12 patients who were noted as having persistent Chest X-ray changes 5 had received treatment with Synagis. There were 36 infants with persistent wheezing and 13 of these infants had been treated with Synagis.

Conclusions Despite the limitations of a retrospective survey, this data demonstrates that RSV can cause significant and prolonged pulmonary disease and is a significant precipitating factor resulting in hospitalization of CF infants. The data also notes that many CF infants infected during the past RSV season have a mild illness and respond to outpatient management. A surprising finding of this survey is that infants at CF Care Centers where Synagis was prescribed for all CF infants continued to have significant RSV related hospitalizations, persistent wheezing or prolonged chest X-ray abnormalities.

There are different opinions among CF physicians concerning the routine use of Synagis and currently the data suggests that there is not a standard of care concerning the prophylaxis of CF infants with Synagis.

This results of this survey should encourage physicians to prospectively study the efficacy of Synagis prophylaxis in preventing hospitalization, persistent wheezing and chest x-ray abnormalities in CF infants.

Milani, A. 1 ; Cisbani, G. 1 ; Macchi, R. 1 ; Vidal-Aroca, F. 2 ; Bertoni, G. 1 1. Biomolecular Sciences and Biotechnology, University of Milan, Milano, Italy; 2. Basilea Pharmaceutica Ltd., Basel, Switzerland With ever increasing frequency, we now observe several examples of bacteria being resistant to every clinically available drug. Therefore this urgently calls for the development of novel and improved antibiotics that may escape the extant mechanisms of bacterial resistance. One recent and promising development of the genome-wide search for target functions for antibiotics led to the identification of essential genes of pathogens by interfering antisense RNAs. The first step is the construction of shotgun antisense libraries (SALs) as follows. Genomic DNA is extracted from the bacterial strain of interest, fragmented by shearing into short pieces of DNA, blunt-ended and cloned in an expression vector under the control of a regulatable promoter. The library is then reintroduced into the cognate bacterial strain and screened by replica plating colonies both in the presence and the absence of an inducer of the vector promoter. By this method, insert sequencing of clones showing conditional growth phenotypes is expected to lead to the identification of essential genes that can be silenced via antisense RNA activity. We adopted this technology in order to generate a panel of essential functions of the cystic fibrosis-related opportunistic pathogen Pseudomonas aeruginosa. So far, we tuned the protocol for SAL generation in P. aeruginosa and identified a number of sequences conferring different levels of growth inhibition We are now characterizing these putative antisense RNAs in order to define the minimal sequence able to cause the toxic effect and, on the other hand, to understand the cellular role of their targets. Pseudomonas aeruginosa (PA) is an extremely versatile microbe with a vast array of pathogenic and metabolic mechanisms that allow it to form persistent infections in select patient populations, especially patients with Cystic Fibrosis. An ineffective immune response is considered partially to blame for failure of PA eradication. We have discovered that some strains of PA express peptidylarginine deiminase (PADI) activity. PADI is an enzyme that post-translationally modifies peptidylarginine to peptidylcitrulline with ammonia as a byproduct. Our lab has shown that human PADI can modulate the immune system through downregulation of TLR4 and IKK-gamma signaling. Characterization of the PA PADI will offer insights into a completely novel method of immune modulation by Pseudomonas aeruginosa. Using a widely published colorimetric assay for PADI activity we have found the specific activity of crude PA cell lysates is very low. However this is similar to the only other known prokaryotic PADI described in Porphyromonas gingivalis. A protein homology search with the Porphyromonas PADI has revealed the likely genetic locus of Pseudomonas PADI in a 4.5 kb operon that appears in the genome of the pathogenic Pseudomonas isolate PA14. This operon appears to contain two candidate genes for PADI activity based on conserved motif searches (PADI 1 and PADI 2). Both have been cloned into expression vectors, partially purified by affinity tag technology and tested in the colorimetric PADI assay. Curiously PADI 1 autocitrullinates itself while PADI 2 has not shown activity. Reaction conditions and substrate specificities for PA-PADI are not like either the human or Porphyromonas PADI. Furthermore analysis of up to 64 clinically diverse strains has demonstrated 15% carry the gene for PADI 1. This work is only the second description of PADI activity in any prokaryote. PA PADI could clearly have a dramatic impact on the local inflammatory milieu if it can access the same targets as human PADI. The description of this activity in Pseudomonas will advance our knowledge of the human-pathogen relationship and give insight into new therapies. which functions by translocating toxins into the cytoplasm of host cells. These toxins cause disease by damaging the surrounding host tissue, promoting dissemination of the organism and paralyzing the phagocytic mechanism of macrophages. PcrV is a factor required for the translocation of the toxins. The bases of these studies were to evaluate PcrV as a protective antigen in "P. aeruginosa" pneumonia and CLDC as a vaccine adjuvant.

Methods: Mice were vaccinated 1, 2, or 3 times with the PcrV antigen combined with either aluminum hydroxide (alum) or CLDC by various routes of administration. Efficacy of vaccination was evaluated by challenge with "P. aeruginosa" and evaluation of survival and/or measurement of various parameters associated with lung injury.

Results: Increase in median survival time was highly significant when CLDC/PcrV was compared with CLDC or PcrV alone. Following 2 subcutaneous administrations CLDC/PcrV showed an increase in median survival time (58 hours versus 43 hours) and overall survival benefit following 3 intraperitoneal administrations (100% versus 80%). Mice with anti-PcrV antibody levels above 10µg/ml were significantly protected.

Conclusion: The investigators establish the efficacy of CLDC/PcrV vaccines via several parenteral routes of administration compared to no treatment as well as CLDC and PcrV-only controls. Differences were demonstrated between performance of CLDC/PcrV and Alum/PcrV in measures of lung injury, median survival, and overall survival. The results correlated with antibody levels and histological examination of the lung tissues. Importantly, these studies indicate that protection can be achieved against "P. aeruginosa" infection by targeting an antigen associated with the type III secretion system. Background: There has been a recent increase in the number of reported cases of acute renal failure (ARF) in cystic fibrosis (CF). Our group have undertaken a national survey, which measured the incidence risk of ARF in CF patients at between 4.6 and 10.5 cases / 10,000 CF patients / year. We have now conducted a case control study to determine which factors which are associated with an increased risk of ARF.

Methods: In our initial survey we confirmed 24 cases of ARF, in CF patients from 20 UK CF Centres, presenting between 1997 & 2004. Using the UK CF database, we identified sex and age (within 6 months) matched controls. Informed consent was sought from the control patients, or their parents, for access to the case notes and clinical data were extracted. Analysis of risk factors was by conditional logistic regression, using STATA (version 9) and by Fisher's exact test.

Results: There were 24 cases of ARF (12 male, median age 10y, range 4m-32y) and 28 controls (17 male, 9y, 10m-32y). In the group of patients with ARF, 21/24 had received an aminoglycoside at the time of their episode of ARF or in the preceding week, compared with only 3 of the controls for the same time period (p<0.001). The median number of days of aminoglycoside in the year prior to the index case developing acute renal failure was 24 (7-99) for cases and 0 (0-47) for controls. Conditional logistic regression showed that the odds ratio for ARF per each day of aminoglycoside was 1.06 (95% CI 1.01 to 1.11, p=0.02 It is well-known that Pa senses the environment and changes its phenotype. For instance, it produces greater amounts of the extracellular polysaccharide alginate in the CF lung, characterized by a microaerobic environment. Little is known about the changes in protein secretion induced by oxygen limitation in PAO1, the proto-typical Pa laboratory strain. No data are available on this regard about Pa clinical strains. Our work was aimed to study the differential regulation of proteins secreted by Pa strains grown in microaerobic or aerobic conditions. A Pa clinical isolate and PAO1 were grown overnight in aerobiosis and in microaerophilic conditions. The supernatants were collected and proteomic analysis was carried over by two-dimensional capillary chromatography -tandem mass spectrometry (Mauri et al., FASEB J 2005) to evaluate the differential protein expression. In the Pa clinical isolate, we identified 66 proteins down-regulated and 42 proteins upregulated in aerobic conditions in comparison with microaerophilic culture while in PAO1 16 proteins were down-regulated and 30 were up-regulated. 6 proteins were down-regulated and 12 up-regulated both in the clinical and in the laboratory strain. These proteins can mediate different biological functions since they are enzymes, heat shock proteins, chaperones, proteins involved in adaptation, motility and in the transport of small molecules. Among all these proteins, as the alkaline metalloproteinase is associated with tissue invasion not only by causing rupture of epithelial tight-junctions but also by degrading several chemokines, we decided to validate its up-regulation observed in aerobic conditions by zymography. We found that the proteolytic activity of the supernatants of Pa grown in aerobic conditions was higher than in microaerophilic culture both for the laboratory and clinical strain, indicating the functional relevance of data obtained by proteomic analysis.

The identification of proteins differentially regulated in aerobiosis and oxygen limitated conditions in Pa laboratory and clinical strains might be helpful for the knowledge of the mechanisms of colonization and lung damage due to Pa in CF patients. For instance, the validation of the upregulation of the alkaline metalloproteinase in aerobic condition may shed light on these mechanisms of CF lung disease. Further studies are in progress to evaluate the function of other bacterial exoproducts regulated in this model and to extend the analysis to other clinical strains.

Supported by the Italian Cystic Fibrosis Research Foundation (FFC-grant#17/2006), Comitato di Vicenza dell'Associazione Veneta per la lotta contro la Fibrosi Cistica and Azienda Ospedaliera di Verona, Italy. Cystic fibrosis (CF) sufferers are subject to repeated lung infections most commonly with the bacterium Pseudomonas aeruginosa. In spite of antibiotic treatment P. aeruginosa tends to become established giving rise to persistent chronic infection. In the lung environment, the bacterium grows as a highly structured biofilm consisting of a complex community of cells embedded within a self-secreted polysaccharide matrix. Investigations of P. aeruginosa biofilm growth using model strains have elucidated mechanisms which appear to govern the biofilm life-cycle. Our research aims to test whether observation of these mechanisms in planktonic culture can be related to the efficiency of biofilm formation.

Biofilm initiation has been linked to cell motility and in particular to the presence of flagella and pili, which are thought to be important for cell attachment and the formation of microcolonies. On testing a large, genetically diverse group of clinical P. aeruginosa isolates retrieved from the lungs of CF patients we found no definitive correlation between the degree of motility of an isolate in planktonic culture and its ability to form a biofilm in vitro. The development of biofilm architecture has been demonstrated in model systems to be coordinated by the production and secretion of N-acylhomoserine lactone (AHL) quorum sensing molecules. However among the clinical isolates tested we observed no obvious correlation between the amount of AHLs produced in planktonic culture and the extent of biofilm formation. Overall we have found observation of phenotypic characteristics in planktonic culture to be poor predictors of efficient biofilm formation.

A proteomics approach was adopted to provide further insight into the physiology of biofilm growth of P. aeruginosa isolates by comparison with planktonic growth of the same isolates. Two genetically unrelated clinical isolates, demonstrated as being capable of efficient in vitro biofilm formation, were selected from our culture collection on the basis of their diverse phenotypic characteristics when cultured planktonically. One displays both twitching and swimming motilities, is mucoid and expresses AHLs, while the other is non-motile, non-mucoid and no AHLs have been detected in planktonic culture. We have developed a simple flow-through bioreactor to provide sufficient quantities of biofilm for proteomic analysis. Gel-based and gel-free techniques were employed to study protein expression patterns for both biofilm and planktonic cultures of each isolate by mass spectrometry. Proteins specific for each growth phase could be detected and may prove suitable biomarkers for monitoring the physiological status of biofilm forming strains when a larger bank of clinical isolates are examined. Liposomal amikacin (Arikace TM ) is a liposome-encapsulated form of amikacin that is formulated to treat chronic P. aeruginosa infections in cystic fibrosis patients. These liposomes carry a zwitterionic surface charge and are composed of lipids found naturally within the lung. A key aspect of the activity of the formulation is the ability to penetrate to the sites of Pseudomonas biofilm-like growth in the lung. Experiments were designed to investigate the penetration of liposomes into P. aeruginosa biofilms and in vitro activity.

Methods and Results: Model liposomes of the same size and lipid composition as liposomal amikacin (Arikace TM ) were prepared with membrane-associated or encapsulated fluorescent labels, a hydrophobic carbocyanine dye and calcein, respectively. A mucoid strain of Pseudomonas aeruginosa (PA3064) was used to establish biofilms in rectangular optical grade glass flow cells. Biofilms were observed after four days of growth by confocal laser scanning microscopy using a focal plane set to view within the biofilm cluster or outside as a control. Time dependent accumulation of fluorescent liposomes within the biofilms was measured by the spatial distribution of fluorescence intensity in regions within or outside of the biofilm. Images indicated significant penetration of liposomes into the interior of biofilms under these conditions. The rate of penetration was considerably slower than typical rates for small molecules, consistent with the size of the liposomes. Liposome concentrations were higher near the periphery than the interior. However, even the interior concentration was at least as high as the concentration of liposomes in the fluid outside of the biofilm, suggesting some binding or trapping of the liposomes within the biofilm. Penetration of liposomes was observed under flow or static conditions. In a "washout" experiment, where medium is passed through the biofilms previously treated with liposomes, a significant portion of the liposomes remained associated with the biofilms for an extended period of time.

The penetration of liposomes was reflected in the observation of killing of bacteria in colonies in the interior of agar beads. Exposure of these cultures to liposomal amikacin resulted in a large reduction of viable bacteria throughout the beads as monitored by a fluorescent DNA content assay. Similar colony forming unit reductions in animal models (to be shown in other poster presentations) suggest that these principles also operate in vivo.

Liposomes similar to liposomal amikacin (Arikace TM ) readily penetrate into biofilms of Pseudomonas aeruginosa and may even have enhanced binding to biofilms. This binding along with localized release can explain the substantial efficacy observed in animal models.

Coates, A.L. 1 

Adherence to recommended therapy in CF has always been a challenge, in part, due to the time demands of the daily therapy. While twice daily inhaled tobramycin for those infected with Pseudomonas aeruginosa (PA) has become an accepted standard of care, as much as 40 minutes a day may be consumed inhaling 300 mg in 5 mL of tobramycin (TOBI ® ) from the PARI LC PLUS ® breath enhanced jet nebulizer. The purpose of this study was to determine if equivalent levels of pulmonary deposition could be achieved in a much shorter time period using 1.5 mL of a more concentrated (100 mg/mL) tobramycin solution delivered by a perforated vibrating membrane nebulizer (eFlow ® membrane configuration 35L) both, developed by PARI Pharma; Germany.

Methods With a goal to study 7 children and 8 adults, to date, the subjects are 4 children 10 years and older and 2 adult males, all with an FEV1 > 50% predicted, with stable CF. All were receiving inhaled tobramycin for positive sputum cultures of PA. Following pretreatment with albuterol, they inhaled both preparations on two occasions with 99m Tc-DTPA added to the tobramycin in the Nuclear Medicine facility. In vitro preliminary work demonstrated that the radiolabel tracked with both formulations of tobramycin. Deposition was measured by a gamma camera taking both tissue attenuation and mucociliary clearance during nebulization into account (Pediatric Pulmonol Suppl 29: A343; 2006) . In order to have a continuous rate of deposition, the PARI LC PLUS ® was run for a timed 10 minutes and then both the total deposition and time of nebulization "scaled up" from in vitro testing when the nebulizer was run to dryness. This was done by multiplying the deposition by the total output when run to dryness divided by the total output in 10 minutes. (Blood samples were taken for quantification of tobramycin in the serum but not yet analysed). The rate of output per minute was calculated from the 10 minute run and the total time was total output from in vitro testing divided by the rate of output. The eFlow ® pro-vides a continuous output and stops automatically at dryness. Quality assurance was the agreement between total radioactivity pre nebulization (in the nebulizer) and post which included the subject, the nebulizer, the connectors and the expiratory filter.

The PARI LC PLUS ® delivered 38.1±6.1 mg in 15.8±1.6 minutes compared to 42.7±10.2 mg in 4.2±1.2 minutes for the eFlow ® . Only the time of delivery was significantly different with p<0.0001(paired t-test). Tolerability of the treatment was comparable for both inhalation regimes, but the shorter treatment was preferred by all patients.

These results demonstrate the possibility of delivering equivalent levels of tobramycin in much shorter periods of time into the lungs of CF patients when using eFlow ® , a very efficient electronic nebulizer. This time saving may improve adherence to recommended therapy. (Pediatrics 2006; 118:1260) . In order to properly interpret OP cultures from NBS infants, especially those with non-classic CF mutations, we need to know the OP flora of non-CF infants. We obtained OP cultures from 100 healthy infants under 1 yr of age. OP specimens were plated on standard CF culture media. Exclusion criteria included a first degree relative with CF, respiratory illness at the time of culture, or positive newborn screen for CF. Data on cigarette smoke exposure, animal exposure, and exposure to hot tubs/swimming pools was collected. 57 samples have been collected to date. In healthy, non-CF infants, the most common finding is non-specific mixed Gram negative and positive growth. However, 3 infants have grown PA (ages 3 months, 10 months, and 10 months). Many infants have grown multiple organisms. The following bacteria have been found in (n) number of infants: S. aureus (10), E. coli (8), E. cloacae (6) , H. flu (4), Klebsiella (4), Pseudomonas aeruginosa (3), H. parainflu (3), unidentified non-lactose fermenting (4), other (9). Data on 100 infants including correlation to environmental factors will be presented. Conclusion: Non-CF infants commonly have S. aureus and many Gram negative organisms including PA in their oropharynx. These results may have some bearing on interpetting colonization and clearance of PA in infants identified through NBS and in EPIC study participants. #300) ). Identification of SMG by sputum cultivation represents a significant challenge because the organisms are phenotypically diverse, grow poorly on routine culture media, and are very difficult to discriminate from other members of the oropharyngeal flora. We have developed a solid media for the selective isolation of SMG from sputum. The value of the media is highlighted by the identification of SMG as the quantitatively dominant organism in sputum samples of three CF patients admitted to hospital for an acute pulmonary exacerbation. In all three, the SMG species failed to be identified on routine or selective media currently described for the culture of CF-specific sputum pathogens. Antibiotic treatment directed against the SMG correlated with clinical resolution of acute symptoms as well the reduction of SMG on daily serial sputum cultures during hospitalization. This novel selective media makes use of antibiotics (colistin, sulfadiazine and oxolinic acid) inhibitory to the growth of principal CF pathogens and much of the usual oropharyngeal flora. SMG agar utilizes a colorimetric indicator to uniquely identify SMG colonies. The sensitivity and specificity of the selective media has been evaluated by molecular methods using Terminal Restriction Fragment Length Polymorphism analysis. SMG organisms do not respond well to anti-pseudomonal therapy, therefore proper detection and culture-directed antibiotic therapy is paramount. We believe that SMG represent significant respiratory pathogens in CF, and because of the inability to effectively culture and identify SMG they have largely gone unrecognized. Pseudomonas aeruginosa releases substantial amounts of the blue antibiotic pigment pyocyanin. In presence of a reductant (such as NADPH), pyocyanin redox-cycles and generates superoxide and H 2 O 2 . In infected CF airways, pyocyanin concentrations can reach high micromolar levels and contribute to oxidative stress of the airways. The structural basis of the pyocyanin molecule that underlies the redox-cycling with reductants of the airways is not clear. We therefore investigated i) the ability of physiologically or pharmacologically relevant reductants of the airways to support the redox-cycle activity of pyocyanin, and ii) the molecular features of pyocyanin that support redox cycling. Dose-and time-dependent H 2 O 2 production by pyocyanin was measured by Amplex Red oxidation in presence of horseradish peroxidase. Rates of H 2 O 2 production by 100 µM pyocyanin in presence of 150 µM reductant were: NADPH (392 pmole/min) > L-ascorbate (108 pmole/min) > reduced glutathione (41 pmole/min) > α-tocopherol (12 pmole/min). In contrast, lipoic acid, genistein, or resveratrol did not significantly support pyocyanin-mediated H 2 O 2 production. In absence of a reductant, pyocyanin showed no measurable formation of H 2 O 2 . To identify the structural characteristics of the pyocyanin molecule that allow for its redox-cycling activity we synthesized a number of new pyocyanins containing electron-donating or electron-withdrawing substituents. Functional assays were performed in presence of L-ascorbate as reductant. Molecular substituents that donated electrons to the positively charged core of pyocyanin, either by hyperconjugative or resonance effects, reduced the H 2 O 2 output of the corresponding pyocyanin. In contrast, a closely related analog (2-hydroxyphenazine-5,10dioxide) showed significantly increased activity (2.2x compared to pyocyanin) suggesting that the electron-withdrawing effect of the N-oxide functionality led to an increase in the redox-cycle activity. These data indicate that the functional characteristics of pyocyanin as a redox-cycling compound are governed by its positively charged core. In the airways, pyocyanin is predicted to utilize several reductants that are present in the airway surface liquid or intracellularly, thus contributing to CF airway disease. Because pyocyanin utilizes a variety of reductants, it appears prudent to test whether inhaled small-molecular CF therapeutics support pyocyanin function.

Supported by NIH (HL-071829, P01 AT002620), CFRI, Philip Morris USA Inc and Philip Morris International, and CFF (Fischer07G0). Taccetti, G.; Braggion, C.; Ravenni, N.; Zavataro, L.; Neri, A.; Festini, F.; Campana, S. Meyer Hospital, University of Florence, CF Center of Tuscany, Florence, Italy For practical purposes, after early eradication treatment, at least three consecutive negative respiratory cultures over a 6-month period would indicate that the organism has been eradicated (CF Trust Guidelines). This recommendation is based on opinion/clinical experience of respected authorities in the absence of directly applicable studies of good quality.

Aims: Using molecular biological techniques, we evaluated whether this 6-month interval is really trustworthy for distinguishing between regrowth of the same strain, suppressed but not eradicated by treatment, and new Pseudomonas aeruginosa (Pa) colonization.

Patients and Methods: Cystic fibrosis (CF) patients were treated with oral ciprofloxacin and nebulized colistin at detection of Pa. All Pa colonization episodes were recorded in an appropriate database. Molecular study of each bacterial isolate from each colonization episode was performed with the RAPD-PCR.

Results: Between 1998 Between -2006 of patients in follow-up in our Center had repeated Pa colonization. The patients' mean age at first Pa colonization was 78 ± 4.9 months. A total of 78 episodes was observed (mean of 2.75 episodes per patient, median of 2, range 2-7). Molecular typing on 78 strains indicated that 66 (84.6%) were a different genotype from later colonization episodes while 12 (15.4%) isolates had the same genotype as those of the preceding episode. The same genotype as preceding colonization was observed in 7 (46.6%) of 15 isolates from patients in which a successive colonization was verified in less than 6 months, and was verified in 8 (12.6%) of 63 strains in patients having a successive colonization in over 6 months (OR=6.01). Colonization was due to a genotypically diverse strain in 54% of cases where colonization occurred within 6 months of eradication. During the observation period 7 (25%) of 28 patients acquired chronic Pa infection.

Conclusion: Re-colonization by Pa following eradication therapy is mostly (84%) caused by strains with a different genotype, suggesting acquisition from an external source. A short Pa-free period is mainly due to transient suppression of Pa growth, and true eradication followed by acquisition of a new Pa genotype occurred in most cases only after a Pa-free period longer than 6 months. Those patients with a Pa-free interval of less than 6 months had a six-times higher chance that the Pa was not eradicated compared to those with a germ-free interval of over 6 months.

This evidence demonstrates that the definition of successful eradication should be reconsidered, taking into account additional parameters such as molecular analysis.

Cystic fibrosis (CF) patients appear to have an increased risk of urolithiasis. While a number of possible explanations for this have been proposed and investigated, no definitive mechanism has yet been demonstrated. As CF patients frequently get respiratory tract infections they regularly receive ciprofloxacin treatment, often given in doses well in excess of conventional prescription regimes. There are occasional reports of ciprofloxacin crystals urine and stones in the urinary tract. Here we investigate the hypothesis that ciprofloxacin excreted in urine might act as a promoter of crystallisation of calcium or magnesium salts and thereby increase the risk of kidney stone disease.

The effect of ciprofloxacin was tested in artificial urine (AU). In vitro crystallisation was tested using a 96 well plate turbidity method, to identify a metastable limit of oxalate concentration (ML) and a growth and nucleation parameter, the turbidity rate index (TRI). The nucleation pH of urine was examined by tritrating oxalate free AU through a pH range of 5.3 to 9.0 and monitoring the solution/suspension turbidity.

In AU at pH 6.0, ciprofloxacin, at 300, 600, 1000 or 1800mg/L, had no detectable effect on initiation of calcium oxalate crystallisation (ML) or its progress (TRI) (n=6 for each concentration). When AU with 6mM Ca, 3mM Mg and 25mM PO4 was titrated there were two distinct nucleation points; a slow event starting at about pH6.7 and a much faster event at about pH 7.4. Omitting Ca or Mg confirmed that the first event was due to calcium phosphate precipitation and the second to struvite. Including ciprofloxacin at 400, 500 or 600mg/L did not alter these nucleation pH values, but the magnitude of the turbidity rise showed that the ciprofloxacin co-precipitated with the struvite. Ciprofloxacin at 1000mg/L and without Ca or Mg began to precipitate at pH 6.2 and could be held in solution until pH 7.2 when Ca and Mg were included.

Even at high concentrations, ciprofloxacin does not influence calcium oxalate crystallisation. Nor does it promote calcium phosphate or struvite precipitation; on the contrary, while the calcium and magnesium remain in solution, they help to prevent precipitation of the ciprofloxacin itself. Urinary ciprofloxacin does not appear to act as a stone or crystal promoter. Pseudomonas aeruginosa is a significant cause of mortality in cystic fibrosis (CF) sufferers. CF patients were thought to acquire P. aeruginosa from the environment; however genotyping over recent years has revealed clonal strains in sputa from CF patients in the UK, Australia, and Canada that are transmitted person to person or from a common source. One clonal strain, Australian Epidemic Strain-1 (AES-1), (formerly Melbourne epidemic strain, m16 or PI) currently infects up to 40% of patients in five CF clinics on the eastern seaboard of Australia. Most CF clonal strains have been associated with increased virulence not fully explained by greater antibiotic resistance. Both genotypic and phenotypic differences have been postulated as important in enabling transmission of clonal strains.

In order to compare the expression profile of AES-1 to the type strain P. aeruginosa PAO1, the CF research group at the University of Sydney compared the genome expression data of four clonal AES-1 isolates and PAO1, when grown as planktonic and as 72-hr biofilm cultures. In AES-1, a set of 22 significantly differentially expressed genes (all downregulated) were identified, including the quorum sensing genes lasA, lasB and rhlL. In contrast, both upregulated and downregulated genes were differentially expressed in PAO1 biofilm compared to PAO1 planktonic culture. Expression data was validated using quantitative real-time PCR.

To compare biofilm growth at the phenotypic level, the four clonal strains and PAO1 were grown as 72-hr biofilms in a double-blind study, and the size of ten randomly selected biofilms per isolate, stained with Syto® 9 green fluorescent stain was measured. At 72 hr, the biofilms formed by AES-1 isolates were significantly larger (ca. 18-fold) than PAO1 (average size: 33389±14211µm 2 vs 1791±498µm 2 )(p<0.01). The average thickness of three biofilms per isolate, measured by confocal microscopy, showed AES-1 biofilms to be approximately 1.5-fold thicker than those formed by PAO1 (12.6±0.2µm vs 8.5±1.4µm).

The general gene downregulation observed in AES-1 biofilms suggests an adaptation to the CF host, while a larger biofilm would provide for more effective bacterial dispersal. Thus the transmissibility of AES-1 may be linked to enhanced biofilm formation upon colonisation. Background -Objective: whilst influence induced by bacterial colonization in cystic fibrosis (CF) is established, risk induced by fungal colonization is less defined. Prevalence of other species than Aspergillus sp. or Candida sp., and factors associated with fungal presence are also poorly documented. Our preliminary study aimed to determine which fungal species were present in sputum collected from adult CF patients, and which factors were associated with fungal presence.

Methods: in a monocenter, transversal prospective study, 21 CF adult patients were included to determine fungal presence in sputum using semi selective growing media. Clinical parameters (Shwachman Score, respiratory function, nutritional status, gastro-oesophageal reflux, pancreatic insufficiency and diabetes); therapeutics used (including oral or intravenous antibiotics, systemic or inhaled corticosteroids or bronchodilatators, antifungal treatments); microbiological data of bacterial colonization and environmental parameters (potted plants or domestic animals presence) were determined for each patient. Correlation between fungal, clinical, environmental, therapeutic or microbiological data was evaluated by Mann-Whitney non parametric U test.

Results: 16 patients (76%) presented with fungal presence in sputum. 61% presented with yeasts species, 52% with moulds. Aspergillus fumigatus and Candida albicans were the predominant species in moulds and yeasts respectively, but less common mould species such as Exophiala dermatidis or Paecilomyces variotii were also recovered. Factors associated with fungal presence were pancreatic insufficiency (p=0,04); malnutrition (p=0,039), bacterial colonization and inhaled corticosteroids. Candida albicans was correlated with more severe Shwachman Score (p=0,019), bacterial colonization (p=0,017), notably with Pseudomonas aeruginosa (p=0, 03) , and intravenous antibiotics use (p=0,015). Moulds species were significantly associated with inhaled corticosteroids (p=0,049). Antifungal use was associated with frequent resistance to azoles treatments (4 resistant isolates out of 5 patients treated).

Conclusion: fungal presence in CF appears frequent. Some species could have been previously overlooked due to diagnosis difficulties. The effect of corticosteroids on moulds species, already found in other pathologies, appears important in CF. Influence of fungal presence on CF course needs prospective studies, in order to establish if patients could benefit from antifungal treatments or preventive measures. dren on a ventilator without a previous diagnosis of cystic fibrosis is unknown.

The aim of our study was to investigate the prevalence of these microorganisms in routine sputum cultures in young children on a ventilator in a Pediatric Intensive Care Unit (PICU).

Methods: From all ventilated children aged 0-3 years admitted from 1998-2004, sputum culture results obtained from tracheal aspirates within the first week of admission were retrospectively analysed. Three patient subgroups were identified: respiratory failure due to pulmonary disease (group 1), ventilation after elective surgical procedures (group 2) and other ventilated children (group 3). Children with a previous intensive care admission or CF were excluded. The CF database was checked (2007) to identify any children with a new diagnosis of CF that were included in the study.

Results:12.1 % of all ventilated children had a positive sputum culture with one of the "CF-bacteria" S. Aureus, H. Influenzae and P. Aeruginosa These were found mainly in group 1 (see table) . 27/54 of these children were admitted for treatment of Respiratory Syncytial Virus (RSV) bronchiolitis. A sweat test was performed in 7 children, all admitted with pulmonary disease and because of co-existing clinical signs or symptoms. One sweat test was positive and subsequently CF was diagnosed in this child. No other children out of the study group have since then been diagnosed with CF. As the population in the northern part of the Netherlands is very stable, it is unlikely a diagnosis of CF has been missed.

Conclusion: S. Aureus and H. Influenzae are cultured frequently in ventilated children without CF, especially when ventilated because of pulmonary disease. 

The opportunistic pathogen P. aeruginosa causes both acute and chronic human infections. The balance between systemic infection and mortality or chronic persistence and morbidity depends on complex relationships in which the immunological status and genetic potential of the host but also the bacterial biodiversity are determinant factors. P. aeruginosa extensive genetic adaptation and microevolution have been repeatedly observed in chronic infections of CF patients in contrast to what it is documented in acute infections. Whether these P. aeruginosa clonal variants differ in their pathogenic potential is not yet known.

A total of 19 clinical P. aeruginosa isolates from six CF patients which carried unique clonal lineage from the onset of colonization over 4-18 years were selected (Bragonzi et al, 2006; Montanari et al, 2007) . Five P. aeruginosa environmental strains, which represent the source of acquisition for CF patients, and two laboratory strains PAO1 and PA14 were also used as references. Multiple genotypic analysis of sequential P. aeruginosa isolates which included PFGE, ATchip and multilocus SNPs showed intraclonal diversity with genome rearrangements, variations in pathogenic islands and acquisition of prothoadaptive mutations in the muc genes. The largest divergence was observed between the completely sequenced reference strains PAO1 and PA14, the latter represented in our panel by three CF isolates. P. aeruginosa virulence has been assessed by monitoring its capacity to induce bacteremia and to establish chronic infection in a murine model (Bragonzi et al, 2005) . Overall, P. aeruginosa environmental strains increased five times the risk of bacteremia when compared to clinical strains (Test of proportion: p<0.001). The high risk of mortality was also evidenced for P. aeruginosa CF strains isolated at the onset of the infection when compared with those isolated after years of colonization (p=0.002) indicating that environmental strains or newly acquired strains were similar in their virulence. P. aeruginosa clinical strains isolated after years of colonization increased chronic persistence and reduced the risk of bacteremia (p=0.002) when compared to strains isolated at the onset of the infection. The strain PA14 was found to be lethal in contrast to P. aeruginosa clonal strains of clinical origin that established chronic persistent infection in the murine lung. Furthermore, our data showed that adaptive traits commonly associated with the chronic P. aeruginosa infections in CF patients, such as transition to the mucoid phenotype, did not confer a selective advantage to bacterial cells in colonizing the murine lung (p=0.294).

These results suggest that P. aeruginosa pathogenicity is independent of the strain's genotype but rather genetic adaptation to the CF airways plays an important role in the development of persistence and in the resistance to host defences.

Supported by Telethon and Italian CF Research Foundation.

Beringer, P. The combination of antipseudomonal beta-lactam and aminoglycoside antibiotics is frequently prescribed during acute pulmonary exacerbations. Since the introduction of the beta-lactam compounds there has been numerous reports citing altered pharmacokinetics of several compounds (but not all) within the beta-lactam class. Recently it has been suggested that substrate specificity for the renal transporter Pgp may explain the variability in renal drug clearance observed in patients with CF. Pgp is structurally related to CFTR and evidence suggests it serves a complementary role in modulating alternative chloride channel function. In a CFTR knockout model MDR1 expression was reported to be increased nearly four-fold. In an effort to elucidate the potential contribution of Pgp to the renal clearance of drugs in patients with CF, we conducted a controlled clinical trial comparing the pharmacokinetics of fexofenadine (FX) in patients with CF and age matched healthy volunteers (HV). Fexofenadine is not significantly metabolized and is a relatively specific substrate for the membrane transporters Pgp and OATP. Probenecid (Pb) is a selective inhibitor of OATP and was used pharmacologically to block the activity of this transporter in-vivo. Our hypothesis is that if Pgp is upregulated within the renal tubules of patients with CF we would expect to see an enhanced renal clearance of fexofenadine in these patients when compared with control subjects.

Methods: 16 (n=8 CF, 8 HV) subjects underwent this prospective controlled study which FX was received alone or in combination with probenecid (PB). Iothalamate was given each day to measure glomerular filtration rate (GFR). Blood and urine samples were obtained at specified times over 12 hours each study day for determination of fexofenadine and iothalamate concentrations. Fexofenadine and iothalamate concentrations were assayed using liquid chromatography-tandem mass spectrometry and HPLC-UV respectively. Pharmacokinetic analysis was performed using a 1-compartment cumulative urinary excretion model with the ADAPT II software. Differences between groups were determined using a paired t-test or Mann Whitney-U.

Results: 16 patients were included, CF patients were slightly older and had lower body mass index when compared to HV, (mean±SD, 27±5.6 vs. 33.4±7 years p=0.0499, 20.03±1.6 vs. 25.2±2.2 p=0.0019 respectively) but did not differ in GFR (120±21.7 vs. 108±25 mL/min-1.73m 2 p=0.38 respectively). No significant differences were found between CF and HV in CLr of FX alone and FX in combination with PB CLr (93.3±20 vs. 83.3±23.3 mL/min-1.73m 2 p=0.31 and 38.3±10 vs 36.7±23.3 mL/min-1.73m 2 p=0.79 respectively).

Conclusion: The results of this study indicate the enhanced clearance of certain antibiotics previously reported in patients with CF do not appear to be due to upregulation of renal tubular Pgp or increased GFR.

MacEachran, D.P. Department of Microbiology and Immunology, Dartmouth Medical School, Hanover, NH, USA P. aeruginosa is the leading cause of morbidity and mortality in Cystic Fibrosis (CF) patients. Shortly after birth CF patients are colonized by P. aeruginosa, which leads to a lifelong chronic infection. The mechanisms behind P. aeruginosa colonization of the CF lung are still poorly understood, however several secreted toxins have been associated with this phenomenon. CF is characterized by a loss of mucocilliary clearance, a key component of innate immunity in the lung, and an increase in sputum viscosity as a result of mutations in the gene encoding the Cystic Fibrosis transmembrane conductance regulator (CFTR). We have previously shown that the novel P. aeruginosa secreted protein Cif is capable of reducing apical membrane expression of CFTR. We have begun to characterize cif expression and have identified a key regulator of cif transcription. Mutations affecting the divergently transcribed putative TetR family repressor encoded by the PA2931 gene result in a significant increase in Cif expression. Furthermore, we have demonstrated that the PA2931 gene product is capable of binding the promoter region located immediately upstream of the cif-containing operon. This work has been supported by the NIH T32-DK007301-29.

Ryall, B. 1 ; Davies, J.C. 2, 4 ; Wilson, R. 3 ; Shoemark, A. 3 ; Williams, H.D. 1 1. Division of Biology, Imperial College London, London, United Kingdom; 2. Department of Gene Therapy, Imperial College London, London, United Kingdom; 3. Host Defence Unit, Royal Brompton Hospital, London, United Kingdom; 4. Paediatric Respiratory Medicine, Royal Brompton Hospital, London, United Kingdom Pseudomonas aeruginosa is the most important respiratory pathogen in patients with cystic fibrosis (CF) and non-CF bronchiectasis. The factors which enable P. aeruginosa to establish chronic, debilitating respiratory infections are unclear. However P. aeruginosa is one of a limited number of bacteria that is able to synthesise hydrogen cyanide, a potent inhibitor of cellular respiration. This study examined whether hydrogen cyanide is produced by P. aeruginosa in CF and non-CF bronchiectasis airway infection.

Cyanide concentration was measured in fresh sputum from CF and non-CF bronchiectasis patients with and without P. aeruginosa lung infection using a cyanide ion sensing electrode. Cyanide was detected in sputum from 15/25 CF and non-CF bronchiectasis patients with current P. aeruginosa infection, whereas it was not detected in any of the 10 patients without this organism (p<0.01). Maximum lev-els were 130µM (mean ±SE: 72 ± 6.6 µM), which compares with levels of above 40µM which would be considered toxic in blood. Concurrent lung function data were available on all 21 P. aeruginosainfected CF patients; the group with measurable sputum cyanide (n=11) was not different from those without (n=10) on the basis of age, gender or P. aeruginosa phenotype (mucoid/ non-mucoid). However, those with detectable cyanide had significantly poorer lung function than those without (FEV1% predicted: 26.8±3.8% versus 46.0±6.7%, p<0.01; FVC% predicted: 44.4±4.9% versus 60.1±7.7, p<0.05). We have shown that cyanide is detectable at clinically significant levels in sputum from CF and non-CF bronchiectasis patients infected with P. aeruginosa and is associated with significantly impaired lung function.

We have recently reported a dose regimen for once-daily tobramycin dosing in patients with cystic fibrosis which was based on a retrospective study (Lam W et al., J Antimicrob Chemother 2007) . We present here an interim analysis of an ongoing prospective study to evaluate the proposed dosing in paediatric patients with cystic fibrosis. As part of this protocol pharmacokinetic parameters were determined in cystic fibrosis (CF) patients receiving intravenous antibiotic therapy for a pulmonary exacerbation. Serum creatinine, audiometric testing, respiratory cultures and FEV 1 were assessed at baseline and after a 2 week treatment period. Serum tobramycin concentrations were interpreted by clinical pharmacists performing therapeutic drug monitoring (TDM) to optimize dosing. Serum tobramycin levels were drawn after the first dose, repeated with each dose adjustment and weekly thereafter. The proposed sample size of this study is 60 patients; an interim analysis was performed to identify trends for pharmacokinetic and safety outcomes from 18 patients admitted from November 2006 through April 2007. Seven (38.9%) patients achieved the target C max of 25-35mg/L, 10 (55.6%) patients remained under the target range and 1 patient was above the target range with the 1st dose. With 2nd, 3rd, and 4th TDM performed for either routine weekly levels or dose adjustment levels, 6, 16, and 17 patients met the target C max respectively. The percentage of patients who achieved target C max increased to 88.9 % with the 3rd set of TDM levels and 94.44 % with the 4th set of TDM levels (p-value<0.0001). There were no significant changes in k e , t 1/2 and V d /kg for each patient from 1st to 2nd set of TDM levels over time. Compared to the previous retrospective study the mean V d /kg was significantly higher (0.327L/kg versus 0.267L/kg, p=0.0021) and the mean t 1/2 was significantly shorter ( 1.73 hrs versus 2.14 hrs; p<0.0015). None of the patients' serum creatinine increased ≥ 50 % from baseline. Baseline audiometric tests were within normal limits in 12 (70.6 %) patients. P. aeruginosa was isolated from respiratory specimens of 6 patients on initial culture. Of the 5 patients with follow-up cultures, 3 patients grew P. aeruginosa sensitive to tobramycin, demonstrating no decrease in sensitivity. Our preliminary results suggest the pharmacokinetic data differ from the expected profile suggested by the retrospective study. All study patients treated according to the guidelines were clinically well after their course of therapy and did not experience clinically significant toxicity from once-daily tobramycin dosing. The once-daily tobramycin dosing will continue to be evaluated until the proposed sample size is achieved for final analysis.

Anderson, G.G.; O'Toole, G.A. Dartmouth Medical School, Hanover, NH, USA Pseudomonas aeruginosa chronically colonizes the lungs of individuals with cystic fibrosis (CF). Evidence suggests that this infection progresses through distinct phases, wherein initial colonizing strains undergo a switch to mucoidy that corresponds to formation of bacterial microcolonies in the airways and life-long persistence. Currently, the environmental signals and mechanisms regulating conversion from the acute, highly virulent form to a less virulent biofilm state remain obscure. To investigate these issues, we developed a tissue culture model system for growth of P. aeruginosa biofilms on CF-derived human airway epithelial cells in vitro. Biofilms grown on these cells appear similar to those found in CF airways as well as to abiotic biofilms. Eventually, epithelial cells were killed by the P. aeruginosa biofilms. However, we discovered that addition of the antibiotic tobramycin preserved the monolayer integrity for at least 24 hours. Bacteria were not completely eliminated by this treatment, suggesting that tobramycin influenced the virulence of the biofilm bacteria. Microarray analysis of CF cell grown biofilms treated with tobramycin revealed marked alterations in transcriptional profiles compared to untreated controls. Two tobramycin-induced genes were identified from this screen which, when deleted, exerted altered virulence phenotypes on airway epithelial cells in culture. Quantification of cytotoxicity of these mutants revealed either increased or decreased ability to kill the epithelial cells. Complementation with full-length copies of these genes restored wild-type function. These studies suggest a complex relationship between acute and chronic forms of P. aeruginosa, and that antibiotic treatment may influence transitions between these modes. It is this interplay that may determine the virulence status of the microorganism upon initial infection as well as during CF exacerbations.

Neville, M.; Sardaryan, G.; Ejaz, S.; Scotto, A.W.; Gupta, R. Transave, Inc., Monmouth Junction, NJ, USA Purpose: During the course of treatment, patients with cystic fibrosis and chronic Pseudomonas aeruginosa infections in their lungs may be treated with nebulized tobramycin (Tobi®). Reduction in treatment times or dosing frequency may improve compliance and outcome. Nebulized Arikace™ may offer advantages to tobramycin by producing sustained lung levels of drug, reducing dosing frequency and increasing antimicrobial efficacy. Herein are presented biodistribution data and efficacy data in a chronic rat model of P aeruginosa comparing Arikace™, soluble amikacin and tobramycin.

Method: Female rats (Charles River) received (~6 mg / kg) of Arikace™, soluble amikacin and tobramycin by inhalation. Test articles were aerosolized in a PARI LC star nebulizer at 30 PSI using Devilbiss compressors (Sunrise Medical) for 80 min. 3-4 rats were euthanized at time 0 and 6 hr post inhalation by CO2 asphyxiation. Blood was collected by a cardiac puncture. Serum was collected from coagulated blood by centrifugation and stored at 4oC until analysis for drug content. In addition lungs, kidneys, intestinal contents, urine and brains were harvested. Tissues and biological samples were homogenized in saline and antibiotic concentrations determined by TDx analyzer (Abbott).

Chronic lung infections were produced in rats by intratracheal instillation of agar beads containing P. aeruginosa. Three days post instillation, rats (n=12/group) were treated with aerosolized saline or once daily with Arikace™, or twice daily with tobramycin (total daily dose = 6 mg /kg) for 14 days. Rats were euthanized 3 days post final dose and the Pseudomonas CFU counts per lung were determined.

Results: The biodistribution and pharmacokinetics of aerosolized Arikace™ was significantly different from that of amikacin and tobramycin. The primary difference was the amount of antibiotic retained in the lungs. Rats that inhaled soluble amikacin cleared approximately 84% of the amikacin from the lungs after 6 hr with concomitant 86% increase in drug levels in the urine. Rats that inhaled Arikace™ cleared only 41% of the deposited amikacin from the lungs after 6hrs. This initial clearance of drug most likely reflected the amikacin released during nebulization. Extended clearance of Arikace™ from the lungs was measured by recovery of amikacin in the urine at later time points. In a separate longer-term study comparing Arikace™ and tobramycin, AUC (0-168 hr)for Arikace™ in the lung was 56233 ug*hr/g while only 1150 ug*hr/ g for tobramycin which demonstrates a 4-fold increase in retention of antibiotic in the lung with Arikace™. Eradication of P. aeruginosa from the lungs of rats with Arikace™ was equal to tobramycin (given equivalent daily doses). However, Arikace™ reduced the number of CFU in the lungs with a single daily exposure to the same degree as inhalation of tobramycin twice a day. Conclusions: Inhalation of Arikace™ achieved higher concentrations and increased retention time of the antibiotic in the lungs of rats when compared to inhaled amikacin or tobramycin. It is expected that these properties of Arikace™ will reduce dosing frequency and improve antimicrobial efficacy in patients.

Nielsen, X.C. 1 ; Johansen, U.R. 1 ; Nørregaard, L. 1 ; Vandamme, P. 2 ; Høiby, N. 1 1. Clinical Microbiology, Rigshospitalet, Copenhagen, Denmark; 2. Laboratory for Microbiology, University of Ghent, Ghent, Belgium Background: The Burhkolderia cepacia complex (Bcc) is a diverse group of bacteria with at least 9 genomovars (GV) or species. Accurate species identification is necessary for better understanding of the epidemiology and pathogenesis of Bcc. Species identification based on phenotypic characters is difficult when it is based on few if any differences. Crossed immunoelectrophoresis (CIE) has been performed routinely to identify all Burkholderia and other difficult isolates in our department (Høiby et al.: "Taxonomic application of crossed immunoelectrophoresis." Int. J. Syst. Bact. 37 (1987): 229-240.) . In this study, we compared the results obtained by CIE with those obtained by the standard molecular biological method using RecA gene PCR and RFLP to identify Bcc at the species level (golden standard).

Methods: Between 1996 and 2005, Burkholdreia isolates were recovered from 34 patients in the two Danish Cystic Fibrosis (CF) centres. 1. CIE method: Immunological cross-reactivity between antigens from the reference strains and the strains isolated from our CF patients were compared by employing CIE and rabbit standard-antibodies (purified IgG) raised against water-soluble antigens from reference strains of the Bcc. Species identification was made based on matching coefficient (MC). 2. RecA gene-based identification: Bcc specific primers were used to amplify recA gene. Species-specific PCR is then performed using B. multivorans specific primers to identify the B. multivorans GV II. RFLP patterns were generated for those non-multivorans Bcc by digesting the Bcc-specific PCR products with HaeIII. The patterns were then compared with the patterns generated from reference strains in the database. Pulsed field gel electrophoresis (PFGE) genotyping was performed for all 34 strains to detect patient-topatient transmission.

Results: 32/34 isolates were Bcc-specific PCR positive. Species-specific PCR identified 29 isolates as B. multivorans (GV II). The other three isolates were identified as B. cenocepacia (GV III) based on RFLP patterns. Two isolates with negative Bcc-specific PCR reactions were identified as B. gladioli by CIE (confirmed by Reference laboratory in Belgium). Results from CIE showed that strains from Bcc were immunologically closely related with only a few non-cross-reactive antigens. Compared with the standard recA gene-based method, CIE method resulted in correct identifications in 27 isolates (79.4 %), misidentifications in 4 isolates (11.8 %) and uncertain identifications in 3 isolates (8.8%). Genotyping by PFGE showed unique patterns for all isolates except for those from two sisters. This suggested that patient-to-patient transmission of Bcc among the Danish CF patients has remained very rare.

Conclusions: RecA gene-based PCR and RFLP is a quick and reliable method for identification of Bcc at the species level. CIE proved useful in initial classification of Bcc strains, but additional information is obtained by the RecA gene-based species identification and PFGE method. BACKGROUND: In childhood, the cystic fibrosis airway is characterized by persistent inflammation with high quantities of neutrophils. In this setting, recurrent exposure to low numbers of Pseudomonas aeruginosa occurs routinely. Chronic infection by P. aeruginosa is associated with a significant increase in morbidity and mortality in CF, and is believed to occur by formation of a biofilm in the CF airway. Previously, we have shown a significant neutrophil dependent enhancement of P. aeruginosa biofilm density that is greatest for low initial concentrations of P. aeruginosa. Biofilm enhancement is facilitated via neutrophil-derived DNA and F-actin, which form a framework that P. aeruginosa can exploit for growth. DNA and Factin polymerize via positively charged molecules such as histones. P. aeruginosa then associates with these polymers, which results in increased early biofilm density, displaying an increase of over 400 fold at 24 hours compared to P. aeruginosa in the absence of neutrophils.

HYPOTHESIS: Compounds that disperse DNA and/or F-actin can disrupt neutrophil-enhanced biofilms.

METHODS: Poly(Aspartic acid), a negatively charged amino acid chain with the capacity to disrupt F-actin, was examined singly and in combination with DNase and/or antibiotics (ciprofloxacin and tobramycin). The Nunc-TSP system was used to allow for high throughput assessment of the density of biofilm formation in the presence of neutrophils and the effect of various agents in disrupting the biofilm. The biofilm assay is independent of cellular settling, as the reactor is vertically suspended in the culture. P. aeruginosa strains tested were PAO1, and two isogenic CF isolates recovered from an initial infection (Early), and following established infection ~4 years later (Late). Biofilms were formed by incubation of P. aeruginosa with human neutrophils for 24 to 48 hours in RPMI with 2% heat inactivated platelet poor plasma.

RESULTS: Antibiotics were incapable of disrupting biofilms. Poly(Asp) significantly disrupted neutrophil-induced biofilms. In the presence of neutrophil products this action is sensitive to proteolytic degradation, but can be protected by the presence of protease inhibitors. DNase also disrupts a biofilm formed in the presence of human neutrophils. The combination of poly(Asp) + DNase resulted in an increase in biofilm disruption for a 24 hour biofilm. Biofilms allowed to form for 48 hours were more resistant to disruption by DNase and poly(Asp) as single agents, but the combination of both demonstrated a synergistic effect. Similar disruption of biofilms were observed when PAO1 was compared to Early and Late CF strains.

CONCLUSIONS: Pseudomonas aeruginosa biofilms formed in the presence of neutrophils in vitro can be disrupted by agents that disperse the F-actin and DNA framework. Support: CFF, Max and Yetta Karasik Foundation Cystic fibrosis (CF) patients may suffer increased morbidity and mortality through colonisation, allergy and invasive infection from fungi. The black yeast, Exophiala dermatitidis (synonym Wangiella dermatitidis) has been found with increasing frequency in sputum specimens of CF patients, with reported isolation rates ranging from between 1.1 and 15.7%. At present, no species-specific diagnostic PCR exists to aid with the clinical laboratory detection and identification of this organism. A novel species-specific PCR-based assay was developed for the detection of E. dermatitidis, based on employment of rDNA operons and interspacer (ITS) regions between these rDNA operons. Two novel primers, (designated ExdF & ExdR) were designed in silico with the aid of computer-aided alignment software and with the alignment of multiple species of Exophiala, as well as with other commonly described yeasts and filamentous fungi within CF sputum, including Candida, Aspergillus and Scedosporium. PCR employing ExdF (forward primer [16-,mer], 5'-CCG CCT ATT CAG GTC C-3') and ExdR (reverse primer [16-mer], 5'-TCT CTC CCA CTC CCG C-3'), was employed and optimized on extracted genomic DNA from a well characterized culture of E. dermatitidis, as well as with high quality genomic DNA template from a further 16 unrelated fungi, including Candida albicans, C. dubliniensis, C. parapsilosis, C. glabrata, Scedosporium apiospermum, Penicillium sp., Aspergillus fumigatus, Aspergillus versicolor, Pichia guilliermondii, Rhodotorula sp., Trichosporon sp., Aureobasidium pullulans, Fusarium sp., Mucor hiemalis, Bionectria ochroleuca, Gibberella pulicaris. Results demonstrated that only DNA from E. dermatitidis gave an amplification product of the expected size, whilst none of the other fungi were amplifiable. Subsequent direct employment of this primer pair detected this yeast in the sputum of 2/50 (4%) adult CF patients, employing a nested PCR approach with panfungal outer flanking primers of the ITS1-ITS2 region. These two patients were the only patients who were previously shown to have a cultural history of E. dermititidis from their sputum. E. dermatitidis is a slow-growing fungus, which usually takes up to two weeks to culture in the microbiology laboratory and therefore is slow to detect conventionally, with the risk of bacterial overgrowth from common co-habiting pan-and multiresistant bacterial pathogens from sputum, namely Pseudomonas aeruginosa and Burkholderia cepacia complex organisms, hence this species-specific PCR assay may help detect this organism from CF sputum more specifically and rapidly. Overall, employment of this novel assay may help in the understanding of the occurrence, aetiology and epidemiology of Exophiala dermatitidis, as an emerging fungal agent in patients with CF. Cystic fibrosis patients are particularly susceptible to infection by strains of Pseudomonas aeruginosa and Burkholderia cepacia complex. Since 2001, the Pseudomonas Genome Database has been a resource for peer-reviewed, continually-updated annotations for the Pseudomonas aeruginosa PAO1 reference strain's genome and, more recently, comparative analyses to several closely-related Pseudomonas species. In order to facilitate better cross-strain and cross-species genome comparisons, we have developed or are incorporating methods to improve the identification of orthologs (genes diverged due to speciation) and identify genes undergoing unusual selection. Our analysis of several completely sequenced Pseudomonas genomes has revealed that approximately 5% of ortholog predictions by the classic "reciprocal best BLAST hit method" are likely incorrect. We have also performed an analysis of unusually large intergenic regions in P. aeruginosa PAO1 that appear to be essential (according to saturation transposon mutagenesis) or involved in virulence (according to signature tagged mutagenesis). We identified at least 20 new putative protein-coding regions and non-coding RNAs that were not previously annotated and may play critical roles in microbial pathogenesis or viability. We are also focusing our attention on facilitating better cross-strain comparisons between recently sequenced Burkholderia cepacia complex genomes through development of a new online Burkholderia Genome Database. In addition to improved ortholog prediction, this new database will provide access to a very flexible Boolean search feature that allows researchers to search and compare annotations within or between the genomes of Burkholderia strains, returning annotations from multiple genomes suitable for simultaneous viewing and downstream analyses. We have also included new, very accurate protein subcellular localization predictions for the deduced proteome from each of these genomes -predictions that can aid the identification of new cell surface or secreted therapeutic targets or vaccine candidates. Further comparative analyses with other newly-sequenced, related strains should provide insight into strain-specific features that may be exploited to better understand virulence and antimicrobial resistance exhibited by CF-relevant pathogens. Funding provided by Cystic Fibrosis Foundation, Therapeutics (USA).

Popova, A.P.; Verma, R.; Zanni, R.L.; Sembrano, E. Pediatrics, Monmouth Medical Center, Long Brnach, NJ, USA Background: Cystic fibrosis (CF) is a chronic, progressive, genetic disease leading to poor lung function, chronic bronchial infection with certain bacteria and recurrent pulmonary exacerbations. Despite evidence of inflammation within the airway, it is unclear which circulating inflammatory biomarker best reflects the airway inflammation and lung function in CF.

Objective: To determine the utility of C-reactive protein (CRP) value in evaluating patients with Cystic Fibrosis pulmonary exacerbation.

Design: A retrospective chart review was carried out on 87 patients followed at the Cystic Fibrosis Center at the Children's Hospital. Information about serum CRP values, Pulmonary function test results(PFT), presence of pulmonary exacerbation as documented in the chart, sputum culture results, current medications, as well as demographic data, genetic compositions and weight percentiles between January 2005 and March 2007 was collected from the charts and was analyzed.

Results: Of 87 patients whose charts were reviewed, 5 were excluded because they underwent lung transplant, 3 were excluded because they had other mild infection, not diagnosed as pulmonary exacerbation. For 31 patients CRP was not measured and they were also excluded from the analysis. Total of 48 patients were included in the study. For 9 patients information about CRP was obtained on more than one occasion. The patients were divided into groups based on the presence of pulmonary exacerbation and elevation of CRP values -in the group with pulmonary exacerbation, 10 patients had elevated CRP and 23 had normal CRP and in the nonexacerbation group 1 had elevated CRP and 26 had normal. Fisher's Exact Test (two -tail) was applied (p=0.043) and confirmed statistically significant elevation of CRP in patients with pulmonary exacerbation of Cystic Fibrosis. The means and standard deviations of the CRP values for the patients without pulmonary exacerbation and the patients with pulmonary exacerbation were calculated and analyzed using Independent Group t-test (two tail). Mean CRP value for the exacerbation group was 16.34 with standard deviation of 33.72 and mean CRP value for the non-exacerbation group of 1.83 with a standard deviation of 2.43. Statistically significant difference between the two means was noted (p=0.019).

Conclusion: Serum CRP values in patients with CF pulmonary exacerbation were significantly higher than in patients without pulmonary exacerbation. Serum CRP may be a useful marker of pulmonary exacerbation in children. In the absence of pulmonary exacerbation, CF is not associated with elevated CRP values in children. The findings of this study could aid in identifying patients with pulmonary exacerbation earlier and initiate appropriate therapy sooner. Further studies are needed to determine whether CRP values are associated with the severity of the underlying disease, to what extent the PFT are affected, the age of the patient, the type of bacterial pathogen, and genetic factors. Background Pulmonary infections with nontuberculous mycobacteria (NTM) are commonly associated with structural lung diseases or airways clearance disorders such as cystic fibrosis and primary ciliary dyskinesia. We follow approximately 100 individuals with NTM pulmonary disease who share common clinical features but have no identifiable systemic immune defect or clearly defined predisposing conditions. The pulmonary disease and clinical presentation suggest these individuals may have disorders of ion transport or ciliary function. While some CFTR mutations and ciliary abnormalities are difficult to detect through standard genetic and ultrastructural analyses, in vivo physiologic tests of ion transport and measures of ciliary function may allow us to identify ion transport abnormalities in subjects with no identified CFTR abnormalities, determine the functional significance of novel CFTR mutations, and distinguish disorders of ciliary function from other airways clearance disorders.

Methods We evaluated the medical and family histories, chest and sinus CT's, sputum cultures, CFTR genotypes, and sweat chloride levels of 15 individuals with nontuberculous mycobacterial pulmonary disease and 3 individuals with chronic airways disease and bronchiectasis. We systematically tested the pulmonary function (PFT), nasal nitric oxide production (nNO), and nasal potential difference (NPD) of each subject. Nasal mucosal scrape biopsies were performed and sent to UNC-Chapel Hill for assessment of ciliary ultrastructure as part of the Genetic Diseases of Mucociliary Clearance Consortium. Pulmonary function, nNO, and NPD were measured on 8 healthy controls.

Results All subjects had sino-pulmonary symptoms consistent with those associated with CF and PCD. 5 subjects had low nNO (27.2 ± 20.7 nl/min) consistent with levels reported in subjects with PCD (<100 nl/min, Noone 2004). One subject with variant CF (∆F508/R347H) displayed low basal PD (-46.1 mV), elevated Na+ absorption (34.2 mV), and no CFTRmediated Cl-conductance (-0.6 mV). 5 subjects with CFTR mutations, 1 subject with isolated elevated sweat chloride (72 mmol/L), and 1 subject with isolated borderline sweat chloride (44 mmol/L) displayed normal basal potential difference (-12.4 ± 5.1 mV) and Na+ absorption (8.0 ± 4.4 mV) but decreased CFTR-mediated Cl-conductance (-9.3 ± 5.5 mV) relative to controls (-15.0 ± 9.1 mV).

Conclusions NTM patients heterozygous for CFTR-mutations or with elevated sweat chloride (and no identified mutations with coding region sequencing of CFTR) may have detectable subtle differences in CFTR Clconductance, while measurement of nNO may identify disease associated with ciliary abnormalities. These findings suggest subtle abnormalities in airways clearance may predispose to airway infection with environmental organisms such as NTM. Support: Intramural funds, National Institute of Allergy and Infectious Diseases

Muhlebach, M. 1 ; Goodrich, J. 2 ; Sutton-Shields, T.N. 2 ; Wedd, J.P. 3 ; Henegar, C. 1 ; Miller, M.B. 4 ; Gilligan, P.H. 4 1. Dept. of Pediatrics, UNC Chapel Hill, Chapel Hill, NC, USA; 2. University of North Carolina Hospitals, Chapel Hill, NC, USA; 3. School of Public Health, University of North Carolina Hospitals, Chapel Hill, NC, USA; 4. Dept. Pathology and Laboratory Medicine, University of North Carolina Hospitals, Chapel Hill, NC, USA Introduction: Proportion of methicillin resistant (MRSA) vs. susceptible SA infections increased from 36% in 1992 to 64% in 2003 in US hospitals. Similarly, the prevalence of MRSA in the community has risen in recent years. These community acquired (CA) MRSA isolates are genotypically different, have different antibiotic susceptibility, and often carry the pvl virulence gene. This virulence gene is associated with more invasive, mostly skin infections in healthy individuals. Reports on impact of MRSA in CF show conflicting data. 1 , 2 .

The aim of this study was to determine the prevalence of CA-MRSA strains at our Pediatric and Adult CF center and to assess the impact of these infections on clinical outcome.

Methods: All MRSA isolates recovered from routine clinical cultures, were prospectively collected and analyzed for 18 months (10/05 -4/07). Using molecular assays to determine the PVL status and to identify the SCCmec (staphylococcal cassette chromosome) type, the organisms were classified as CA or HA. Classification as CA MRSA was based on presence of pvl and SCCmec IV type. Clinical outcomes included lung function measurements in patients reliably performing spirometry and nutritional parameters (BMI%). Patients who had undergone lung transplant were excluded as their lung functions vary depending on transplant outcome.

Results: Of the 104 patients identified in this 19 month period, 83% had HA and 17% CA MRSA. Neither current age (mean 18.1 yrs. in HA vs.14.6 years in CA) nor age at acquisition of MRSA (14.6 vs. 13 years) was significantly different. Sixty-eight % of all MRSA infected patients had chronic P. aeruginosa infection, (70% in the HA and 51% in CA, relative risk 2.3 of having HA in presence of chronic P. aeruginosa infection). Crosssectional comparison of patients infected with CA vs. HA acquired strains showed no difference in nutritional status (BMI 37 vs. 38%). Lung function (FEV1 and FEF25-75) was not different in CA vs. HA patients, neither when including all MRSA infected patients or those who had no concomitant infection with other organisms (n=22). We will present further comparisons to age and gender matched children with OSSA infection in the final report.

Conclusion: Comparison to OSSA in matched patients will assess clinical outcomes of infections with MRSA. Initial data collected at our center did not show a significant difference in outcomes between patients infected with CA vs. HA MRSA. A multicenter study would most efficiently assess the impact on clinical outcomes and determine regional differences of MRSA patterns (CA vs. HA) in different CF centers.

Ref : The stringent response is a global regulatory mechanism used by bacteria to adapt to nutrient limitation and other environmental stresses, and is mediated by the signaling molecules (p)ppGpp (phosphorylated guanosine nucleotides). Increases in (p)ppGpp repress metabolic processes and growth, and regulate genes involved in long-term stress and starvation survival in several species ((i)E. coli, M. tuberculosis(i)), bacterial multicellular behavior ((i)Myxococcus xanthus(i)), and pathogenesis ((i)P. aeruginosa(i)).

While many studies have investigated bacterial functions (such as attachment and motility) involved in biofilm development, less is known about the physiological adaptation occurring in biofilm growth. When cells grow to high densities entrapped in a polymer matrix, they must adapt to heterogeneous micro-environments where gradients in nutrients, oxygen and metabolic waste exist. Nutrient may thus become depleted as they are consumed by overlying cells, and some biofilm subpopulations are likely starved.

We hypothesized that the stringent response mediates adaptation to nutrient starvation in (i)P. aeruginosa(i) and is important for biofilm formation. We tested a (i)relA(i) deletion mutant unable to synthesize (p)ppGpp in response to amino acid starvation, and its isogenic wildtype parent PAO1.

We measured the starvation survival of planktonic bacteria in phosphate buffered saline and observed that the (i)relA(i) has a 10 to 100fold greater loss in viability compared to wild type after 48h. We also tested biofilm formation in a static biofilm assay using polystyrene peg lids and estimated biomass accumulation by crystal violet staining. In this system, the (i)relA(i) accumulates 5 times less biomass than wild type. In a flow-cell reactor system, the wildtype strain forms complex mushroom-like biofilm structures, but the (i)relA(i) mutant forms flat, thin biofilms. We determined cell attachment to polystyrene by crystal violet staining, and to glass surfaces by direct visualization under microscopy. The attachment to both surfaces is similar in the (i)relA(i) and wildtype strains, and the strains have identical planktonic growth rates.

Our findings suggest that the stringent response is involved in starvation survival, a physiological condition relevant to growth in biofilm. Furthermore, (i)relA(i) inactivation impairs biofilm formation, suggesting that the stringent response may be an important stress adaptation in biofilm. Belfast, United Kingdom; 2. Northern Ireland Public Health Laboratory, Belfast City Hospital, Belfast, United Kingdom; 3. Northern Ireland Regional Adult CF Unit, Belfast City Hospital, Belfast, United Kingdom; 4. Department of Respiratory Medicine, Queen's University, Belfast, United Kingdom Pseudomonas aeruginosa (PA) is a clinically significant bacterial pathogen responsible for increased morbidity and mortality in patients with cystic fibrosis. Small non-coding micro(mi)RNA species (generally 80~100nts or less) in prokaryotes are involved in numerous cellular processes as is the case with eukaryotes. As in eukaryotes, these miRNAs act by base pairing with target mRNAs imposing translational and stability changes of mRNAs and thus culminating in the control and regulation of target mRNAs, crucial for bacterial stress responses and virulence to changing environments surrounding the infected zone or host cell (e.g. human airway epithelial cells). We recruited a commercial kit (Biodynamics DynaExpress miRNA Cloning Kit AV, Tokyo, Japan) and isolated miRNAs from a variety of Pseudomonas pathogens from plant, soil / agricultural environment (Pseudomonas syringae -PS), Type strain (Pseudomonas aeruginosa -PA) and a collection of clinical isolates of Pseudomonas aeruginosa (cfPA) from adult CF patients, who were chronically infected with P. aeruginosa. The results on the miRNA clones obtained exhibited a wide variation in the occurrence of miRNAs in this bacterium. The putative miRNAs obtained, cfPA in particular, yielded small non-coding RNAs of sizes from as little as 16nts to well over 45nts. An initial search using the bioinformatic tool sRNAPredict (www.waldorlab/tufts.edu) has not as yet, fully revealed the genomic annotations for these novel sRNA sequences in the inter-generic regions of the pathogenic pseudomonad group of bacteria. We present our data on these unknown cellular miRNAs cloned from cfPA isolates from CF patients and discuss their significance as novel regulators to bacterial stress responses and in the context of the bacterium's involvement as the predominant pulmonary pathogen, associated with cystic fibrosis.

Airway function is diminished in infants with CF diagnosed clinically but whether this is true for those identified by newborn screening remains unclear. We investigated whether lung function is diminished in infants with CF diagnosed by newborn screening and if this occurs in association with pulmonary infection. Methods:Lung function was measured in asymptomatic infants with CF following sedation with chloral hydrate (60-100mg/kg) using the raised volume rapid thoracic compression technique at an inflation pressure of 30cmH 2 0. We measured forced vital capacity (FVC), forced expired volume after 0.5 seconds (FEV 0.5 ) and forced expired flow at 75% expired FVC (FEF 75 ) which were then expressed as z scores. 1 Broncho-alveolar lavage (BAL) was performed under general anesthesia within forty-eight hours following infant lung function testing. Three separate aliquots of 1mL/kg sterile saline were instilled into the right middle lobe and a further aliquot in the left lingula and the BAL fluid analyzed for bacterial, fungal and viral pathogens using routine culture techniques. Results: BAL was successful in all 24 infants studied aged 0.2 to 2.1 yrs (median 1.0 yrs) and lung function in 21 infants. Median (range) z scores for FVC, FEV 0.5 and FEF 75 were -1.3 (-2.7 to 1.6), -2.0 (-5.0 to 0.6) and -1.7 (-3.5 to 0.47) respectively. These decrements in lung function were all significant (p<0.01). BAL cultures were positive in 12 (50%) infants of whom 4 (17%) had evidence of infection with S. aureus and 1 (4%) with P. aeruginosa. In 12 infants there was no growth in BAL or growth only of mixed oral flora. Viruses were not identified in any infants. There was no association between any of the lung function parameters and pulmonary infection detected by BAL. Lung function was successful in a subgroup of 7 of 9 infants who were less than 0.5 yrs (median age 0.3 yrs) when tested. Diminished airway function was identified in this younger subgroup even though only one infant had evidence of pulmonary infection with respiratory organisms in BAL. In this younger subgroup median FEV 0.5 z score was -1.6 which was significantly below that predicted (95% CI: -2.3,-0.2 and p=0.03). Conclusions: Airway function is diminished in infants with CF diagnosed by newborn screening and occurs irrespective of infection identified by BAL. Lung function is abnormal in very young infants before pulmonary symptoms develop and when no evidence of pulmonary infection can be detected in BAL. 1 Inhaled high-dose tobramycin appears to transiently clear P. aeruginosa (PA) from airway secretions in young children with CF, though inflammation may not be markedly reduced (Am J Resp Crit Care Med 2003; 167:841) . Because inhaled antibiotics may not reach some areas of infection, we hypothesized that intravenous (i.v.) antibiotics may be more effective than inhaled high-dose tobramycin for reducing lower airways inflammation in CF children with early PA infection. Study Design: We initiated a single-center, randomized, prospective study comparing the effects of two antibiotic treatment regimens on bronchoalveolar lavage fluid (BALF) inflammatory markers. Clinically stable CF children with a recent isolate of PA from surveillance cultures are randomized to receive 4 weeks of TOBI® (Group 1) or 2 weeks of i.v. ceftazidime and tobramycin (Group 2). BALF is obtained just before treatment, and repeated 4-6 weeks after completion of treatment. The primary study outcome is change in BALF % neutrophils (PMN). Multiple secondary outcomes are assessed including quantitative bacterial cultures and cytokines in BALF. This report summarizes interim analysis at the halfway point of the study.

Results: To date, 18 subjects have enrolled. Five subjects (3 in Group 1, 2 in Group 2) dropped out after the initial bronchoscopy due to a decision by the primary CF physician to use other treatments based on BALF culture results (2) , or because of the development of a new respiratory illness before the second bronchoscopy (3). A total of 5 adverse events (1 significant, unrelated to protocol) were reported in 3 subjects, none resulting in a change of protocol request by the IRB or Data Safety Monitor. Of the 13 subjects completing the protocol to date, 5 were in Group 1, and 8 were in Group 2. The groups were well matched in terms of age, PA in initial BALF, and initial BALF %PMN. A majority of subjects in each group were children with their first isolation of PA on surveillance cultures. There was a tendency for Group 2, but not Group 1, to have mild reductions in BALF %PMN (-9 ± 6%) after treatment. Interim efficacy analysis did not suggest a significant difference between the treatment groups at this stage. Similar changes were seen in Group 2 for PMN/ml (-589 ± 351 x 10 3 /ml) and several cytokines including GM-CSF (-54 ± 27 pg/ml) and MCP-1 (-368 ± 231 pg/ml). Two Group 2 subjects were unable to have stable i.v. access established and were treated with oral ciprofloxacin and inhaled TOBI® for 2 weeks as alternative systemic antibiotics, per protocol. Dropping these subjects from Group 2 data did not change data trends.

Conclusions: The protocol appears to have adequate safety. Parents of children with first-time PA infection may be more likely to consent to treatment randomization, than those with repeated PA infection. Interim data analysis suggests that continuation of the study toward the original sample estimate is appropriate.

Acknowledgements: Supported by CFF(NOAH04A0) and NIH (GCRC R00046). We thank Tracy Callahan RN, Lupe Haynes RRT, Justin Hubbard, Paula Murphy, Cassidy Henegar, Nancy Lee RN, Benjamin Butler RN, and Sheree Berckmans RN for logistical and technical support. Methods: Our local CF database was searched for patients born 1985-2000. Patients were included in the analysis if they were followed in our CF center before the age of 5. Patients were separated into two birth cohorts, born 1985-1992 (cohort 1) and 1993-2000 (cohort 2). Yearly peak FEV1% predicted from ages 6-12 were obtained using Knudson reference equations. Yearly peak BMI% and weight-for-age% at each age were also obtained. Type 3 tests of fixed effects and repeated measures analyses were used to assess mean peak BMI%, mean peak weight-for-age%, mean peak FEV1% predicted at first PFT, mean peak FEV1% predicted from ages 6-12, and mean rate of change (slope analysis) in peak FEV1% predicted from ages 6-12. Chi-square test for association was used to analyze differences in demographic parameters. Wilcoxon rank-sum test was used to assess differences in age at diagnosis. Wilcoxon non-parametric test was used to analyze differences in mean clinic visits per year. Chi-square testing with a two sided alpha of p <0.05 was used to determine significance.

Results: 144 CF patients were included in the analysis. There were no significant differences between birth cohorts 1 and 2 for age of diagnosis (p=0.94), sex (p=0.3792), race (p=0.35), genotype (p=0.16), pancreatic insufficiency (p=0.61), or the mean number of clinic visits per year from ages of 0-5 (p=0.33). There was a significant difference in the mean number of clinic visits per year from ages of 6-12 (p=.0009; see Table 1) .

Birth cohort had a significant effect on the rate of decline in lung function from ages 6-12. Birth cohort and age had significant effects on the mean weight-for-age%, mean FEV1% predicted at first PFT, and mean FEV1% predicted from ages 6-12. Sex had a trend for significant effects on lung function parameters.

Adjusted for age and sex, there were significant differences between birth cohorts 1 and 2 in mean BMI% and mean weight-for-age% at ages 6 and 12, mean FEV1% predicted at first PFT, mean FEV1% predicted from ages 6-12, and in the mean rate of decline in FEV1% predicted (see Table 1 ).

Conclusions: Improvements in lung function and nutritional outcomes in our CF center are associated with an increase in the mean number of clinical visits per year for CF patients ages 6-12. The frequency of clinical follow-up is a marker of CF care that may have positive impacts on CF outcomes. BACKGROUND: Current management of CF airway disease includes the use of key pharmacological therapies designed to combat the chronic cycle of obstruction, infection, and inflammation. Randomized controlled trials have demonstrated that maintenance therapy with azithromycin (AZM), dornase alfa (DNAse), and tobramycin for inhalation (TOB) each improves pulmonary function and reduces the frequency of exacerbations in patients with P. aeruginosa. The purpose of this study was to determine if CF patients in an uncontrolled environment (clinical setting), experience a reduced rate of pulmonary function decline when receiving AZM + DNAse + TOB compared to those not on all three therapies. METHODS: Adult CF patients were selected based on positive sputum cultures for P. aeruginosa, pulmonary function test results (FEV 1 % predicted) for the past year, and receiving one of the key pharmacological therapies. In addition data was collected on CFRD, depression, medication adherence, airway clearance, and nutritional status. Patients were stratified according to baseline pulmonary function: mild (FEV 1 ), moderate (FEV 1 of 40-70%), and severe (FEV 1 <40%). Patients were then sub-grouped as to whether they were receiving all 3 or ≤ 2 of the key therapies. The annual rate of decline in % predicted FEV 1 was determined using linear regression on non-exacerbation data and was categorized as stable (<2.3% decline), intermediate (2.3-4.1%) , or rapidly declining (> 4.1%). Time to intermediate or rapid decline in pulmonary function was determined using Cox proportional hazard modeling. RESULTS: 107 patients (54 males/53 females, median age 28 years) were included in this study; 28 with mild, 42 with moderate, and 35 with severe pulmonary disease. Overall, the mild and moderate groups experienced median declines of 2.12% and 1.50%, while the severe group experienced a 2.41% improvement in FEV 1 % predicted respectively. The use of regimens containing AZM therapy significantly delayed time to intermediate or rapid decline in lung function in the mild group (median 210 days vs. 49.5 days, HR=0.19, p=0.013). Similarly, regimens containing DNAse + AZM + TOB significantly delayed time to intermediate or rapid decline in lung function in the mild group (median 231 days vs. 105 days, HR=0.27, p=0.048). No significant differences were seen in the moderate or severe group for patients receiving all 3 or individual therapies. CON-CLUSION: Administration of pharmacological regimens containing DNAse + AZM + TOB, or AZM + TOB or AZM + DNAse significantly delay progression of pulmonary disease in CF patients with mild lung disease.

Sawhney, V. 1 ; Arthur, B. 2 ; Seltzer, R. 3 ; Kraynack, N.C. 4 1. Internal Medicine, Akron General Medical Center, Akron, OH, USA; 2. Pediatrics, Akron Children's Hospital, Akron, OH, USA; 3. Biostatistics, NEOUCOM, Rootstown, OH, USA; 4. Pulmonology, Akron Children's Hospital, Akron, OH, USA

Early and efficient diagnosis of pulmonary exacerbations (PEX) in patients with cystic fibrosis (CF) is crucial in the management of CF. In Oct 2004, we implemented a pulmonary exacerbation scoring (PES) system to uniformly and consistently identify PEX in CF patients as part of a quality improvement project at the CF Center at Akron Children's Hospital. We have previously reported improved pulmonary function outcomes in pediatric patients at our center after implementation of the PES. We now describe the impact of the PES on clinical practice in the management of PEX and pulmonary function in our adult CF patients.

The PES was developed by our multidisciplinary CF team after an extensive literature review and was implemented in Oct 2004. The PES has been previously described and consists of 13 clinical questions divided into three domains: systemic and pulmonary symptoms and signs, and objective measurements. These elements are scored individually and a cumulative PES of ≥ 5 (range 0-17) is considered a PEX and treatment with antibiotics is recommended. The course and choice of antibiotic regimen is left to the physician's discretion.

We examined median percent predicted FEV 1 decline from quarterly percent predicted FEV 1 measurements obtained from Port CF. We calculated this for all the patients seen in our adult CF clinic (age>19) from Oct 2004 till Sep 2006 during which the PES was in use. We compared this rate of decline to median percent predicted FEV 1 decline for the same cohort of patients over the preceding two years (Oct 2002 till Sep 2004 , during which the PES was not in use. The percent utilization of the PES was calculated individually for all adult CF care providers. The influence of individual components of PES in decision to treat was also studied.

We found no significant difference between the rates of decline in percent predicted FEV 1 during the pre-PES and post-PES periods (Mean = -.25, SD = 1.58 vs Mean = -.59, SD = 1.66, p=0.29) as measured by paired samples t-test. Logistic regression analysis was done to evaluate the relative importance of each PES component in predicting decision to treat a PEX. We found that changes in FEV 1 (OR=1.87), cough (OR=2.64), sputum (OR=3.28), and chest examination (OR=1.51) are most likely to affect the decision to treat where as dyspnea (OR=0.62) has the little effect on decision to treat. We found a high degree of correlation between adult provider use of the PES and an independent chart reviewer (correlation coefficient 0.66-0.87; p<0.001).

In contrast to CF patients ages 6-18 years, we found no significant effect of use of the PES on pulmonary function in our adult population, despite the high frequency of use by our adult providers. It is unclear why this is the case given the robust effect of the PES on our pediatric population. This study suggests that a simplified score that includes FEV 1 , cough, sputum, and chest exam may be useful in identifying a PEX in adult CF patients.

Special thanks to Tasha Capozzi for her help with chart review. This project was supported by the Akron Children's Hospital Foundation. Objective Chronic pulmonary infection with Pseudomonas aeruginosa is the main cause for morbidity and mortality in cystic fibrosis (CF) patients. Tobramycin and colistin, widely used inhaled antibiotics in CF patients, were found to affect elastase activity in vitro. We showed recently that CXCR1 on neutrophils mediates bacterial killing, but is cleaved in CF airways by elastase. Therefore, we examined the effects of inhalation with tobramycin and colistin on CXCR1 expression and antibacterial capacities by airway neutrophils in CF patients in vivo.

Methods CXCR1 expression was quantified by flow cytometry on neutrophils in peripheral blood and induced sputum of 25 CF patients without inhalation therapy, 25 CF patients with tobramycin and 25 CF patients with colistin inhalation therapy. The longitudinal effect of a 2 weeks tobramycin or colistin inhalation period on CXCR1 expression, free elastase levels and bacterial killing by airway neutrophils was assessed.

Results Inhalation with tobramycin increased, while inhalation with colistin decreased CXCR1 expression on neutrophils in induced sputum of CF patients. Neutrophils isolated from sputum of CF patients after tobramycin therapy had higher, whereas patients after colistin therapy had lower bacterial killing capacity and α-defensin release as compared to neutrophils before antibiotic inhalation therapy.

Conclusions Tobramycin and colistin differentially modulate the pulmonary host defense in CF patients in vivo via CXCR1 on neutrophils. These findings may have clinical implications when considering antibiotic treatment in CF patients.

Sharp, J. 1 ; Sheehan, D. 1 ; Ren, C.L. 2 1. Pediatrics, Women & Children's Hospital, Buffalo, NY, USA; 2. Pediatrics, University of Rochester, Rochester, NY, USA Background: Airway infection and inflammation occur early in infants with CF. The raised volume/rapid thoracoabdominal (RV/RTC) technique is more sensitive than older infant pulmonary function test (iPFT) methods in detecting airflow obstruction in these patients. However, there are limited data regarding lung function measured by RV/RTC in infants with CF NBS. We hypothesized that abnormalities of lung function measured by RV/RTC are present in CF NBS infants. To test this hypothesis, we reviewed the iPFT data from CF NBS infants followed at our CF Centers.

Methods: This study was a retrospective chart review. We identified infants with CF NBS followed at our CF Centers who had had at least 1 iPFT performed in the first year of life. IPFTs were performed using the Collins IPL. Forced expiratory flows were measured using RV/RTC as described by Feher et al (J Appl Physiol 80:2019 , 1996 and fractional lung volumes were measured by whole body plethysmography (BP) as described by Castile, et al (Ped Pulm 30:215, 2000) . Data from the CF infants were compared to normal data reported by Jones, et al (AJRCCM 161:353, 2000) for RV/RTC and Castile, et al for BP.

Results: We reviewed data from 22 infants. Their mean age was 9.7 m (range=4-24). Pancreatic insufficiency was present in 20 (91%) infants, and 3 (12.5%) had been treated for Pseudomonas infection. Their PFT findings are summarized in the Table. FVC and FEV0.5 were not significantly different from the normal predicted values. However, the mean FEF75 and FEF25-75 Z scores were significantly below the mean of normal infants (P=0.02 and 0.03 respectively). Gas trapping was present, as evidenced by significantly increased FRC, RV, and RV/TLC (P<0.001 for all 3 measures). Six infants (27%) had an FEF75 Z score of < -2.0, and 20 (91%) had an FRC > 100% predicted.

Conclusions: Infants with CF NBS have abnormal lung function early in life. Our results suggest that PFTs in CF NBS infants can be useful in detecting early changes in lung function. We speculate that early therapeutic interventions may improve lung function in this group of patients. Table. PFT Findings in Infants with CF NBS. Forced expiratory flows are reported as Z scores, while fractional lung volumes are reported as % predicted. Objective: Clinicians use PFTs to assess effectiveness of therapy in hospitalized CF patients. However, the variability of multiple PFT efforts has not been described in this group of patients. Previous study of within day variation of PFTs has been reported in healthy adults and children (Enright, 2004, and Nickerson, 1980) and clinically stable outpatients with CF (Nickerson, 1980, and Cooper, 1990) . Nickerson and Cooper found coefficient of variations (CV) for FVC of 1.9-6%, FEV 1 of 2.4-5.3%, and FEF 25-75 of 7.4-9.3% for adults and children with CF. No one has studied the within day variation of PFTs in CF patients during pulmonary exacerbations. Our objective was to determine the within day variation of PFTs during hospitalizations for CF pulmonary exacerbations. We hypothesized that within day variation would improve during treatment.

Methods: We retrospectively reviewed PFT data for all patients admitted to Childrens' Hospital and Regional Medical Center for a CF pulmonary exacerbation in 2005. Patients performed PFTs under the direction of experienced pulmonary function technicians. The test was complete when the patient performed 3 efforts that met ATS/ERS criteria for acceptability. If the patient was unable to produce 3 efforts because of coughing, 2 efforts were accepted if the FVC and FEV 1 were within 5% or 150 mL of each other. For analysis, we included only those PFTs: (1) with at least 2-3 flow-volume curves that fulfilled ATS/ERS criteria, and (2) were performed within 3 days of hospital admission or discharge. For patients with multiple admissions, we used PFT data for their first admission only. For patients with admission and discharge PFTs available, we compared the coefficient of variation (CV) for PFTs (FVC, FEV 1 , ). There were at least 8 days between admission and discharge PFTs. The Wilcoxon rank-order test was used to compare the CV for admission and discharge PFTs.

Data: Sixty-one patients were admitted 114 times in 2005. Excluding repeat hospital admissions, 17 patients had individual PFT trials available at the time of admission and 13 patients at the time of discharge. Seven admission and 2 discharge PFTs included only 2 trials. Three patients at admission, and 2 patients at discharge, were unable to produce PFT curves that met ATS/ERS criteria, despite previously having done so when well. Nine patients had paired admission and discharge PFTs from the same hospitalization. The CV for FVC at discharge was significantly better than at admission, p = 0.02 (see Table) .

Conclusions: Within day variation of PFTs during hospitalizations for CF pulmonary exacerbations is comparable to published outpatient data. CV for FVC improves significantly during exacerbation. Changes in FVC > 4%, FEV 1 > 3%, and FEF 25-75 >10% can be considered a significant spirometric response to therapy in hospitalized CF patients. 

Background: Lung Clearance Index (LCI), a measure of ventilation heterogeneity, can be calculated from multiple breath washouts of inert gas. In cross sectional studies it is a more sensitive measure of early airway dysfunction in CF than spirometry alone. In normal subjects, the technique is reproducible, with a narrow normal range of 5.9-7.5. We hypothesized that it may be helpful in detecting change longitudinally, and may have value in assessing novel therapeutic interventions such as the upcoming gene therapy by the UK CFGT Consortium.

Methods: Patients aged 10 years and over, presenting with an exacerbation requiring intravenous antibiotics, were recruited. LCI was assessed in triplicate by multiple breath washout of 0.2% Sulphur Hexafluoride (SF 6 ), using a novel gas analyser (Innocor, Denmark) and custom-built software within the first 24 hours of starting treatment. The test was repeated at the end of parenteral antibiotic therapy and again 4-10 weeks later when clinically stable. Patients were recruited at three sites and assessed using standardised equipment and a data analysis protocol.

Results: Paired LCI measurements were available in 33 patients, mean age 24 (range 11-44) years, 18 male. In a further 4 patients, technically acceptable washouts could not be obtained at one or other time point. Mean (SD) LCI improved from 14.7 (2.9) to 13.9 (2.4), p<0.005. In 11/33 patients there was an improvement in LCI of >10% after antibiotics. 2 patients showed a >10% deterioration in LCI. Mean FEV 1 was significantly greater after treatment (2.34 v 2.11 L/s, p<0.05, n=18). There was a weak correlation between percent change in FEV 1 and LCI (r 2 =0.33, p<0.05). 26 patients also had LCI measured at the follow up. Mean (SD) LCI at this visit was 13.9 (2.5), p<0.05 vs visit 1, no significant difference vs visit 2.

Conclusions: LCI can be used to demonstrate changes in the lung after antibiotic therapy. This is the first study in CF to demonstrate a lon-gitudinal improvement after intervention in a marker of ventilation heterogeneity, and offers different but complimentary information to spirometry. The higher sensitivity of LCI may make possible measurement of improvement in mild patients after novel therapies such as gene therapy.

This work was funded by the Cystic Fibrosis Trust. Aim: Inhaled hypertonic saline positively affects sputum production in patients with cystic fibrosis (CF) by improving mucociliary clearance. This study was conducted to compare the aerosol delivery performance of various nebulizers upon aerosolization of hypertonic saline (Hyper-Sal™) 3.5% and 7%, available as preservative-free, sterile 4 ml solution in single dose, patient friendly blow-fill seal vials.

Method: 4 ml hypertonic saline (Hyper-Sal™) 7% was nebulized by three jet nebulizers the PARI LC PLUS®, LC STAR® and Hudson RCI MICRO MIST as well as the eFlow® for Hypertonic Saline (HS) generating aerosols via a perforated vibrating membrane. HS 3.5% was characterized in the eFlow®HS. Droplet size distribution patterns were assessed by laser diffraction at a flow rate of 15 L/min and aerosol delivery performance by breath simulation tests mimicking a sinusoidal breathing pattern. The delivered dose (DD) was determined by gravimetric analysis of the inhalation filters. Inhalation and exhalation filters were changed after 2 min and the drug delivery rate (DDR) was calculated as NaCl found on inhalation filters per min. Respirable Drug Delivery Rate (RDDR; DDR x Respirable Fraction <5 µm) indicates how much NaCl can reach the lungs per minute.

Results: Aerosol characteristics of the different nebulizers after aerosolization of 4 ml HS 7% are summarized in the table. Data for HS 3.5% nebulized by the eFlow®HS were as follows: DD= 61.1%, DDR= 9.3 %, nebulization time = 6.6 min and comparable to data for HS 7%. The lower recoveries for the jet nebulizers are probably associated with increased evaporation and longer nebulization times.

Conclusion: About 18 -59% of the charged sodium chloride from 4 ml HS was delivered and 57% -76% of these droplets were in a respirable size range <5 µm resulting in vitro respirable doses of about 10% -40% of the label claim. DDR and RDDR are about 3-to 10-fold higher for eFlow®HS indicating a much higher delivery efficiency compared to conventional jet nebulizers. Further studies will be needed to demonstrate that about 2 ml HS may be sufficient to deliver a comparable dose to the lungs using an eFlow®HS compared with jet nebulizers. This bears the potential to further reduce nebulization time and increase acceptance by patients. Objectives: Individuals with CF frequently have complicated and time consuming treatment regimens consisting of inhaled antibiotics, mucolytics, and airway clearance. Patients and clinicians may have difficulty determining which specific therapies are effective for any individual patient or whether a particular treatment is causing adverse effects. This study assessed the utility of home spirometry for providing an objective measure of pulmonary status in between scheduled clinic visits.

Methods: Consecutive adults with CF were recruited at a single academic medical center and were asked to measure FEV1 twice daily for six months using the PiKo-1 meter (nSpire Health™, LLC). 35 subjects agreed to participate and 24 completed the study. Subjects were given diaries to record changes in respiratory symptoms and changes in medications during the study period. Subjects were evaluated in clinic at least every three months and FEV 1 values obtained in the pulmonary function lab were compared to home FEV 1 from the same day using linear regression and Bland-Altman analyses. Graphical displays of home FEV 1 measures were visually inspected for changing trends and these were compared with medication changes.

Results: The mean FEV 1 measured at home was 2.14±0.2L. The mean FEV 1 in the pulmonary lab was 2.35±0.22L. The home FEV 1 was significantly correlated with the clinic value from the same day, R 2 =0.94, p<0.001. FEV 1 measures fell within the limits of agreement (±2 SD) by Bland-Altman plot. In 7 of 24 subjects (29.2%) there was at least one sustained change in FEV 1 that correlated with either stopping or starting a particular treatment. The figure shows two significant changes in FEV 1 in one patient related to medication changes. There is an increase in FEV 1 at the initiation of aerosolized 7% saline and a fall in FEV 1 that corresponded to inhaled tobramycin use. The improvement on hypertonic saline increased this patient's enthusiasm for this treatment. Conclusions: Home FEV 1 monitoring is feasible and appears to provide accurate measurement of pulmonary function. It can provide clinicians with objective data to tailor treatments without necessitating a clinic visit. Feedback from home monitoring may improve medication adherence and may allow treatments to be individually tailored based on patient responses to the medications. 1 1. Respiratory Unit, Royal Hospital for Sick Children, Glasgow, United Kingdom; 2. Diagnostic Imaging, NHS Greater Glasgow & Clyde, Glasgow, United Kingdom; 3. Medical Statistics, Royal Hospital for Sick Children, Glasgow, United Kingdom; 4. Medical Physics Department, Gartnavel Hospital, Glasgow, United Kingdom Background: There have been calls for caution regarding the suggested use of regular chest computed tomography (CT) scans for monitoring disease progression in CF 1 . Consideration of the associations of higher ionizing radiation exposure must be made before regular chest CTs are performed on the wider CF population.

Aim: We tested the hypothesis that higher ionizing radiation exposure from chest radiographs (CXR) and chest CTs would be associated with markers of severity including gender, genotype (Class I/II or III-V), height SDS, weight SDS, BMI, FEV 1 SDS, length of hospital admission (days), Scottish Index of Multiple Deprivation (SIMD) 2 scores and PA/BC microbiological infection (chronic or intermittent colonisation with Pseudomonas aeruginosa and/or Burkholderia complex).

Methods: Effective doses of ionizing radiation (mSv) from CXRs and chest CTs (including HR scans) were determined for all CF children at our centre between 1st January 2003 and 11th July 2006 (184 weeks). The most recent anthropometric and lung function data to July 2006 were used. SIMD scores report a data zone that encompasses ~750 individuals and comprises 31 indicators in six domains (current income, employment, health, education, housing and access).

Results: 134 CF patients (67 male:67 female) underwent 1853 radiation exposure procedures (mean of 3.9 procedures per patient per year). The average effective doses were calculated for 1806 of the 1853 procedures (97.5%) including 1591 CXRs and 35 chest CTs. Lung function data were available on 89 patients (66%). Univariate analysis revealed that higher ionizing radiation exposure was significantly associated with FEV 1 SDS (p<0.001), length of stay (p<0.001) and PA/BC infection (p<0.001). The regression line indicates that for each drop of two FEV 1 SDS, each child receives an extra 3mSv in radiation exposure from CXRs and chest CTs alone. Similarly, each child receives an extra 2mSv in radiation exposure from CXRs and chest CTs for every 100 days of hospital admission. Multivariate analysis with backward selection demonstrated that length of hospital admission (p<0.001) and PA/BC infection (p=0.05) were associated with higher ionizing radiation exposure.

Discussion: In CF, higher ionizing radiation exposure from CXRs and chest CTs in childhood are associated with poor lung function, PA/BC infection and length of hospital admission but not other indicators of severity such as gender, severe genotype, anthropometric measures or socio-economic status. Using effective doses of ionizing radiation as an outcome marker in CF is an alternative means of assessing the impact of the disease and its management on the child. Background: Exhaled volatile compounds have attracted attention as potential non-invasive biomarkers of chronic lung disease. Early sustained airway inflammation and bacterial infection are characteristic features of pulmonary involvement in cystic fibrosis (CF). The cellular response is dominated by neutrophils, which have been shown to generate oxygen-and halogen-derived radicals as part of their host defense armamentarium.

Aims: To gather pilot data on the pulmonary output or uptake of selected organic trace gases in CF patients and healthy controls; and to explore the impact of pulmonary exacerbations and antibiotic treatment.

Methods: Samples of ambient air and exhaled breath were collected from 30 children, adolescents and young adults with CF and from 39 healthy controls. The patients' age ranged from 9-29 years and their FEV 1 between 21-114 % predicted, 27 were Pseudomonas positive. Twelve patients were studied at the start of i.v. antibiotic treatment for a pulmonary exacerbation; repeated sampling after the course was performed in 7 subjects. All samples were analyzed on a customized gas chromatography system with flame ionization and electron capture detectors. After determining the mixing ratios of a variety of volatile organic compounds, alveolar gradients were calculated from the corresponding exhaled and ambient values.

Results: Compared to healthy control subjects, CF patients showed a significantly lower respiratory output of methanol, ethanol, acetaldehyde and dimethylsulfide but a higher net release of benzene, methylchloride, trichloromethane and trichloroethene. These differences were most pronounced in patients with acute exacerbations. A relevant scatter and overlap between groups was noted for ethane, pentane, acetone, isoprene, toluene, and tetrachloromethane. After antibiotic treatment, the output of benzene, methylchloride and trichloromethane decreased by 40-250%.

Conclusion: Although important pathways remain incompletely understood, the local biochemical milieu in CF lungs may be informative and accessible through non-invasive breath testing. In this context, the diagnostic potential of exhaled organic traces gases (especially oxygenated, halogenated and aromatic compounds) should be further explored.

Supported by: German Research Foundation (DFG BA 1895/3-2).

Toy, E. 2 ; Weiner, J. 2 ; Sacco, P. 1 ; Duh, M.S. 2 1. Health Economics, Novartis Pharmaceuticals, East Hanover, NJ, USA; 2. Analysis Group, Inc., Boston, MA, USA Objective: To review up-to-date economic outcomes data in patients with cystic fibrosis (CF), and in particular, costs related to respiratory infection by Pseudomonas aeruginosa (Pa), which is the leading cause of morbidity and mortality in CF patients.

Methods: A systematic search of the MEDLINE database from 1990-2007 was conducted to identify major and review articles that contained the terms "cystic fibrosis" and "cost" and which were published in English language journals. Selected conference abstracts were searched, and additional articles were identified through bibliographies of retrieved articles. Recent articles that contained economic data on antibiotic and mucolytic therapies were selected for in-depth review. Cost estimates were converted to 2006 US dollars for comparability.

Results: Approximately 300 articles fulfilled the initial search criteria, and 27 were selected for in-depth review. Articles were divided into four categories: economic impact of inhaled tobramycin (2 articles), the effect of home-vs. hos-pital-based antibiotic therapies for pulmonary exacerbations (4 articles), economic impact of recombinant human deoxyribonuclease (rhDNase) (10 articles), and cost-of-illness for CF (11 articles). Inhaled tobramycin (actual or recommended dose of 300 mg/5ml) has been associated with reductions in health care costs; these cost savings offset between 37% and 57% of the cost of this therapy. Antibiotic therapy for pulmonary exacerbations generally resulted in lower health care costs when administered in a home setting rather than a hospital setting. Use of rhDNase led to reduction in health care costs, with higher reductions observed in patients with higher levels of use; the cost savings offset between 17% and 38% of the drug cost. Cost-of-illness studies have been conducted in 7 different countries; the economic estimates varied widely across studies due to differences in treatment patterns, health systems, methodologies, and patient populations. Most cost-of-illness studies were retrospective observational studies from the perspective of a hospital or third-party payer, and the cost of CF ranged from $9,000 to $64,000 per patient per year. The largest cost categories included hospitalizations, out-patient visits, rhDNase and antibiotics, while disease severity and presence of Pa infection were common determinants of cost. Cost-of-illness studies have underestimated societal costs because they have rarely considered patient time costs and none have considered indirect costs (e.g., burden on lost productivity or informal caregivers).

Conclusions: Studies show that inhaled tobramycin and rhDNase can partially offset medical costs; home-based IV antibiotic therapy is likely to reduce costs; and direct costs associated with CF can be high but vary widely across countries and analytical methodologies. Areas for future research include direct comparisons of inhaled antibiotic therapies, examination of the relationship between treatment adherence and economic outcomes, and estimation of societal cost-of illness.

This study was sponsored by Novartis Pharmaceuticals Corporation, East Hanover, NJ. 

Background: Sputum interleukin 8 (IL8) and myeloperoxidase (MPO) are advocated as markers for infective exacerbations in CF. We have previously shown that the neutrophil protein, calprotectin, is a sputum biomarker of disease activity in cross sectional and longitudinal studies. In the present study we evaluate longitudinal measurement of sputum and serum calprotectin during infective exacerbation.

Methods: Sputum and venous blood samples were taken at the beginning (visit 1) and end of a course of IV antibiotics (visit2), and at time of clinical stability following the infective episode (visit 3). Processed sputum supernatant was analysed by ELISA for calprotectin, IL8 and MPO. Serum calprotectin was assayed by ELISA.

Results: 36 paired sputum samples samples were available for visits 1 and 2 and 25 paired samples for visits 2 and 3. All data are presented as mean(SEM). Sputum calprotectin decreased from 1123(134) to 852(136) µg/ml, p<0.05, from visit 1 to 2. There was a non-significant increase to 1052(159) µg/ml at visit 3. Sputum IL8 did not change from visit 1 to 2 [13.3(1.5) 6(4.2) ]. Serum calprotectin decreased from 51(9.5) to 16.5 (2.5) ng/ml, p<0.0001 (n=35 paired samples) from visit 1 to 2. At visit 3 there was an increase to 18(2) ng/ml, p<0.05 (n=28).

Discussion: Calprotectin appears a valuable marker both in sputum and serum for tracking changes in lung inflammation during an exacerbation of CF and may be more sensitive to change than IL8 or MPO. On stopping antibiotic therapy sputum and serum calprotectin increase implying that this marker may be sensitive to small sub-clinical changes in lung inflammation. These data suggest that sputum and serum calprotectin might be employed to assess responses to drug therapies in CF such as gene therapy.

This research was funded by the CF Trust UK. Chronic lung disorders are usually associated with a hypoxia driven increase in red cell mass. However, patients with cystic fibrosis often have normal or decreased haemoglobin levels. The present prospective observational study in cystic fibrosis (CF) patients was performed to determine which factors were involved in alterations in the hematopoetic response to corresponding arterial oxygen pressure.

Sixty adult patients (age 21-51) with stable CF were included. They all had vitamin A, D, E and K but no vitamin B12 supplementation. 25 patients were on oral Fe2+ (100mg/day). Resting arterial blood gases, lung function, complete blood counts, parameters of iron status, CRP, sputum microbiology and serum erythropoietin were measured at recruitment and after 3 and 6 months.

Patients had varying degrees of pulmonary functional impairment and 9% were hypoxemic (arterial oxygen pressure < 60 mm Hg). Low-grade systemic inflammation (CRP > 0.5 mg/dl) was present in 40% of the patients, who all had bacterial colonization. None of the patient had erythrocytosis and 12 patients had anemia . There was no significant difference in iron status between patients with or without chronic iron supplementation and erythropoietin levels were normal. During the 6 months observation period no significant changes occurred.

The patients exhibited an impaired erythropoietic response to hypoxemia with normal or low hematocrit in spite of chronic lung disease. Linear multivariate regression analysis (Table) revealed CRP and iron levels but neither iron substitution, nor erythropoietin levels nor lung function parameters as independent determinant of haemoglobin levels.

CF may be associated with anemia of variable severity as expression of the chronic inflammation present in these patients. The therapeutic consequences are to treat the underlying inflammation rather than to supplement iron.

Variables included in initial model: Background Improvement in the treatment and survival of females with CF has led to many leading a normal adult lifestyle, and some can now expect to raise a family. However, pregnancy can have a profound impact on wellbeing, and caring for a young child can affect compliance with CF treatment. To look at this further, we surveyed the experience of pregnancy and motherhood in CF women attending our large adult CF centre (230 patients). Methods We reviewed the records of 10 CF women who chose to remain pregnant (12 babies [7 male] from 11 pregnancies), looking at changes in lung function, BMI, respiratory exacerbations during pregnancy, clinic attendance, and oral and IV antibiotic therapy (home and inpatient care) during pregnancy and also in the first year post delivery. We devised a questionnaire to examine how patients coped with pregnancy and the demands of motherhood, focussing on changes in compliance with therapy, worries regarding the effect of CF treatments on the baby, and support given by the CF team.

Results At conception mean FEV1 was 82% predicted (range 61 to 107) and mean BMI 21 (range 18 to 26).There were no miscarriages or neonatal mortalities and all children remain well, but the sickest mother died at 4 years. Mean FEV1 following pregnancy was 73% [range 51 to 113] (P=0.01), with 1 mother losing 21% by one year post delivery. At one year, BMI was unchanged (mean 21, range 16 to 27). Rates of respiratory exacerbations and IV antibiotics were similar pre and during pregnancy, although during the first year post delivery 2 mothers avoided hospital based treatment preferring home care. Six mothers' clinic attendances increased during and post pregnancy (1 attended outpatients 17 times in the first year post delivery).

Nine mothers worried about the effects of CF medication on their unborn child: the remaining mother was in her 2nd pregnancy. All managed their health differently during pregnancy: whilst 6 increased compliance, worked harder to stay healthy, and increased clinic attendances; 4 avoided clinic, postponed treatment, avoided any medication in the first trimester, and reduced oral antibiotic use to prevent harm to their unborn child.

Following delivery, 7 mothers spent less time on their own health due to motherhood demands, and 8 missed treatments during the first year: 4 missed all except IV antibiotics. Seven mothers ceased chest clearance and nebuliser therapy, and whilst 4 missed oral therapies only 1 avoided IV therapy. One mother who returned to work early felt this had contributed to her post natal depression.

Nine mothers indicated that CF team support was adequate, but the remaining mother felt that she should have been offered more frequent home visits post delivery due to lack of family support.

Conclusions The experience of pregnancy in our centre has positive outcomes for most mothers. However, compliance with treatment was affected and a number of mothers lost significant pulmonary function due to pregnancy. Our study has shown that the demands of motherhood especially in the first year challenge the management of their own health. We offer a proactive service to women considering pregnancy, with early participation of the multidisciplinary team, and have increased home visit frequency to during the first year post-delivery.

Geller, D.E.; Kesser, K.C. Research, Nemours Children's Clinic, Orlando, FL, USA Background: Alpha-1 antiprotease (α-1AP) is being developed as an inhaled drug in CF to inactivate excess elastase in the lungs. It is estimated that tens of milligrams of α-1AP need to be deposited in the lungs to be effective. Delivering a high dose with conventional jet nebulizers may be too time-consuming. We speculated that slowing and prolonging inspiration could optimize α-1AP delivery. The I-neb® Adaptive Aerosol Delivery (AAD® system, Respironics) is a portable, electronic, battery-powered, mesh nebulizer that uses the AAD algorithms to deliver drug in the first portion of inspiration. Interchangeable mouthpieces allow the I-neb to be used in a conventional tidal breathing mode (TBM) or in the target inhalation mode (TIM) that guides the patient to inhale very slowly. The low residual volume after nebulization reduces drug waste and optimizes efficiency. We compared delivery efficiency of α-1AP in vitro with the I-neb in the TBM vs. 2 TIM modes. Methods: We studied 3 new I-neb devices using TBM (tidal volume [Vt] 0.5L, respiratory rate [RR]=15, I-time=2 seconds), TIM-6 (Vt 1.6L, RR=6, I-time=6 sec), and TIM-9 (Vt 1.6L, RR=4, I-time=9 sec) using a breath simulator. Nebulizers were filled with 0.5 mL α-1AP (50 mg/mL). Nebulized drug was captured on a filter and measured via spectrophotometer (inhaled dose), and particle size was measured with an Insitec laser system. Results: Median particle sizes for the 3 modes were similar (4.4-4.8 µm) . The dose emitted from the I-neb was very high, and similar between modes (82-90% of nominal dose). Predicted lung doses for these experiments were calculated from a previous scintigraphy study (Nikander 2006 ) that showed 63% of the emitted dose deposited with TBM and 73% with TIM. Predicted lung dose of α-1AP (as % of nominal) and delivery times were TBM: 56.6% in 7.5 min, TIM-6: 59.9% in 4.4 min, and TIM-9: 64.5% in 2.5 min. Discussion: The I-neb has a very high predicted lung dose with both TBM and TIM. However, TIM reduces the time of administration to as little as 1/3 that of tidal breathing. If larger predicted lung dose is necessary to see a clinical effect, either the initial drug volume or drug concentration may be increased. We conclude that slow, deep, controlled inspirations using I-neb is a very efficient method to deliver α-1AP, and is faster than tidal breathing. We also speculate that TIM has the additional potential advantage of better distribution of drug in the lungs because of slower flow rate and particle velocity. Objective: early inflammation was observed in infants with CF regardless the occurrence of bacterial infection. Lung function (LF) is a tool to assess bronchial inflammation consequences. Few data are available on the pronostic value of impaired LF in infancy.

Methods: In infants, airflow obstruction (T PTEF/TE , curve shapes, tidal volume, functional residual capacity (FRC, nitrogen washout technique) were assessed with the Sensor Medics 2600 and interpreted with Stocks reference values. At 6y, FEV 1 and pulmonary flows (Spiro 2000, Nova medical SC, Zapletal reference values) and FRC He , were measured. Symptomatic children were defined when cough, wheeze or rattle were recurrent. All symptomatic infants have been treated with inhaled steroids as any infantile wheezer.

Results: 30 screened infants (11 ∆F508/∆F508, 14∆F508/other, 5 other/other) were investigated at mean age 16mo and 6 years. 16 (54%) had respiratory symptoms (RS); the mean tidal volume was normal (9.5ml/Kg), 19 (64%) had a marked obstructive curve shape, 9 (30%) a decreased T PTEF/TE , and 7 (23%) hyperinflation. Then, 7 (23%) had normal LF, 16 (53%) AO, 3 (10%) hyperinflation and 4 (13%) both. At age 6, 50% had RS, and LF with mild alteration as mean flows ≥ 95%, and FRC He = 97.7%. Accurate analyses showed that 10 (33%) had normal LF, 9 (30%) AO and 11 (47%) small airway diseases (only distal AO and/or hyperinflation). FRC in infancy and at age 6 were correlated (r=0.579; p=0.03). Conversely, no correlation between AO parameters in infancy and at age 6 was found. The LF alteration was independent of the genotype, bacterial colonization and recurrent symptoms at both ages.

Conclusion: AO in infancy is frequent as found in half of the screened CF infants. The lack of correlation with the AO observed at age 6 suggested that AO in infants might be due to a non specific inflammation (i.e viral), transient and then relevant from a treatment. Hyperinflation, although mild, is conversely an early marker of the CF disease as persistent between infancy and age 6. We describe the efficacy of a novel system using an electronic real-time distant home monitoring of symptoms and spirometry in CF patients with the aim of early detection of P Exs

Patients and methods: This is a 6 month prospective observational study. All patients were provided with a set consisted of a mobile phone-palm PC (the X DAII, O2-UK) and with a spirometer (vitalograph) with an attachment to the PC. The PC contained a purpose-built software on which CF symptoms and spirometric values were recorded once daily. These were then sent on a real time to a computer server where the data were plotted on a time-magnitude graph. The site was accessed by a password-permitted CF physicians.

Patients were asked to score four symptoms (cough, sputum, breathlessness and fatigue) once a day and to record at least three attempts of spirometry. These were accepted by the software only if they met the ATS criteria. Preset criteria for P Exs were stipulated. When the recordings met the criteria for P Exs patients were invited to come to the CF clinic for review and were managed by either watchful waiting, oral or IV antibiotics. A prodrome phase is deemed to be present when a patient had an increase in one or more symptom for one week prior to the P Ex.

Twenty one evaluable patients were included in the study. In 19 (90%) there were an increase in one or more than one recorded symptoms throughout the study. A total of 50 P Exs median 1.2 per patient were detected within the 6 months study period. In 21(42%) P Exs, symptom increase and/or decline in FEV1 met the preset criteria for P Ex and in 29 cases patients presented despite that symptom changes were not sufficient to meet the pre-set definition of P Ex. Prodromal phase prior to P Exs were seen in 36 (72%) patients. Adherence and technical issues were the main two problems throughout the study.

Distant daily monitoring of patients symptom is easy and sensitive. It can be used to monitor symptoms and early detect P Exs. The vast majority of our patients have an increase in their symptoms even when they perceive their disease to be stable. A prodormal phase preceded P Ex is seen in the majority of patients. The pre-set criteria for defining P Ex may have been too stringent.

The research was funded by O2 telecommunication, UK

The monitoring system used in the study. A combination of a mobile phone-palm computer (X DAII, O2 UK) and a spirometer (Vitalograph UK) .. The aim of this research was to examine whether systemic inflammation is present in adult patients with CF during their stable status.

Patients and methods: Blood samples were obtained from a total of 17 CF patients 6 female, mean age 23, range(18-26)11 males mean 23.5 range15-36 and from 32 control subjects, 20 female, mean age 21.3, range(21-40)12males,mean age27.1yrs, range(21-33) All CF patients were included when they were at least 2 weeks away from any pulmonary exacerbations. All control subjects were not known to have any pulmonary or systemic disease.

Venous blood samples were taken and spun at3000 rpm.Separated sera was aliquted and frozen at -70 degrees C. They were analysed in batches for 6 cytokines IL-1B,IL-6,IL-8, IL-10, ILp70,and TNF-ALPA using a CBA human inflammatory kit(BD Bioscience). Whitney-U test was used to compare values of serum cytokines between stable CF patients and control subjects..

Serum IL-6 and IL-8 were raised in CF patients compared to control subjects. The mean value for IL-6 was 8.2pg/ml (SD 2.0) for patients and 2.3pg/ml (SD 0.3) for controls, P<0.0001. Similarly, the mean value of serum IL8 was 29.6pg/ml (SD 6.7) in CF patients compared to 8.6pg/ml (SD 0.64) in the control group, P<0.0001 (figure below).

There was no difference between CF patients and control with regards to serum IL-1b, IL-10 and IL-12P70.

Conclusion: serum IL-6 and IL-8 were raised in CF patients compared to an agematched control subjects. Inflammatory markers were raised in the sera of CF patients even during disease stability. Even during disease stability, the lungs remain an inflamed organ which could explain the decline in lung function tests and the countinued daily systemic symptoms experienced by patients with CF.

Serum IL6 and IL8 in CF patients with stable disease compared to the control group (please see abstract body) We have shown that even during disease stability, an increase in inflammatory markers are present in the sera of CF patients compared to an agematched control group. This longitudinal study examines changes in serum inflammatory markers during P Exs.

We investigated 22 CF patients 6 female, mean age 23, range15-36 years. Their mean baseline FEV1 was 62.8% of predicted (range 27.7-99 %).

A baseline venous blood sample was withdrawn and sera were stored at -70 degree C from 22 patients during disease stability. Out of those there 17 experienced P Exs. Blood samples were obtained at the beginning and at the end of the P Exs.The following inflammatory markers were analysed: White blood counts, C -reactive protein (CRP), IL-6, IL-8, IL-10, IL-1b, IL-12p70 and TNF-α.

Patients recorded their symptoms and their FEV1 on a daily basis on an electronic diary and the data were beamed daily to a research centre where data were analysed. P Ex was defined as 1). an increase in two symptoms for 3 successive days, or 2). an increase in 2 symptoms and 10 % reduced FEV1 from baseline over 3 successive days or 3). 10% decline in FEV1 over 3 successive days or 4).when a patient felt that he/she was going through a P Ex. Paired t tests were used to analyse changes in inflammatory markers.

There was no change in the number of circulating Neutrophils, lymphocytes or eosinophiles at the start compared with the end of P Ex. There was an increase in mean CRP values at the start of P Ex and reduction at the end of P Ex, but this did not attain statistical significance. Mean (SD) of serum IL6 increased from 8.8pg/ml (2.8) at baseline to 16.18pg/ml (4.0) at the start of P Ex (P=0.007) and returned to base line 8.88pg/ml (2.5) at the end of P Exs (P=0.012 compared to values at the start of P Ex). (Figure 1) There were no changes in the values of serum IL-8, IL-10, IL-1b, IL12p70 or TNF-α at the start and at the end of P Exs.

Serum IL6 increased at the start of P Ex and returned at the baseline at the end of P Ex. CRP was the closest inflammatory marker to follow that pattern. None of the other inflammatory markers changed with the diagnosis of P Ex's. This and other studies call for the price of analysis of IL-6 to be reduced to become available in routine clinical practice. Background. ABPA can be a therapeutic challenge in children with CF. Aim was to describe the tolerability of voriconazole and nebulised liposomal amphotericin in the treatment of ABPA in children with CF.

Methods. We performed a retrospective case note review of voriconazole and nebulised liposomal amphotericin use, in the treatment of ABPA in children with CF, at our centre.

Results. 7 children diagnosed with ABPA according to the current criteria, failed to improve despite previous treatment with itraconazole and steroids, and received voriconazole as a second line agent. The dose of voriconazole was as recommended by the manufacturer's summary of product characteristics and the duration of treatment was two weeks. Significant side effects noted in 4 children. Observed side effects included visual hallucinations (n=1), photopsia (n=2), sleep disturbances (n=2), severe headache (n=2), influenza like symptoms (n=1), gastrointestinal side effects (n=1) and elevation of alkaline phosphatase (n=1). Of these 4 children, one child tolerated voriconazole when concomitant treatment with omeprazole was withdrawn.

A second course of voriconazole, given at a lower dose and without the loading dose was tolerated by two children. Voriconazole blood levels were not measured in any patient.

We used nebulised liposomal amphotericin in 4 of these 7 children. One child received amphotericin as a second line agent when voriconazole had to be stopped because of significant side effects. 3 children had active ABPA despite treatment with first line (itraconazole and steroids) and second line (voriconazole) agents and received amphotericin as a third line agent. Amphotericin was used at a starting dose of 2mg nebulised twice daily, with weekly increments as tolerated, up to a maximum dose of 15mg twice daily for four weeks. The starting dose and all subsequent increments in dose were supervised. All children received salbutamol immediately before nebulised amphotericin. The standard formulation available for intravenous infusion was made up into single doses for nebulisation by the pharmacy. This was stored at 4 0 C and diluted with water for injection for use with the nebuliser. The only side effect, experienced by all 4 children was a dry irritant cough at the start of nebulisation that settled soon without need for discontinuation of the medication. Improvement in clinical symptoms, lung function and total serum IgE was noted in all 7 children. The improvement in FEV1 and FVC, expressed as mean (range) of percentage predicted were 18 and 20 respectively. The mean (range) fall in serum total IgE and aspergillus specific IgE were 1395kU/L (240-3299) and 11 mg/L (4-18) respectively.

Conclusion.

Voriconazole use was associated with significant side effects. Avoidance of drugs that may potentially result in interaction is important. Omitting a loading dose and using a lower maintenance dose may help selected patients tolerate voriconazole. Nebulised liposomal amphotericin was tolerated by our children with CF and ABPA who did not tolerate or responded poorly to voriconazole. Recent studies have shown that high resolution computed tomography (HRCT) is a more accurate method of evaluating early lung disease in cystic fibrosis (CF) than pulmonary function tests (PFTs) or chest x-rays. Since young children can not perform the breathing maneuvers required for high quality HRCT, lung volume control is necessary. Positive pressure facemask ventilation using chloral hydrate allows full inspiration and end exhalation images, but requires special expertise. Controlled ventilation can also be produced with general anesthesia (GA). GA has not been reported in children undergoing HRCT for CF research. The advantages of GA over sedation include rapid onset, faster recovery and reliability of scan completion. The undesirable effects of sedation are unpredictability and the higher failure rates. Additionally, failed sedation contributed significantly to parental dissatisfaction. (Malviya, et al., 2000) . The primary outcome of our study was to examine the safety of using GA for HRCT scans and the quality of images obtained. A secondary outcome was parental satisfaction.

Methods: Participants were recruited from a larger NIH/CFFTI funded multi-site randomized clinical trial focusing on behavior and nutrition treatment to improve growth in preschoolers with CF. HRCT in subjects 2 to 5 years old (mean age 38 ±10 months) was performed with GA. Parents accompanied the child during induction and left after the child was asleep prior to the intravenous (IV) being placed. After induction of anesthesia, a ProSeal®, specialized laryngeal mask airway (LMA), was placed which allowed ventilation with higher pressures and the ability to suction the stomach. Volumetric thin section CT scans were obtained. GA using inhalation (sevoflurane) and/or IV (propofol) agents was used and 6 out of 7 families who were eligible (85% recruitment rate) participated in the sub-study.

Results:

The quality of the HRCT scans in all 6 subjects were optimal. There were no anesthetic complications upon emergence and none of the patients experienced any side effects after follow-up phone calls at 24 hours. Parents also expressed that the clinical interpretation of the HRCT scan provided beneficial information about their child's current lung disease status. The mean time from completion of scan to discharge was 38.8±15 minutes.

Conclusions: The use of GA in Radiology departments is becoming more common in pediatric centers. Our initial findings of this study demonstrate that GA can be safely administered for children with CF. In addition to very good quality scans being obtained, parental satisfaction was also noted. The use of the ProSeal® LMA for HRCT may improve the image quality and aid in the diagnosis of early lung disease in preschoolers with CF.

Supported by NIH Grant R01 DK054915 and a CF Foundation Clinical Research Grant. Whey protein is rich in sulfhydryl (SH) groups and is recognized for its ability to increase glutathione and reduce oxidative stress. Hyperbaric pressure treatment of whey protein increases its digestibility, promotes the release of novel peptides for absorption, increases intracellular glutathione in healthy subjects (Zavorsky et al, Int J Vit Nutr Res, 2007) , and reduces in vitro production of IL-8 (Vilela et al, Inter Immnuopharmacol, 2006) . We present here the initial results of a pilot study of supplementation with pressurized whey in CF patients. Children and adults with CF were supplemented with hyperbaric whey protein for 1 month (20 gm/day in patients less than 18 years of age, 40 gm/day in older patients). Anthropometric measures, pulmonary function, serum C-Reactive Protein (CRP) and plasma lipid peroxides were measured before and after supplementation. The results of the first 11 patients to have completed the study are reported, with further data to be presented. In 7/11 subjects weight increased and in 7/11 subjects lean body mass increased. There were no significant changes in FEV1 or FEV/FVC ratio. In 7/11 patients the CRP fell (4/7 by more than 50%) and in 2/11 the CRP increased. In the remaining two patients, the CRP was undetectable at baseline and did not change at follow-up. Plasma lipid peroxides decreased in the three patients in whom it could be measured (2/3 to undetectable levels). These preliminary results suggest that supplementation with pressurized whey protein can reduce systemic inflammation in CF patients and has the potential to beneficially modulate the inflammatory response in CF patients.

Supported by the BREATHE Initiative of the Canadian CF Foundation with Pseudomonas aeruginosa. Soon after its introduction, the mesh technology based Pari eFlow® nebulizer gained preference over jet nebulization in the CF community because of its short nebulization time, portability and noiseless operation. Concerns were raised about the durability of the device due to expected vulnerability of the mesh. The aim of this study was to compare the particle size distributions in the aerosol, delivered doses and output rates from the eFlow and the Pari LC Plus®(LC+)with PariTur-boBoy compressor before and after use in daily practice. LC+ nebulizers in use by the adult patients taking part in the study were taken in for testing and new eFlows, cleaning instructions and a disinfection device were provided. After 6 months use (3 cycles of 2x28 doses TSI: 300 mg/5 ml), the eFlows were collected by the local CF center. Particle size distributions were measured at a constant flow rate of 30 l/min with a laser diffraction apparatus. Because of uncertainty regarding the cleaning procedure after last use, 8 additional e-Flows were collected from the same patients after another period of 6 months use. Conclusions: New eFlows produce slightly larger droplets in a narrower size range than new LC+ nebulizers. After use in daily practice, droplet size distribution changes for both devices. Output rate (delivered dose per minute) decreases for both devices. For used eFlows delivered dose is lower than for used LC+ nebulizers, because the eFlow is programmed to switch off after 10 min, even when nebulization is still incomplete (9 out of 21 tested eFlows). This may result in a considerable reduction of delivered dose. However, when a mesh is replaced, performance is the same as that of a new device, suggesting that the changes in performance originate in the vibrating mesh. For daily practice these results indicate: (1) the performance of both nebulizers changes during use but the changes are smaller for eFlow than for LC+ (2) changes in droplet size distribution are relatively small for both devices;in contrast with output rate and delivered dose (3) proper functioning of eFlow can be improved by timely replacement of the mesh, ultimately when eFlow switches off after 10 minutes. Without replacement, the delivered dose is unpredictable.

mean values for size distribution, delivered dose, nebulization time and output rate (X 50 is volume median diameter) Backgrounds: Lung surfactant protein A (SP-A) belongs to the family of collagen containing, C-type lectins, called collectins. SP-A plays a critical role in the innate immune defense of the lung. SP-A binds calcium dependent to carbohydrate structures on pathogens via its carbohydrate-binding domain (CRD). Cystic fibrosis (CF) lung disease is characterized by chronic inflammatory and proteolytic processes in the airways, leading to destruction of the lungs and early death.

Aims: We hypothesized that SP-A might have a reduced function in CF. This might result from SP-A degradation by neutrophil proteases, an altered macromolecular organization of SP-A, or inhibition of CRD-binding by airway components.

Methods: We studied bronchoalveolar lavage fluid (BALF) from 31 clinically stable patients with CF and from 13 children suffering from bronchitis. BALF samples were purified by carbohydrate-binding assay on fucose attached to Sepharose resin with and without calcium. In the absence of calcium SP-A cannot bind via its CRD. Each fraction was analysed for SP-A and IgG by immunoblotting. Macromolecular organization of SP-A was determined by gel chromatography.

Results: Only 49 % of total SP-A of CF BALF bound to the fucose column, whereas 92 % bound from the bronchitis BALF. The structural organization of the SP-A oligomers did not differ between the non-binding and the binding fractions of both groups. 4 of the CF BALF had SP-A at 27 kD that did not bind to fucose. These BALFs had a high percentage of neutrophils (> 80%). In vitro human leucocyte elastase decreased SP-A to less than 25% of its initial levels and SP-A fragments at 27 kD were detected.

Conclusions: Neutrophil proteases degrade SP-A in CF lungs. The carbohydrate dependent binding of SP-A is diminished due to a reduced function of the CRD. These alterations contribute to the reduced ability in CF to remove specific pathogens from the lungs.

Chiron, R. 1,6 ; Murris-Espin, M. 2, 6 ; Crampette, L. 3,6 ; Varrin, M. 4 ; Didier, A. 2, 6 ; Wallaert, B. 5, 6 ; Chanez, P. 1, 6 ; Leroy, S. 5, 6 1. CRCM, CHU, Montpellier, France; 2. CRCM, CHU, Toulouse, France; 3. ORL, CHU, Montpellier, France; 4. IURC, CHU, Montpellier, France; 5. CRCM, CHU, Lille, France; Montpellier, France In cystic fibrosis (CF) patients, upper airways involvement is frequent but its impact and relationships on the course and severity of the disease has been rarely investigated.

OBJECTIVE: To relate upper airways involvement with lower airways parameters and quality of life in 91 CF patients.

METHODS: Clinical and radiological prospective study of subjects with CF at 3 French CF centres. Clinical rhino-sinusitis (RS) was defined by the presence of more than 3 episodes of nasal symptoms during the last year, current nasal obstruction or rhinorrhea. Nasal investigations included: anterior rhinoscopy, sinus CT scan analysis and RS quality of life questionnaire. Lower airways were assessed using a thoracic CT scan (Bhalla score), spirometry and sputum culture and a CF quality of life questionnaire (CFQ14+).

RESULTS: 91 (47F/44M) patients aged 25.4 (14-71) years old were enrolled. Endoscopy demonstrated polyposis in 12% and nasal hyper secretions in 67.5% of the patients. Spirometry, Bhalla score, sputum colonization did not discriminate patients with (n=61) or without RS (n=30), and regardless the presence of polyps. CFQ14+ was significantly lower in patients with of RS (p=0.04).

CONCLUSIONS: RS involvement altered quality of life in CF patients, although, they do not have a more severe pulmonary disease. A specific attention and treatment to RS should be paid in CF patients. The impact on different outcomes deserves to be examined in longitudinal studies.

As part of the UK Cystic Fibrosis Gene Therapy Consortium "Tracking Study", EBC was measured (ECoScreen, Jaeger) in adult and paediatric patients presenting with an infective exacerbation requiring IV antibiotics. Exacerbation was defined by pre-determined criteria. EBC was repeated at the end of parenteral antibiotic therapy and again 2-4 weeks later. Patients were recruited at three sites and assessed using standardised equipment. Samples were obtained according to ERS guidelines. Exhaled breath pH was assayed (Shindengen pH meter, Camlab, UK) immediately after collection without a de-aeration step.

Results: Paired EBC measurements at the start and end of antibiotic therapy were available in 40 patients. Mean age was 24yrs (range 11-44). Data from 6 patients were excluded because at least one pH measurement was >7, suggesting salivary contamination. For the remaining 34 patients mean EBC (SD) pH increased from 5.9 (0.5) to 6.1 (0.3), p<0.01.

Of these patients, 28 also had measurements available at a third visit, 2-4 weeks after completion of antibiotics. 2 were excluded because of salivary contamination. No difference in mean(SD) EBC pH was found (6.1 (0.4) vs. 6.1 (0.4)) Conclusions: EBC pH increases after treatment with antibiotics, and may offer a non-invasive means of assessing airways inflammation in chest exacerbations. Further work is required to follow the longitudinal change in EBC pH and other inflammatory markers in clinically stable patients, as EBC may be a useful tool in assessing response to novel treatments, such as gene therapy.

This work was funded by the Cystic Fibrosis Trust. Background: Children with cystic fibrosis (CF) undergo multiple procedures exposing them to ionizing radiation in hospital. Some specialist centres have advocated regular chest computed tomography (CT) as a means of better quantifying lung pathology. The clinical benefit of this approach is unproven and previous calculations of risk have failed to account for other ionizing radiation exposure procedures 1 .

Aim: To quantify the exposure of ionizing radiation in the pediatric CF population of a tertiary level children's hospital. To determine the potential difference in exposure of introducing biannual chest CT scans.

Methods: Effective doses of ionizing radiation from chest (CXR) and abdominal (AXR) radiographs, fluoroscopy and CTs (sinuses and chest including HR scans) were determined for CF children at our centre between 1st January 2003 and 11th July 2006 (184 weeks). The maximal effect was tested i.e. introduction of biannual chest CT scans from age 2 years abolishing the requirement for CXRs and the assumption was made that AXRs, sinus CTs and fluoroscopy rates were unchanged (median exposure values calculated for age groups 0-<1year, 1-<2year, 2-<4year, 4-<8year, 8-<12year, >12year) .

Results: CF patients (n=134, gender ratio 1:1) underwent a mean of 13.8 ionizing radiation exposure procedures per patient over the study period (total of 1853 procedures). The average effective ionizing radiation doses (mSv) were calculated for 1806 (97.5%) procedures. Each CF child received a mean (median) effective radiation dose of 1.47 (0.04) mSv/year with a range of 0.004 -25.76mSv/year. The reason for the skewed distribution was identified by analysing the proportions of the procedures to the total effective dose in this population: fluoroscopy 83.2%, chest CT 7.1%, sinus CT 6.3%, CXRs 2.5%, AXRs 0.7% and HRCT 0.2%. It was calculated that these patients undergo 3.4 CXRs per patient per year. We calculated that an average patient diagnosed by newborn screening could expect to receive 1.24mSv from ionizing radiation procedures between birth and age 16. Biannual CT chest scans in addition to median exposures for each age group from AXRs, sinus CTs and fluoroscopy would potentially result in 14.34 mSv by age 16, an 11 fold increase.

Discussion: There is a 1 in 100 lifetime risk of developing a solid or haematological cancer with each exposure to 100 mSv of ionizing radiation 2 . Although fluoroscopy forms the most significant contribution to a pediatric CF patient's ionizing radiation exposure in our centre, introducing biannual chest CTs would markedly increase exposure. As CF survival improves, CF physicians must be cognisant of the potential health cost of survival.

References : Previous studies have used generic measures of HRQOL, which are not sensitive to important changes that may occur as a result of treatment. This study assesses the impact of IV antibiotic treatment for a pulmonary exacerbation on both generic and disease-specific HRQOL of children and adolescents with CF.

Method: Participants included 45 patients (M age = 13.4 years) from the Cincinnati Children's Hospital and University of Florida CF Centers. Fiftyfour percent of the sample were male. Patients completed two HRQOL measures, the PedsQL™ (generic) and CFQ-R Child or Teen/Adult version (CF-specific), which have excellent reliability and validity. These measures were completed within 48 hours of hospital admission (pre) and 48 hours of discontinuation of IV antibiotics (post). Note that approximately 32% of patients went home on IV antibiotics and 68% remained in the hospital for approximately 2 weeks.

Results: Significant improvements were found in pulmonary functioning (FEV1 % predicted) pre to post IV antibiotic treatment (paired t (44) = -6.9, p < .0001), with an average of 14% increase in FEV1 % predicted. Paired t-tests were conducted to examine changes in HRQOL scores, using Holm's procedure to correct for multiple comparisons. Significant improvements were found on the CFQ-R Respiratory (paired t (45) = -5.3, p < .0001) and Weight scales (paired t (23) = -4.1, p < .0001). Trends were also noted for improved Emotional Functioning (paired t (45) = -1.8, p = .08) and Vitality (paired t (23) = -1.9, p = .07), and worse Treatment Burden (paired t (45) = 1.7, p = .09). On the PedsQL, only Emotional functioning improved significantly (paired t (44) = -2.6, p < .05).

Conclusions: Results of this study confirm the effectiveness of IV antibiotics for the treatment of pulmonary exacerbations, with significant improvements found for both pulmonary functioning and HRQOL. These results also highlighted the importance of using a disease-specific HRQOL instrument, which proved to be more sensitive than the generic measure. On the CFQ-R, significant improvements were observed in respiratory symptoms and weight, aspects of functioning that are not assessed on generic measures.

Fukushima, L.K.; Hsu, E.; Woo, M.S. Division of Pediatric Pulmonology, Childrens Hospital Los Angeles, Los Angeles, CA, USA Do Hispanic and Caucasian CF patients share the same frequency of pulmonary exacerbations? Increased frequency of pulmonary exacerbations is linked to increased risk for CF morbidity and mortality. We have previously reported that pediatric Hispanic CF patients have greater mortality than our Caucasian CF patients who attend the same accredited CF Center. We performed a retrospective review of CF pulmonary exacerbations (defined as physician prescription of intravenous or oral antibiotic therapy) in our CF patients who routinely attended the CF Clinic during the period of January 1, 2005 through December 31, 2006. Data collected included demographics, number of intravenous antibiotic courses and number of oral antibiotic courses. Unpaired Student's t-test was used to compare age of the groups and Chi-square was used to compare use of intravenous and oral antibiotics between Hispanic and Caucasian CF patients as well as between males and females. During the study period, there was a total of 197 CF patients (106 Males:91 Females; mean age 10.8 ± 5.7 years). Of the 197 patients, 82 (42%) were classified as Hispanic (46 Males:36 Females; mean age 11.0 ± 5.3 years) and 99 (50%)were classified as Caucasian (49 Males:50 Females; mean age 10.8 ± 6.2 years; p=0.42, ns). Hispanic CF patients (51 patients had 299 episodes; 26 Males:25 Females) had significantly greater number of pulmonary exacerbations (received treatment with intravenous and/or oral antibiotics) than Caucasian CF patients (35 patients had 182 episodes; p = 0.0006). Hispanic CF patients who required intravenous antibiotic therapy were significantly younger (12.3 ± 5.1 years vs 15.0 ± 5.9 years; p<0.00001) than the Caucasian CF patients who were treated with intravenous antibiotics. There was no significant difference in age between Hispanic and Caucasian CF patients who received oral antibiotic courses (11.2 ± 5.3 years vs 13.1 ± 6.4 years; p=0.08, ns). Gender did not have a significant impact on pulmonary exacerbation risk in our population (Hispanic IV use by gender p=0.82; Hispanic oral use by gender p=0.12; Caucasian IV use by gender p=0.38; Caucasian oral use by gender p=0.37). We conclude that Hispanic CF patients had increased incidence of pulmonary exacerbations than Caucasian CF patients who attend the same CF Clinic. Hispanic patients who were treated with intravenous antibiotics were younger than the Caucasian CF patients. We speculate that Hispanic ethnicity has a major impact on the incidence of pulmonary exacerbations. Other factors, such as modifier genes, environmental exposures, or inflammatory responses, may play a role in the increased risk of pulmonary exacerbations in the Hispanic CF population. Introduction: One of the main goals of the multidisciplinary team of the CF Association of Minas Gerais, Brazil is to find strategies to preserve lung tissue from the bacterial exacerbation in cystic fibrosis patients.

Objective: To investigate clues on onset of bacterial exacerbation. To our knowledge, no studies were found regarding the identification of the early warning signs of bacterial exacerbation with CF patients. We wanted to know whether CF patients present early warning signs before onset of fever and/or breathing difficulties.

Methods: Telephone interviews were carried out with the 330 CF patients registered in the Association, by a physiotherapist academic who did not know the patients and just ask two questions. The first question was about the use of antibiotics during the year of 2006 and the second was related to early signs before the onset of fever.

Results: From the 330 patients, only 207 patients replied.

Patients who were using antibiotics therapy during 2006 and had fever: 117/207

Patients who were using antibiotics and did not have fever: 37/207 Patients who had fever and did not use antibiotics: 9/207 Patients who did not have fever and did not use antibiotics: 44/207

The results revealed the following warning signs and also that some patients had more than one sign.

Tiredness, quietness, slowness (51/117); increased cough (27/117); lack of appetite (19/117); sticky sputum (13/117); sleepiness (12/117); headache (14/117); increased excitement, nervousness (3/117); crybaby (11/117); dry mouth (1/117); clammy (1/117); difficult to walk (1/117); gastric upset (5/117); itchy throat (6/117); abdominal pain (7/117) ; itchy eyes (10/117); breathing difficulties (7/117); tremulous (3/117); paleness (3/117); burning ear (1/117); chest pain (1/117); hissing (1/117) ; and syncope(1/117).

Conclusions: It is helpful to know these early warning sings. We assume that the earlier the diagnosis of the exacerbation is made, the easier it is to resolve the problem and the less destruction of the lung tissues.

Reference: Lung Line, National Jewish center for Immunology and Respiratory Medicine 1988. Background: Cystic Fibrosis patients are experiencing increasing survival and are being treated with more chronic therapies. How these changes are reflected in day to day clinical symptoms has not been evaluated.

Objective: We examined data from the Epidemiologic Study of Cystic Fibrosis, a large longitudinal observational study, to characterize the change in respiratory symptoms over time. For each year between 1995 and 2005 data from patients with at least one visit recorded as occurring while clinically stable were included. Data from all visits, including sick as well as stable, were used for each year in which at least one clinically stable visit was recorded. Patients were separated into age groups less than 6, 6 to 12, 13 to 17, and 18 or older. Note that in 2003 data collection changed, potentially confounding the results.

Results: The average number of patients per year was 16393. The percent of patients with no reported cough progressively increased over time (table) . Results were similar for pateints with no reported sputum production at clinically stable visits. When sick visits were added there were similar results. The frequency of cough reported at greater than 90 percent of visits progressively decreased over time (table) . Again the results for sputum production were similar. These findings were seen in all age groups although older patients are more likely to experience these symptoms.

Conclusions: These data suggest that CF patients are experiencing fewer respiratory symptoms of cough and sputum production over the last decade. This improvement has occurred in patients with either intermittent or chronic symptoms. How clinical care has impacted on these symptoms has yet to be determined. Also, more people are being diagnosed with CF at an earlier age or with milder disease, potentially resulting in the entire CF population having fewer symptoms.

Lung disease in children with CF begins in early childhood, with intermittent and then chronic infection, associated with an exaggerated inflammatory response. The extent to which inflammation contributes to bronchiectasis has not yet been identified. The aim of this study was examine relationships between inflammation and early lung damage Methods: Twenty eight children with CF were assessed at diagnosis and annually thereafter with a bronchoalveolar lavage (BAL). BAL fluid was used to assess micro-organisms and inflammation, including total and differential cell counts. A three slice HRCT technique (inspiration and expiration) was performed under general anesthesia at age 3-6y, immediately preceding the annual BAL. HRCT scans were scored by an independent and experienced radiologist using a modification of the Brody Score.

Results. Twelve Children (43%) had evidence of bronchiectasis on HRCT. Inflammation present in BAL at 12 months of age predicted bronchiectasis at 3-6 y. Total cell count at 12 mo, predominantly neutrophils, was higher in those who developed bronchiectasis compared to those who did not (0.81 vs 0.43 x10 6 cells/ml BAL; p=0.047). The total number of infections detected in BAL was not related to bronchectasis. However, children with bronchiectasis had a higher incidence of Pseudomonas aeruginosa within the first 3 years of life (50% vs. 25%). Conclusions: Bronchiectasis is evident within the first five years of life and appears to be associated with an exaggerated inflammatory response and the early acquisition of Pseudomonas aeruginosa.

Supported by: Cystic Fibrosis Foundation (USA), National Health and Medical Research Council (Australia). Fortunately the survival of patients with cystic fibrosis has improved remarkably over the last few decades. Many patients can life quite "normal". Therefore family planning becomes more important. Several reports docu-mented the good maternal and fetal outcome in pregnant women with cystic fibrosis. Besides the data from the French CF registry no longterm data about the impact of paternity of male patients with cystic fibrosis are available until now.

In a retrospective analysis we gathered data about all our male patients who became father while being followed-up in our cystic fibrosis center for adults. From 43 male patients, 6 patients became father between 1999 and 2006. Patient no. 1 became father twice (second time father of twins); patients no. 6 became also father of twins .We looked at FEV1 % 1 year prior to childbirth, perinatal and 1 and 5 years after the childbirth. For the FEV1% one year prior to birth as well as for FEV1% 1 and 5 years after birth, we calculated the difference to the FEV1% at time of birth.

Between 1999 and 2006, 7 male adult patients with cystic fibrosis became fathers of a total of 9 children, all by the assistance of reproductive techniques. Their FEV1% are shown in table 1.

Although the male patients are not directly influenced by a pregenancy (as the pregnant women are by growth of the uterus), there seems to be more than ususal differences in this lung function parameter. There were also more fluctuations in FEV1 per year. The biggest differences in FEV1% were seen 1 year after birth of their children.

Further investigations in a larger group (in cooperation with other centers) are planned; especially with a focus on the frequency of infections and quality of life compared to male patients without children. Intravenous (i.v) aminoglycosides are widely used in cystic fibrosis (CF) patients as treatment for pulmonary exacerbations. In an effort to reduce undesiderable effects of these antibiotics it is recommended to measure drug serum levels. Recent experience suggests that tobramycin (TO) may be administered as a single daily dose with equal efficacy as multiple doses but less risk of nephrotoxicity.

Aim: We evaluated the efficacy of once-daily i.v TO treatment to improve pulmonary function expressed as forced expiratory volume in one second (FEV 1 ) in 15 CF patients with pulmonary exacerbations compared to the same group of patients who were previously treated with TO in multiple daily i.v doses, as historic control. We evaluated baseline and peak drug serum levels of TO administered once daily compared to multiple dose levels.

Methods: All patients were treated with i.v TO in combination with a beta-lactam antibiotic for two weeks. Respiratory exacerbation was defined according to CFF criteria. All CF patients were colonized by non-fermentative Gram-negative bacteria.

Serum concentrations of TO were measured (fluorescent polarisation immunoassays) before the fifth infusion (basal) and 30 minutes after the fifth infusion (peak). Reference TO blood levels were considered to be ≤ 2 mg/L (basal) and <12 mg/L (peak) for multiple daily doses, and ≤ 1 mg/L (basal) and 20-30 mg/L (peak) for single daily dose.

Treatment efficacy was evaluated on the basis of improved clinical condition and increased FEV 1 values compared to the beginning of therapy.

Results: 15 patients with respiratory exacerbation were evaluated, with a mean (± DS) age of 27.05 years (± 6.09; median 25.83, range 16.8-39.4) . 14 out of 15 patients were chronically colonized by Pseudomonas aeruginosa. The total mean (± SD) daily dose of TO administered in multiple doses was 410 (± 68.66) mg whereas the mean single dose was 406.6 (± 37.1) mg.

Of those patients given multiple doses of TO, only one patient (6.6%) had basal TO values outside the range and 4 patients (26.6%) had peak values outside the range. Of those patients given a single daily dose, no patients had basal TO values outside the range whereas 12 out of 15 (80%) had peak values of less than 20 mg/L and 3 patients out of 15 (20%) had peaks between 20 and 30 mg/L. The mean (± SD) increase in FEV 1 was 5.6% (± 9.09) in patients treated with a single dose and 7% (± 9.41) in those treated with multiple doses. The mean (± SD) period between FEV 1 measurements before and after exacerbation was 59.53 days (± 53.4, median 37, range 6-201) in case of monodose therapy, and 36.93 (± 24.79, median 29, range 17-105) in case of therapy with multiple doses.

Conclusions: Both single and multiple-dose i.v TO therapy caused a comparable FEV 1 increase and were effective in resolving the respiratory exacerbation. Although clinical improvement was observed, the mean i.v TO dose of 400 mg/day in patients treated with once-daily i.v TO determined peak values between 20 and 30 mg/L in 20% of subjects. Intravenously administered aminoglycosides are efficacious for the treatment of pulmonary infections of cystic fibrosis (CF) patients, with the principal adverse effects of these molecules being acoustic nerve damage and nephropathy. Damage to the eighth cranial nerve varies from 1.8 to 32% in CF patients treated with multiple daily doses of aminoglycosides. Recent experience suggests that tobramycin can be administered as a single daily dose, resulting in less nephrotoxicity but equal efficacy.

Patients and methods: We evaluated the prevalence of oto-and nephrotoxicity due to aminoglycosides administered to CF patients followed at the Tuscan Regional Center where we are currently following 192 patients. Cochlear damage was evaluated using tonal audiometry (Amplifon Amplaid 171 S). Renal damage was evaluated by measuring patients' creatinine clearance.

Results: 96 (50%) patients (52 males and 44 females) between the ages of 9 and 43 (mean age 22.5 years, median 21, SD 9) were tested with audiometry. All these patients had been repeatedly treated with multiple daily doses of intravenous aminoglycosides. 15 (7.8%) (5 males and 10 females) out of 192 patients between 3 and 45 years (mean age 24.8, median 24.2, SD 13.1) had their creatinine clearance measured.

We found that 7 patients (7.2%) had an abnormal audiometry, typically attributable to aminoglycoside ototoxicity (high frequency deficit). Two patients required a hearing aid. One patient with normal cochlear function had labyrinth deficit.

One out of 15 patients tested had pathological creatinine clearance values.

Conclusions: Our data indicate that 8 (8.3%) of our patients had damage to the eighth cranial nerve due to repeated multiple daily doses of aminoglycosides. We recommend that CF patients under aminoglycoside therapy be routinely given audiometric and creatinine clearance exams in an attempt to reduce the incidence of undesirable side effects of these drugs.

Davies, J. 1, 6 ; Voase, N. 1, 6 ; Dewar, M. 2,6 ; Mullard, K. 1,6 ; Gammie, F. 1, 6 ; Saunders, C. 1, 6 ; Horsley, A. 2, 6 ; Gray, R. 2, 6 ; MacLeod, K. 3, 6 ; Somerton, L. 1,6 ; Higgins, T. 1,6 ; Donovan, J. 1, 6 ; Cornish, N. 1, 6 ; Hansell, D. 4, 6 ; Aziz, Z. 4, 6 ; Ashby, D. 5 ; Geddes, D. 1, 6 ; Greening, A. 2, 6 ; Cunningham, S. 3,6 ; Innes, A. 2,6 ; Alton, E. 1, 6 1. Gene Therapy, Imperial College, London, United Kingdom; 2. Western General Hospital, Edinburgh, United Kingdom; 3. Royal Hospital for Sick Children, Edinburgh, United Kingdom; 4. Royal Brompton Hospital, London, United Kingdom; 5. Queen Mary, University of London, London, United Kingdom; 6. UK CF Gene Therapy Consortium, Edinburgh, London, Oxford, United Kingdom The improving clinical status of patients with CF and the slow rate of decline of conventional outcome measures such as FEV 1 necessitate the development of clinically-relevant surrogate outcome assays for use in clinical trials. In our forthcoming clinical trial of CFTR gene therapy, the UK CF Gene Therapy Consortium will use both established, clinically-available assays and more novel measures designed specifically for this purpose. Both groups of assays are being subjected to rigorous testing prior to use to establish their utility as disease biomarkers. In this study, we examined their performance in the context of an infective exacerbation treated with IV antibiotics. This abstract will report the response of established, clinically-available assays; available data from novel assays will be reported separately.

Children (12 years and above) and adults with CF were recruited from three centres at the time of a clinician-defined infective exacerbation requiring IV antibiotic treatment. Exclusion criteria included FEV 1 <30% predicted and requirement for supplemental oxygen. A panel of assays was performed at the start and end of treatment, which was most commonly 14 days. Data are presented as mean (SD) or median (range) and differences compared with either parametric or non-parametric paired analysis as appropriate. Clinical assays included spirometry, sputum microbiology, sputum cell count and differential, serum inflammatory markers (CRP, white blood cell (WBC) count,) and chest CT. Patients also completed a symptom score sheet.

To date, 40 patients (mean age 24.2 [7.5] years, 21 male) have paired data available from the start and end of the course of IVs. At baseline, 24 were infected with P. aeruginosa, 8 with B. cepacia complex organisms and 14 with S. aureus (2 MRSA). Significant changes from baseline were observed in FEV 1 (53.3[15.1] The clinical response to any novel intervention, for example, CFTR gene therapy is difficult to predict. Prior testing of experimental assays in a study such as this provides data on the variability of the measurements within the disease population and the degree of change observed with an intervention known to lead to clinical benefit. This should aid the design of rational, powered clinical trials.

Funded by the CF Trust Influenza can worsen CF and lead to serious complications and mortality. Influenza vaccine is safe and effective in CF patients. In France, annual influenza vaccination is recommended for all CF patients aged over 6 months and for healthcare workers in regular and prolonged contact with high-risk patients.

Objective:

To estimate the influenza vaccination coverage rate for 2005-2006 season in CF patients and healthcare workers in CF care centres of the Great South region of France.

Methods: a multicentric observational study between February and September 2006 was conducted. Healthcare workers were contacted by telephone and were questioned about their 2005-2006 influenza vaccination status, and the reason for not getting vaccinated. For patients aged over 6 months whose vaccination booklet was available, the physician filled in a written questionnaire.

Results: A total of 128 professionals were interviewed. The survey included 33.6% of doctors, 27.3% of nurses, 9.4% of dieticians, 9.4% of physiotherapists, 9.4% of social workers and 4.7% of psychologists. The overall influenza VC rate was 59.4% and varied considerably according to profession: 81.4% of doctors, 66.7% of physiotherapists, 58.3% of dieticians, 50% of psychologists, 48.6% of nurses and 16.7% of social workers. The overall influenza VC rate in CF patients was 80.6% (86.5% in children (age < 18 years old) versus 70.2% in adults; p<0.05). Receiving a voucher for free vaccination from the National Insurance increased the influenza VC rate. For healthcare workers, the main reason for non vaccination was that flu was considered as a benign disease (useless vaccine); for patients, it was a lack of time.

Conclusion Methods: Adult patients with CF were studied. Patients completed a standardised questionnaire. The questionnaire identified which changes and duration of symptoms would cause them to contact the CF team and/or change treatment.

Results: All patients would contact the CF team for haemoptysis but 40% would wait for more than 24 hours and 20% would wait for more than 48 hours before doing so. Up to 60% of patients would wait a week or more before contacting the CF team because of sputum purulence. 20% of patients would not contact the CF team if they developed new chest pain.

Conclusion: There appears to be a disconnect between recognizing a change in symptoms and the length of time before acting on the change. This has major implications for the delivery of CF care.

Supported by Cystic Fibrosis Association of Ireland. Aim: To determine the efficacy of standardized desensitization protocols in treating antibiotic allergies in adults with cystic fibrosis (CF).

Background: Dependence on high dose and multiple combination intravenous (IV) antibiotics to treat pulmonary exacerbations in adult CF patients has resulted in an increased frequency of allergic reactions when compared with the general population. (1, 2) Strategies to address these allergies are essential to maintain effective antibiotic treatment in this population. Antibiotic desensitization is the process by which one induces immune tolerance by incremental exposure to the antibiotic, which may induce stabilization of mast cells. The mechanism remains undetermined but may be mediated via IgE or memory T-cells. Prior to the study period, 9 standardized antibiotic desensitization protocols were developed by an Allergist/Immunologist based on published reports and were approved through the hospital's Pharmacy and Therapeutics committee.

Methods: The Toronto CF database was accessed to generate a list of hospital admissions for IV antibiotics during the 5-year study period (1999) (2000) (2001) (2002) (2003) (2004) . Patients who underwent desensitization were identified and each of their case notes underwent retrospective analysis. Data were collected with respect to age, organism requiring antibiotic therapy, antibiotic allergy requiring desensitization, reactions during desensitization, reactions post desensitization, treatments required to treat these reactions, and the completeness of antibiotic therapy post desensitization. Statistical analyses were performed using graph pad prism®.

Results: During the study period the CF unit dispensed and initiated approximately 600 courses of IV antibiotics. A total of 63 desensitizations were performed in 22 patients (17 female). Although some patients were desensitized to only one drug or to the same drug more than once, there were 43 patient-drug combinations with a median of 2 (+/= 1.2) unique combinations per patient. Of the 63 desensitizations protocols initiated, 7 patients developed documented reaction to the antibiotic during desensitization however only 1 desensitization protocol was not completed due to an adverse event. There were no episodes of anaphylaxis either during or after desensitization. A total of 9 patients developed documented reactions during the subsequent antibiotic course -5 required termination of the antibiotic and 4 completed the course of IV antibiotics with management of allergic type reactions with antihistamines.

Conclusions: Standardized desensitization protocols were initiated in 22 patients during the study period. The success rate for desensitizations in these patients was 79% however when the data is analyzed for true allergic phenomenon, the success rate rises to 90%. This demonstrates the efficacy of these standardized protocols in the treatment of IgE-mediated allergic reactions to antibiotics in the adult CF patient.

References: 1. Burrows, J., Toon, M., Bell, S., (2003) . Antibiotic desensitization in adults with cystic fibrosis. Respirology, 8: 359-364. 2. Ramesh, S. (2002) . Antibiotic hypersensitivity in patients with CF. Clin Rev Allerg Immun, 23: 123-139. surgery for symptomatic disease. There is a long-held suspicion that poorly-controlled chronic sinusitis negatively impacts the respiratory status of CF patients, but whether sinus surgery improves the clinical course of CF lung disease is unclear.

Aim: Assess the impact of functional endoscopic sinus surgery (FESS) on respiratory exacerbation rate and lung function in adult CF patients Methods: This is a single-center, retrospective study based on chart review. During the study period, 1999-2006, 40 adult patients underwent 55 separate FESS procedures. This abstract contains results so far from the analysis of 26 patients that underwent a total of 37 surgeries. Primary outcomes were 1-year change pre to post FESS in pulmonary function parameters and pulmonary exacerbation rate.

Results: At the time of surgery, the mean FEV1 was 62% of predicted (ppd), the mean FVC 79.6 ppd, and the median age was 24.0 years (range 18.0 -46.0). 56.1% of patients were female. The median number of IV antibiotic courses for respiratory exacerbations in the 12 months prior to FESS was 1 (range 0-5) compared to 2 (range 0-5) in the year after surgery (p = .04). Further,there were fewer days on intravenous (IV) antibiotics for respiratory exacerbations in the 12 months prior to FESS than in the 12month post-operative period (median 14 vs. 23 days), but this difference was not statistically significant (p = .09). The use of oral antibiotics for respiratory flares was comparable in the 12 months before and after FESS (median 122 vs. 104 days, respectively, p = 0.53). The median number of hospital admissions and median number of days hospitalized for respiratory exacerbations did not differ in the 12 months pre vs. post-FESS (p = 0.70 and p = 0.77, respectively). The mean FEV1 and FVC in the 12 months pre and post surgery did not differ significantly (61.9 vs. 57.7 ppd, p = 0.13, and 77.4 vs. 76.0 ppd, p = 0.11). The linear rate of change in FEV1 and FVC did not differ significantly in the 12 months prior to and following FESS (-.01 L/month vs. +.01 L/month, p = .16 and -.01 L/month vs. +.02 L/month, p = 0.11, respectively). There were no significant differences (p > 0.10) for any of the above comparisons when the analysis was limited to 6 months before and after FESS. There were no significant differences (p > 0.10) in antibiotic use and hospitalizations when the analysis was limited to the 29 surgeries in which patients were experiencing both respiratory and sinus symptoms. This latter group had lower FEV1 and FVC ppd in the 12 month post-FESS period (59.9 vs. 56.7 ppd, p = .001, and 75.4 vs. 72.9 ppd, p = .03, respectively). However, the rate of decline in FEV1 and FVC in the 12 months before and after surgery did not differ significantly ( p = 0.17 and p = 0.18, respectively).

Conclusion: FESS did not significantly affect the rate of respiratory exacerbations or rate of decline in pulmonary function in adults with cystic fibrosis. Cumulative decline in lung function as measured by FEV 1 in cystic fibrosis (CF) pulmonary exacerbation is a major cause of CF-related morbidity and mortality. High-dose extended interval aminoglycoside (HDEI AG) use may be more effective in improving lung function than traditional multiple daily dosing in CF-related pulmonary exacerbation. Intermittent dosing of beta-lactam antibiotics are often used with HDEI AG in patients with CF pulmonary exacerbation, which does not optimize betalactam pharmacodynamics. The primary purpose of this study is to determine the effect of HDEI AG plus continuous infusion beta-lactam (CIBL) on patients' return to best baseline FEV 1 from the previous 12 months.

This study was conducted at University of Kentucky HealthCare. This was a concurrent observational trial with patients serving as their own historic controls via review of the medical record. All pediatric and adult patients with CF respiratory disease hospitalized with an acute pulmonary exacerbation between November 1, 2005 and May 1, 2007, with Pseudomonas aeruginosa in their respiratory culture and were followed for at least one year were included. Excluded were patients with known hearing disability or renal insufficiency or an inability to reliably perform spirometry (ie.less than six years old). Patients served as their own historic controls via conventional treatment (control) versus HDEI AG plus CIBL (treatment). The primary endpoint was to determine the percent of patient courses that return to best baseline FEV 1 in the last twelve months after treatment with HDEI AG plus CIBL. Secondary endpoints included determination of patient courses that returned to mean baseline FEV 1 in the last twelve months after study treatments, and description of beta-lactam dosing characteristics.

Thirteen patients were included in the pilot analysis which totaled 21 patient courses (mean 1.6 courses per patient). Average patient age was 24 ± 11.1 years with a mean FEV 1 of 42%; 61.5 % were female. After HDEI AG plus CIBL no patients showed a return to their best baseline FEV 1 in the previous 12 months with the exception of one patient whose value was unchanged. Average overall percent change in best baseline FEV 1 group was -11.0 ± 7.8 at follow up. 62% of patient courses exhibited a return to their mean baseline FEV 1 . Average overall percent change in mean FEV 1 baseline was 1.8 ± 6.8 at follow up. Five of the 8 (62.5%) study patients who had serum beta-lactam concentrations obtained were able to achieve at least four times the MIC of the target organism. More data will be collected on subsequent courses and follow up lung function.

Our preliminary data suggest that our patients lost 11% of their best FEV 1 with an acute exacerbation and returned to 1% above mean baseline after treatment with HDEI AG and CIBL therapy. This evidence suggests that this might be an alternative to intermittent beta-lactam therapy. Follow up data on remaining patients may provide sufficient evidence to validate this hypothesis. Respiratory exacerbation in cystic fibrosis patients is characterized by increased sputum that may become more purulent. The detection of the exacerbation is based mainly on clinical subjective parameters. FEV1 measurements are accepted as gold standard to be used for the life-time of the patient but do not adapt fast enough after resolution of an exacerbation. We compared parameters that are affected by the uneven distribution of ventilation in the lung due to increase sputum. Methods: A Zen body plethysmograph was used to measure TLC, FEV1 and DLCO (CO/CH3 mixture). TLC was also calculated by the volume of CH3, the angle of the slope of phase III was calculated in 20 CF patients at the beginning of the respiratory exacerbation and after 2 weeks of antibiotic treatment. Patients were 17 -30 years old (12M/8F). Results: FEV1 changed by 9.8+7.41% at the end of the antibiotic treatment, the TLC(pleth) however was stable and changed by only 0.9+8.7%. The RV/TLC decreased by 23.8+ 23.4% and the TLC(gas) was increased by 5+36.6%. The difference between TLC (pleth) and TLC (gas) before and after the antibiotic treatment decreased from 24.7+13.6% to 10.2+7.1 %. Conclusion: the changes in the parameters affected by the unevenness of distribution of ventilation (m/p reflecting the increased sputum in the airway) are more pronounced than FEV1 and may be used as a more sensitive parameter for assessing the course of respiratory exacerbation.

Fibreoptic bronchoscopy was performed in 78 children aged 6-16 years: 27 with CF, 16 with non-CF bronchiectasis (BX), 24 with asthma, and 11 control children without lower respiratory disease. Endobronchial biopsies were taken and stained with haematoxylin and eosin. ASM content, myocyte number and size were quantified using stereology.

Results: The volume fraction of ASM in subepithelial tissue (median; IQR) was 0.04 (0.02-0.05) in controls. By comparison, it was 0.12 (0.6-0.21, p<0.01) in CF, 0.16 (0.04-0.21, p<0.01) in BX, and 0.27 (0.12-0.49, p<0.0001) in asthma. Myocyte number (cells per mm 2 of reticular basement membrane) was 1944 (1596-6318) in controls, 4504 (2838-8962, ns) in CF, 4971 (3476-10057, ns) in BX, and 8204 (5270-11749, p<0 .05) in asthma. Myocyte size (um 3 , mean; SD) was 1948 (386) in controls, 3364 (890, p<0.01) in CF, 2857 (873, p<0.05) in BX, and 3249 (801, p<0.01) in asthma. In CF, the volume fraction of ASM in subepithelial tissue was related to myocyte number (r=0.88, p=0.001), but not to myocyte size.

Conclusions: ASM remodeling occurs in CF children. This, however, is not disease-specific and is also found in non-CF bronchiectasis and in asthma. Both myocyte hypertrophy and hyperplasia contribute to increase in ASM in CF. Support: ERS long-term fellowship and Swiss National Foundation grant to NR Purpose: CT scans are increasingly being ordered on a routine basis in patients with cystic fibrosis. While prior studies have investigated their clinical utility in children, there is limited and conflicting information on their utility in adults. The purpose of this study was to determine the relationship between clinical disease severity as assessed by spirometry and CT findings as well as initial CT score ability to predict change in lung function in a nonselected group of adult CF patients.

Methods: A retrospective review of 45 CF patients (mean age 30.2 +/-8.5, male 62%) who had undergone a routine chest CT was performed. The CT scans were scored using the Brody method by a blinded reader with the degree of bronchiectasis, mucous plugging, peribronchial thickening, parenchymal disease, and air trapping assessed individually and a composite score generated. The CT metrics were first compared with spirometric test that were temporally related to the date of the CT scan, and then with future spirometric tests.

Results: For the cohort the patients had a wide range of severity of disease with the average FEV1% 42.9 +/-21.9, and the average FVC% was 63.5 +/-20.7. No significant correlation was found between initial FEV1% and total CT score (0.051 p value 0.739), or initial FVC% and total CT score (-0.082 p value 0.594). For CT metrics and future spirometry, the average time from CT to PFT's was 671 days +/-528 with median of 484 days. Predictors for a significant decline in PFT's were a decreasing BMI and male sex (p-values 0.015 and 0.035 respectively). Further, for patients with a low FEV1% (<40% at baseline), the FVC% change over time was associated with total CT score, male gender, and BMI (p-values 0.006, 0.007, and 0.027 respectively) Conclusions: In adult CF patients, CT scan findings did not correlate with lung function. However in longitudinal analysis, BMI and gender had an effect on FVC change. For patients with low lung function, total CT score did have an association with change in FVC along with BMI and gender. Pulmonary exacerbations (PEs) are episodes of worsening of respiratory symptoms that commonly occur in patients with cystic fibrosis (CF). In clinical trials of new therapies in CF, pulmonary exacerbation is widely used as a primary or secondary endpoint. Despite its importance, no single clearly agreed upon definition exists. Some are based on patient reported symptoms while others require hospitalization or antibiotic prescription. Use of different definitions makes it difficult to compare the results across studies and also to plan future studies. Moreover, the rate of PEs (proportion of patients with at least one exacerbation over a given time-period) is affected by baseline patient characteristics such as age, FEV 1 and medication use at the start of the clinical trial; factors which also contribute to the difficulty in comparing trials.

Using data from the CF Therapeutics Development Network (TDN) data bank, we examined several PE definitions and baseline characteristics for their effect on PE rates. We find that pulmonary exacerbation rates vary substantially depending on both the PE definition and the baseline characteristics. For example, in one study, the proportion of subjects hospitalized and/or prescribed antibiotics during the 6 month treatment period was 29%, while the proportion who met a symptoms based criteria for PE during the same period was 65%.

We also examined the implications of varying control group exacerbation rates on sample size requirements for clinical trials and find that the control group PE rate has a large impact. For illustration, we assume a two group study with two-sided significance level 0.05 and 80% power. If the control group rate is 30%, then approximately 200 subjects per group are required to detect a relative rate reduction of 40% (corresponding to a treatment group rate of 18%). However, if the control group rate is 65%, then only approximately 70 subjects per group are required to detect the same relative rate reduction of 40% (corresponding to a treatment group rate of 39%). Thus, a PE definition that selects for a greater PE rate in the control group will require fewer study subjects to detect the same relative rate reduction given that all other variables remain equal. Sample size requirements can be further reduced by analyzing the number of events or the time to first event instead of the proportion of subjects who have at least one PE during a given time period.

Supported by Cystic Fibrosis Foundation Therapeutics, Inc. (RAM-SEY03Y0), the National Center for Research Resources (NCRR MO1-RR00037) and CSL Behring. Briglia, H.; Hilliard, J.; Chmiel, J.F.; Krenicky, J.; Konstan, M.W. Pediatrics, Case Western Reserve University, Cleveland, OH, USA Cystic Fibrosis (CF) is characterized by a vicious cycle of obstruction, infection and inflammation. The inflammatory response, which is profound and excessive relative to the burden of bacteria, is characterized by a massive neutrophil (PMN) influx. CF patients often experience intermittent declines in lung function associated with increases in both bacterial burden and PMN influx. These pulmonary exacerbations require treatment with antibiotics. There are substantive issues in the identification of pulmonary exacerbations and the assessment of therapies. A marker which indicates the inflammatory state of the lung would be useful to identify infective/inflammatory exacerbations, as opposed to worsening due to pulmonary vascular disease, or simply upper airway infection (cold), and might guide therapy for intensity and duration.

G-CSF and GRO-α are cytokines produced in the lung by a variety of cells, including macrophages and epithelial cells. G-CSF is involved in the proliferation, differentiation and activation of PMN precursors, while GRO-α has known chemotactic and activating effects on PMNs. Transfer of these cytokines through the basolateral membrane into the bloodstream results in increased PMN production and activation in the bone marrow. The plasma concentrations of G-CSF and GRO-α might serve as potential biomarkers for a pulmonary exacerbation.

In this pilot study we measured G-CSF and GRO-α in the plasma of healthy volunteers (n=3), clinically stable CF patients (n=23) and CF patients experiencing a pulmonary exacerbation, both before (n=14) and following (n=10) 10-14 days of antibiotic (abx) therapy. Cytokines were measured using commercially available ELISA kits (R&D Systems, Minneapolis, MN) . T-tests were performed on natural log transformed cytokine data (SigmaStat V3.5, Systat Software Inc., San Jose, CA).

᭹ The lack of a significant difference between G-CSF and GRO-α plasma levels in healthy and clinically stable CF patients suggests that these cytokines are not indicators of chronic inflammation.

᭹ Both cytokines are increased during pulmonary exacerbations in patients with CF and decrease to baseline values following antibiotic therapy, suggesting that these cytokines might serve as potential biomarkers for CF pulmonary exacerbations.

᭹ Although not statistically significant, there was a trend towards an inverse correlation between both cytokines and FEV 1 during pulmonary exacerbations in this small study (data not shown).

᭹ Further studies are warranted to increase our understanding of G-CSF and GRO-α as potential biomarkers of pulmonary exacerbation in cystic fibrosis.

This work was supported in part by the Cystic Fibrosis Foundation through an institutional Research Development Grant.

Pletcher, S.D. 1 ; Koff, J.L. 2 ; Kleinhenz, M. 2 1. Otolaryngology, University of California, San Francisco, San Francisco, CA, USA; 2. Medicine, University of California, San Francisco, San Francisco, CA, USA Introduction: While sinus disease is common in adult patients with cystic fibrosis (CF), the severity of sinus symptoms and relationship of sinus disease to other manifestations of CF are relatively unknown.

Objectives: To evaluate the severity of sinus symptoms in adult patients with CF and correlate these findings with other CF manifestations.

Methods: Twenty-six consecutive adult patients with cystic fibrosis were surveyed during routine clinic visits for sinus specific and overall quality of life outcomes using both the Sino-Nasal Outcome Test (SNOT-20) and the Cystic Fibrosis Questionaire (CFQR). Analysis of SNOT-20 and CFQR scores were compared to %FEV1 and sputum culture growth of pseudomonas for all patients.

Results: SNOT-20 scores ranged from 0.1 to 3.1 with a mean of 1.17 on a 0-5 point scale. For the purpose of data analysis, the patients were divided into two 13 patient cohorts: Group 1 with minimal sinus symptoms (SNOT-20 < 1.0) and Group 2 with mild to moderate sinus symptoms (SNOT-20 > 1.0). The mean %FEV1 for Group 1 was significantly higher than the mean %FEV1 for Group 2 (71.5% vs. 49.9% respectively, p<0.007). Patients in Group 2 were more likely to have sputum culture growth of pseudomonas although this trend did not reach statistical significance (92% vs 62%, p=0.16). CFQR scores differed significantly among the two groups with Group 2 patients reporting more disability in the Physical, Role, Vitality, Body, Eating, Health, Respiratory, Weight, and Digestion domains but not in the Emotion, Social, and Treatment domains.

Conclusions: This cohort of adult CF patients show minimal to moderate symptoms of sinonasal disability. Those patients with mild to moderate sinonasal symptoms have decreased pulmonary function and decreased overall quality of life compared to CF patients with minimal sinus symptoms. 1, 2 1. Internal Medicine, University of Iowa Hospital and Clinics, Iowa City, IA, USA; 2. Translational Lung Imaging Program, University of Iowa Hospitals and Clinics, Iowa City, IA, USA; 3. Pulmonary and Critical Care Medicine, Mount Sinai Medical Center, Miami Beach, FL, USA Mucociliary transport is an important component of airway host defense against inhaled particles and microbial pathogens. Mucociliary clearance is hypothesized to be defective in cystic fibrosis allowing for airway colonization and infection. Current methods for measuring mucociliary transport and clearance lack the ability to measure mucociliary transport in specific proximal and distal airway segments. Using a 64-slice high resolution CT scanner (SOMATOM 64, Siemens), we are developing a method to track radiopaque particle movement in the swine airway. Swine were induced with ketamine/acepromazine, intubated, and then anesthesia maintained with propofol infusions. Animals spontaneously breathed humidified air with the endotracheal tube cuff deflated. Radiopaque Teflon/bismuth trioxide disks (1 mm diameter, 0.8 mm thick) were instilled via a catheter into the airways. Serial CT images (0.6 mm slice thickness and slice increment of 0.5 mm) were obtained every 2 minutes for a total of 20 minutes. Airway tree segmentation was performed using Pulmonary Workstation 1.1 (VIDA Diagnostics, Iowa City, IA). This program allows for fully automated airway tree segmentation. Baseline mucociliary transport rates (2-dimensional) were 15.6 ± 2.6 mm/min. Following aerosol delivery of human sputum leukocyte elastase (150 mU), mucociliary transport was markedly diminished (5.0 ± 1.1 mm/min) at 45 minutes. Thus far, we are able to measure mucociliary transport down to the 3rd generation airways. Future studies include more distal deposition and measurement of particle movement. In summary, this novel method measures mucociliary transport in defined airways and will allow us to examine the effect of altered airway epithelial function on mucociliary transport. lung transplant. Purpose: To design a CT scoring system to quantify structural abnormalities on CT scans from CF patients with Severe Advanced Lung Disease (SALD) and to investigate the correlation between CT scores and survival. Materials and methods: CT scans of 57 CF patients screened for lung transplant between 1990 and 2005 were collected from 3 transplant centres. All scans were scored with the Brody II scoring system. To design a new scoring system sensitive for the specific changes in SALD, a panel of 3 experts reviewed a random set of 10 CT scans on eligible items to be used in a pilot analysis. The resulting SALD scoring system consisted of four items: bulla/cysts; areas with consolidation/mucous; areas with hypo perfusion/air trapping and hyper/normal perfusion. In each CT slice (10 mm spacing) and for each of the items the surface area of corresponding lung tissue involved was estimated on a 0-100% scale. Total surface area for the 5 items and each slice added up to 100%. The final SALD score was a mean volume estimate of abnormal and normal lung tissue.

Results: Pilot analysis of a set of 25 CT scans showed a a spectrum of abnormalities, ranging from predominantly bronchiectasis to predominantly mosaic perfusion. A moderate correlation was found between the SALD components inflammation and hypo perfusion/ air trapping and Brody score (p=0.004, R=0.55 and p=0.006, R=0.536). The inter-observer variability of both scoring systems was comparable. (Ri ranging from 0.70 to 0.83) Analysis of the complete cohort with the Brody II score showed a significant correlation with survival with a hazard ratio of 2.4 (95% CI of 1.07 to 5.33) for dying on the waiting list for every ten point increase in Brody II score. (p=0.02). The Brody component scores for bronchiectasis, airway wall thickening and parenchyma showed a similar correlation with hazard ratios (95% CI's) of respectively 1.9 (0.98-3.73), 1.9 (1.03-3.35) and 5.3 (1.44-19.2) for every ten point increase in score. (p=0.044, 0.028 and 0.01). The preliminary results with the Brody score suggest that it is useful to include CT information in prediction models. It is likely that the predictive value of the CT can for survival can be further improved using the dedicated SALD-CT score. Scoring of the remaining scans according to the SALD system is ongoing and will be completed shortly. Snell, G. 2 1. Physiotherapy, The Alfred, Melbourne, VIC, Australia; 2. Allergy, Immunology and Respiratory Medicine, The Alfred, Melbourne, VIC, Australia; 3. Epidemiology and Preventative Medicine, Monash University, Melbourne, VIC, Australia; 4. Radiology, The Alfred, Melbourne, VIC, Australia Background Despite the widespread use of airway clearance (AC) techniques to clear excessive secretions and improve lung function, little is known about their efficacy following lung transplant (LT). This study aimed to compare the effects of two AC strategies (proactive vs reactive) on a range of clinical outcomes following LT. Methods A prospective randomized trial design with stratification for suppurative pre-LT disease was used. Consecutive patients at 1 month post LT were eligible for inclusion. Subjects were excluded if medically unstable, ventilator dependent or had a contraindication to performing positive expiratory pressure (PEP) therapy. Patients performed AC using PEP either twice daily (proactive strategy) or only in the presence chest infection (reactive strategy). Lung function (FEV 1 and FVC), chest radiography (Brasfield score), exercise capacity (6 minute walk) and bronchoscopic airway characteristics (anastomotic healing, patency and secretions) were assessed at 1, 2 and 3 months post LT. Patient adherence and satisfaction were measured. Results Of 60 consecutive patients, 36 (18 in each group) were recruited and completed the study. Both groups improved lung function (FEV 1 72 ± 4% to 81 ± 4 % p<0.0001; FVC 69 ± 3% to 81 ± 3% p<0.0001), Brasfield scores (17.8 ± 0.5 to 19.8 ± 0.5 p<0.002) and 6 minute walk (451± 16m to 545 ± 16m p<0.0001) over the study period. No significant differences between groups for any outcome were found. The vast majority had fully healed, 100% patent anastomoses without secretions at 3 months. There were no significant differences between airway characteristics and incidence of chest infection. Adherence to both strategies was high (84% proactive, 100% reactive). Conclusion In the absence of significant differences in outcomes, it is recommended that AC only be performed in the presence of chest infection based on greater treatment cost and treatment time required by a proactive strategy. Further studies in those with reduced anastomotic patency and in those with recurrent chest infections later post-transplant are warranted. Supported by an Alfred Trusts Grant.

Munro, P.E. 1 ; Button, B. 1, 2 ; Bailey, M. 3 ; Holland, A. 1, 4 ; Snell, G. 2 1. Physiotherapy, The Alfred, Prahran, VIC, Australia; 2. Allergy, Immunology and Respiratory Medicine, The Alfred, Melbourne, VIC, Australia; 3. Epidemiology and Preventative Medicine, Monash University, Melbourne, VIC, Australia; 4. School of Physiotherapy, La Trobe University, Melbourne, VIC, Australia Background The optimal duration and structure of pulmonary rehabilitation (PR) for lung transplant (LT) recipients is not known. This study aimed to describe changes in functional outcomes in LT recipients who participated in a post LT PR program at The Alfred, Melbourne, Australia. Methods Prospective, repeated measures design. Functional exercise capacity (six minute walk test-6MWT), lung function (FEV 1 , FVC) and quality of life (SF 36) were assessed at 1, 2 and 3 months following LT. Following discharge, all subjects attended a 1 hour supervised outpatient exercise training class 3 days per week until 12 weeks post LT and 6 education sessions. Patients with post-operative complications were excluded. Data were analysed using descriptive statistics and ANOVA with repeated measures. Results 60 consecutive LT recipients from Sept 2003 to Mar 2005 were assessed for inclusion. 36 (50% male) subjects, mean age 46 ± 14 yrs were recruited and completed the study. 81% had undergone bilateral LT. 33% had cystic fibrosis, 31% chronic obstructive pulmonary disease. Significant improvements were demonstrated in 6MWT (451 ± 16m to 545 ± 16m, p<0.001), FEV 1 (72 ± 4 % to 81 ± 4%, p<0.0001), FVC (74 ± 4% to 78 ± 4%, p<0.0001) and all quality of life domains, p<0.05. The greatest changes in 6MWT and lung function were seen between 1 and 2 months. Conclusion Functional exercise capacity, lung function and quality of life improved significantly over the first 3 months in LT recipients who participated in PR at our institution. These data will allow benchmarking with other centres and program structures. Supported by an Alfred Trusts Grant. Does gender play a role in perception of CF quality of life domains before and after lung transplantation? The CF quality of life (CFQ-R) for patients м14 years of age is a disease-specific instrument. Previous studies using this instrument have reported that CF females have higher scores in weight and body perception than CF males. To determine the impact of lung transplantation on gender-based perception of quality of life, we reviewed the CFQ-R on all CF patients м14 years of age and who underwent successful lung transplantation at our center. CFQ-R was administered 1-3 months prior to transplant surgery and then 3 months after lung transplantation while the patients were in the outpatient clinic setting. Domains measured: Physical, Role, Vitality, Emotion, Social, Body Image, Eating, Treatment Burden, Health Status, Weight, Respiratory Symptoms, and Digestion. The CFQ-R responses were computer-scored and Domain scores were then generated (score 0-100, 100=better). During the 2 1 ⁄2 year study period, 14 CF patients underwent successful lung transplantation (5 Males:9 Females; mean age 16.1 ± 2.5 years). Prior to lung transplantation, female CF patients had lower Physical Domain scores but higher scores in Health Status perceptions than the male CF lung transplant candidates (see graph). All patients had marked improvement in all domains 3 months after lung transplantation (see graph). After transplant surgery, CF males had higher scores than CF females in Emotion and in Eating Domains. Scores for Body Image Domain were similar for both genders before and 3-months after lung transplantation. We conclude that although there were differences in Physical and Health Domain perceptions prior to lung transplantation, both male and female pediatric CF patients had marked improvement in all domains after lung transplantation. However, female CF lung transplant recipients had lower Emotion and Eating Domain scores after transplantation. We speculate that lower scores in these domains may reflect increased post-traumatic stress and depression in female lung transplant recipients. Further study on CF Quality of Life in pediatric lung transplant recipients is on-going.

Background CF liver disease (CFLD) with severe cirrhosis develops in 5-10% of patients, usually within the first 2 decades of life. Most patients with CFLD suffer from complications of portal hypertension but hepatocellular function usually remains intact for many years, even decades. Patients with CFLD referred for lung transplantation (Tx) may be offered lung Tx, lung-liver Tx or be excluded from lung Tx. However, variceal hemorrhage due to portal hypertension can be managed by banding, sclerotherapy or shunt procedures without liver Tx. The aim of this study was to examine outcome of CFLD patients who underwent lung Tx and the subsequent progression of CFLD, particularly with the use of potentially hepatotoxic immunosuppressive drugs.

We conducted a retrospective cohort study to compare CF patients undergoing lung Tx with and without CFLD. B. cepacia-negative CF patients undergoing lung Tx at Toronto General Hospital from 1987-2007 were eligible for inclusion. Liver cirrhosis was defined by histology, esophageal varices on endoscopy or imaging evidence of splenomegaly. Each patient with CFLD was matched for age and year of Tx with 3 CF patients without CFLD. Demographic data, survival and liver function tests (LFTs)(AST, ALT, ALP, INR, albumin, protein at week 1, month 1, 2, 3, 6, 9, 12, 24, 36, 48 and 60 post Tx) were obtained by chart review. Results of the 2 groups were compared using unpaired t-test (parametric data) or Mann-Whitney U test (non-parametric data).

Results 110 B. cepacia-negative CF patients underwent lung Tx over this period, 6 (5.5%) having CFLD with cirrhosis. Data from the 6 CFLD patients vs. 18 matched CF patients without CFLD were analyzed. No significant difference was found between the groups in pre-Tx age, gender, BMI, PI status, CFRD status, FEV1% predicted, 6-minute walk distance or LFTs. In the CFLD group albumin was lower 1 week post-Tx (18.2+1.7 g/L vs. 22.2+3.9 g/L, p<0.026), ALP was higher 1 month post lung Tx (389.5 vs. 137.9, p<0 .033) and ALT was higher 60 months post-Tx (26.7+8.3 u/L vs. 16.0+4.5 u/L, p<0.026). There was no significant difference in the other blood tests. Azathioprine was changed to mycophenolate mofetil in 2 CFLD patients due to liver test abnormalities. 1 CFLD patient developed hepatic encephalopathy and ascites 4 years post lung Tx and is being assessed for liver Tx. No difference was seen in the number of episodes of acute rejection or post-Tx survival (66.7% vs. 61.1% alive at 5 years).

Discussion CFLD patients undergoing lung Tx did not have a worse prognosis than patients without CFLD. Transient elevation of LFTs were seen in the immediate post-Tx period but settled either spontaneously or after discontinuation of azathioprine. We conclude that selected patients with cirrhosis due to CFLD may be safely offered lung Tx without concomitant liver Tx. Infectious mononucleosis, commonly called mononucleosis, or "mono," is an illness caused by the Epstein-Barr virus. Mononucleosis has been nicknamed the "kissing disease" because the Epstein-Barr virus commonly is transmitted in saliva during kissing. However, sneezes and coughs also can transmit the virus occasionally. The Epstein-Barr virus (EBV) permanently infects more than 90% of the people on Earth, but it causes symptomatic mononucleosis only in a small number. In developed nations, such as the United States, mononucleosis most often occurs between the ages of 15 and 25. Unfortunately EBV can not only infect, but can also transform, B cells after transplant and may lead to lymphoma with a resulting ~50% mortality rate. The source of EBV is presumed to be passenger lymphocytes in the donor tissue and those at greatest risk are patients who are EBV naive before transplant. Since most CF patients with endstage lung disease in the developed world are now being referred for transplant given the success of this intervention, we reviewed our entire lung transplant database to determine the nature of EBV infection and the likelihood of developing lymphoma following lung transplantation. Of 272 first-time lung transplants, 178 (65%) had CF. EBV serology was not available in 27 (15 CF, 12 others) before transplant. 34 of 163 (21%) CF patients were EBV seronegative before transplant as compared to 4 of 82 (5%) patients with other end stage lung diseases (composed mostly of COPD and IPF) in the control population (Chi Square p < 0.001). The CF EBV seronegative cohort was 24 (SD 9) years old on average, an age when most seroconversion has already occurred in our society. Of the CF EBV seronegatives, 27 survived beyond 3 months and were thus at risk for post transplant lymphoma. Of these 27, 8 developed lymphoma (incidence = 30%) in comparison to 1 of the 4 EBV seronegative controls (incidence = 25%, p = NS). All lymphomas were stage IV at presentation and the majority (9 of 11) arose in the first post transplant year. The CF EBV seronegatives who developed lymphoma did not differ from those who did not with regard to levels of immunosuppression (cyclosporine, methylprednisolone or prednisone) or doses of antiviral therapy (i.e., ganciclovir which is active against both CMV and EBV). In addition, the cessation of anti-lymphocyte antibody induction therapy in the summer of 1995, somewhat surprisingly, had no impact on the development of lymphoma. Two cases of lymphoma developed in CF EBV seropositives (incidence = 1.2%) and no cases of lymphoma developed in the control EBV seropositives (p = NS). Thus EBV seronegativity raises the lymphoma risk 25 fold in CF patients (p < 0.001). Two of the 11 (18%) cases of lymphoma resulted in death (one CF and one other) and one CF patient probably died of complications related to chemotherapy even though he was tumor free at that time. Lymphoma outcome was not stage or clonality dependent. In conclusion, CF lung transplant recipients are at an increased risk for lymphoma largely due to their lack of exposure to EBV before transplant. Kissing more before transplant may lower the risk of lymphoma afterward. University Hospital UZB, Brussels, Belgium; 2. Chest Medicine, Erasme University Hospital, Brussels, Belgium; 3. Dermatology, Erasme University Hospital, Brussels, Belgium; 4. Gynaecology, Erasme University Hospital, Brussels, Belgium; 5. Gastroenterology, Erasme University Hospital, Brussels, Belgium; 6. Pathology, Erasme University Hospital, Brussels, Belgium HPV infection is an underestimated phenomenon in organ transplantation (tx) recipients, being unable during immunosuppression (IS) to mount an adequate anti-viral immune response and risking genital/anal warts (condylomata acuminata) as well as pre-cancerous/cancerous lesions. We retrospectively assessed the incidence/treatment of genital/anal HPV-associated lesions in cystic fibrosis (CF) lung tx recipients. The files of all 74 tx patients transplanted in the ULB center between 1988 and 01/2007 (38 men/36 women, median age 24±7.5, range 8-45 yrs) were reviewed. Maintenance IS consisted in chronic triple therapy combining a calcineurin inhibitor, a cell cycle inhibitor and corticosteroids. Median survival was 46.5 ± 44.4 months (range 0-186). Ten of 60 sexually active patients (17%) who survived ≥12 months post-tx developed HPV-associated genital/anal lesions: 3 men (35-41 yrs) and 7 women (25-41 yrs). Genital/anal HPV-PCR proven condylomata were diagnosed between 62-97 months in the men and between 14-90 months post-tx in the women. All 10 underwent local treatment (cryotherapy ± laser ± topical imiquimod), 3/10 patients underwent multiple treatments under general anesthesia. Recurrence of the condylomata was high. One male patient presented high-grade anal dysplasia and 3 women moderate to high-grade cervical dysplasia; 2 underwent conization and one complete hysterectomy after 2 conizations. One of the females presented both condylomata and a high grade cervical dysplasia.

These retrospective data in a CF population indicate that 1) HPV infection may cause significant morbidity in young subjects with chronic disease needing organ tx. It may even compromise life expectancy after tx. 2) Data on the effectiveness of treatment strategies -including topical immunomodulators -have to be collected 3) The potentially protective effects of the now available HPV vaccination in females and males? with chronic disease at risk for future organ tx, when administrated before sexual activity and before tx, should be rapidly evaluated in a multi-center effort.

Ballmann, M. 1 ; Pfister, E. 2 ; Schlueter, K. 1 ; Becker, T. 3 ; Melter, M. 2 ; Junge, S. 1 1. Pediatric pulmonology and neonatology, Medical School, Hannover, Germany; 2. Pediatric nephrology and hepatology, Medical School, Hannover, Germany; 3. Visceral and transplant surgury, Medical School, Hannover, Germany Liver disease is an important comortality and comorbidity in CF patients even in the pediatric age group. Liver transplantation (LTX) is an accepted option in end stage liver disease. Nevertheless the clinical outcome in CF patients is still under discussion. Here we reported the clinical outcome 12 months after isolated liver Tx in pediatric CF-patients.Since 2000 n=9 LTX were performed in children (male n=7, female n=2; age (mean±SD) 14,1±3 years) . Immunosuppressive therapy was done with ciclosporin ± basilixumab and steroid. We followed pulmonary function, CN-score, nutritional status (BMI Z-score) and body com-position, airway cultures and IgG.Results:All were still alive 12 months after LTX. Before LTX all patients had a mild to moderate pulmonary disease: FVC(%pred) 92,2±16 (71,4%-119), FEV1(%pred) 88,7±16,1 (68-103), MEF25(%pred) 61, 7± 28,5 (35-114) , CN-Score 8±4,3 points. Nutritional status: BMI(mean) 17,2±1,8% (range: 15,3%-20,3%, Z-score: 0,88±0,86), upper arm circumference(cm)(mean±SD) 17,8±0,9, triceps skin fat fold(cm) 5,2±1,2. Body composition: 13,5±7,2% of bodyweight were fat. One year after LTX all patients were still in a stable pulmonary situation: FVC(%pred) 93,9±14 (range:75,5%-114%), FEV1(%pred) 88,6±12,9 (range:68%-108%), MEF25(%pred) 61,6±30,4 (range:24,7%-106%), CN-Score 7,5±4,8 points. There weren't any significant changes in the airway microbiology under immunosuppressive therapy (steroid, Ciclosporin±Basiliximab), the serum IgG levels declined significant from 17,4±3,9g/l to 9,8±2,3/l(p<0.05). The growth over the year was 6,9±5 cm and the increase of weight come to 5,1±6,1kg, while the BMI didn't increase in this first year after transplantation. Certainly we found an increase of the body fat mass (to 18,2±3,8% of body weight), of the upper arm circumference (to 20±4,2cm) und of the triceps skin fat fold (to 11,4± 8,1cm) .Conclusion:Liver transplantation is an effective therapy even in children with CF related liver disease and stabilise pulmonary function and improves nutritional status in patients with CF and mild or moderate pulmonary involvement before LTX. (Neglia JP et al., N Engl J Med 1995; 332:494-499) , and colon cancer has been described in young adults with CF (Chaun H et al., Can J Gastroenterol 1996; 10:440-442) . Three of 62 lung transplant (LTX) recipients from our center developed colon cancer after successful bilateral lung transplant for end-stage lung disease. Data from these 3 individuals are given in Table below .

An additional 8 recipients with cystic fibrosis were screened via colonoscopy. Colonic polyps were detected in 2 and included lesions up to 3 cm in diameter. Five years after a colonoscopy that had shown a 4 mm polyp, one patient underwent a 2nd colonoscopy that revealed multiple polyps, including a 3 cm diameter sessile polyp in the sigmoid colon.

Colonic malignancies appear to arise from mucosal polyps, and screening via colonoscopy can detect and remove premalignant lesions. Lung transplant recipients have an increased risk of GI malignancies and tumor surveillance is impaired by post-transplant immunosuppression, colonoscopic screening should be considered for transplant recipients with cystic fibrosis. Additionally, colonic malignancy should be considered as a potential cause of unexplained gastrointestinal symptoms such as constipation.

Nossent, G. 1 ; Kastelijn, E. 1 ; Teding van Berkhout, F. 1 ; Zanen, P. 1 ; van den Bosch, J. 2 ; van de Graaf, E. 1 1. Respiratory Dis, University medical centre, Utrecht, Netherlands; 2. Respiratory Diseases, St. Antonius Hospital, Nieuwegein, Netherlands Objective: In the Netherlands lung allocation is based on waiting time. To prioritize patients with a poor prognosis and to reduce waiting list mortality the High Urgency (HU) status was introduced in 2001. To be considered for HU status there has to be a limited life expectancy on clinical judgement and all three transplantation centers have to audit this decision. The aim of this study is to analyze the potential effect of LAS on patients with Cystic Fibrosis (CF) who became HU or died on the waiting list. It would be expected that patients who became HU had a higher LAS.

Methods: From 2001 till 2007 in all CF patients placed on the HU waiting list, the median LAS (25th, 75th) was calculated at the moment of listing and at the moment the patients became HU. Also for the CF patients who died on the waiting list the mean LAS (sd) was calculated at the moment of listing.

Results: From the 187 patients on the waiting list 52 patients had CF. Of these patient 19 were placed on HU list and 4 died on the waiting list. In the CF patients that became HU the median LAS on the moment of listing was 35. 1 (33.9;35.9) . At the moment they were placed on the HU waiting list the median LAS was 35. 5 (34.8;37.4)(p= 0.2) . The median LAS of the 4 patients who died on the waitinglist was 35.3 (2.77) . This LAS did not differ significantly from the LAS of listing of the patients that became HU (p= 0.2).

Conclusion: The LAS does not detect the deterioration in patients with CF as diagnosed by clinical judgement. This may be due to the exclusion of specific prognostic factors of CF in the calculation of the LAS. Besides that, the LAS does not differentiate CF patients on the waiting list with a high chance of mortality. Introduction: LuTx recipients have one of the highest rates of IA in solid organ transplantation causing high morbidity and mortality rates. Nowadays antifungal prophylaxis is extensively used but clinical trials are rare.

Methods: All patients who received LuTx in our hospital were included in this retrospective study. Aspergillus airway colonization was defined as Aspergillus cultured twice from airway specimens in the absence of IA or tracheobronchitis (TB). Definition of IA and TB was according to literature. After the death of 2 CF-patients due to IA, a prophylaxis protocol was introduced containing avoidance of use of the cell saver during the operation in preTx colonized patients to avoid hematological fungal spreading and targeted prophylaxis with itraconazole or voriconazole in all pre-and postTx colonized patients during 3 months and inhaled amphotericin during hospital admission.

Results: A total of 93 LuTx were performed (in 92 patients) (35% CF patients) from 15-7-2001 until 31-12-2006. 23 patients (24.5%) were colonized with Aspergillus preTx and 16 patients postTx. Only 4 patients were colonized before and after Tx. Before introduction of the protocol 4 patients (4/69=5.7%) (50% CF patients) developed an Aspergillus infection: 3 (4.3%) patients developed A. fumigatus TB and 1 (1.4%) patient suffered from IA: cerebral A. fumigatus infection (proven IA). Two patients, both CF patients, (2/4=50%) died due toAspergillus infection (cerebral and TB).After introduction of the protocol only one A. fumigatus TB occurred (1/24= 4.1%) but no invasive fungal infection.

Conclusions: The rate of fungal infections in our LuTx patients was comparable with data from literature. Since introduction of azole prophylaxis protocol and avoidance of use of cellsaver no IA occurred.

Fischer, A. 1 ; Müllinger, B. 2 ; Arendt, T. 1 ; Bernhardt, T. 3 ; Hannah, K. 4 ; Scheuch, G. 1 1. Activaero GmbH, Gemuenden, Germany; 2. Biological Products, Bayer Healthcare, Leverkusen, Germany; 3. Activaero GmbH, Gauting, Germany; 4. Talecris Biotherapeutics, Research Triangle Park, NC, USA Background Patient compliance during a study is an important factor in view of assessing the clinical effect of a treatment. This is especially true when patients administer the drug at home. Usually, patient records, counting of returned doses or mechanical counters are used to track compliance, which may be biased by the study subjects. This report is about an aerosol study in CF which used a device for controlled breathing (AKITA® inhalation system, Activaero, Germany). The device works with a patient-individual smart card that records every single breath during a treatment including a date/time stamp in an encrypted manner. Each patient's inhalation protocol can be displayed by loading the smart card into the proprietary "Compliance Manager" software.

Methods For this report, the data set of a recent controlled study (Griese et al., 2007) was analyzed. 72 CF-Patients were instructed to inhale with the AKITA inhalation system at home for 42 days. After the study, the smart cards were returned to Activaero, and the inhalation records on the chip were analysed. We analysed the compliance of patients who participated at least 21 days (59 out of 72 patients, others deemed to be drop-outs . Most of the patients showed a DSC which is lower than the TDC, indicating that they had missing treatment days, which were compensated by additional inhalations on other days. We found 9 patients with a DSC more than 10% lower than their TDC (max difference: 23.81%).

Conclusion This aerosol study with home treatment demonstrated a high compliance rate for TDC and DSC. This information is more reliable for compliance than study participation in days alone. For future studies it is recommended to define in the protocol not only a minimum of study participation in days but also a minimum in TDC and DSC. In addition the compliance thresholds may be defined with regard to the drug's pharmacokinetic profile. In general, the inhalation record of the AKITA smart card provides an unbiased view of the inhalations treatments during a study especially when the subjects perform the inhalations at home. Compliance calculations as shown above may be performed and linked to other outcomes of a study for validation. Compliance data like these may also be used routinely by the treating physician in order to guide and supervise his patients. Introduction: Adaptive Aerosol Delivery (the I-neb ® AAD ® System) has been developed to deliver precise, reproducible doses of aerosol into inhalation. I-neb utilizes a medication chamber, which incorporates a metering chamber to deliver a preset (metered) dose of aerosol. Inhalation of hypertonic (7%) saline has been shown to aid mucociliary clearance in patients with cystic fibrosis. [1] We performed an in vitro test to determine the emitted, delivered and residual doses of hypertonic saline from the I-neb AAD System with 0.5 mL metering chamber, and a conventional jet nebulizer (LC Plus ® ) with a 4 mL fill. The results were then entered into a model to predict the likely lung dose from the in vitro tests, in order to determine the equivalent I-neb dose to the 4 mL conventional jet nebulizer fill.

Methods:An I-neb device configured to operate at power level 10 (of 15) was fitted with a metering chamber designed to deliver a preset dose of 0.5 mL. The I-neb was weighed, loaded with 1 mL of 7% saline and run on the CEN simulated breathing pattern. This process was repeated using an LC Plus nebulizer loaded with 4 mL of 7% saline. Aerosol was collected on a filter placed between a Harvard pump and the device. Emitted dose and residual dose were determined by gravimetric output, and delivered dose was determined by ion analysis. All tests were conducted in triplicate.

Results: As shown in table 1, the emitted dose approximated the delivered dose for I-neb, whereas for LC Plus the emitted dose was approximately twice the delivered dose, due to the wastage of aerosol upon simulated exhalation. As I-neb has a delivered dose of 513 mg, and I-neb has been shown to deposit 62.8% of the dose in the lung, [2] this would result in a lung dose of 322 mg. The Pari LC Plus has a higher delivered dose, but the overall lung deposition is only 47% of the fill volume, i.e. 613 mg. [3] Conclusion: The results of this test suggest that two treatments would be required in order to deliver an equivalent lung dose of hypertonic saline from an I-neb device fitted with a 0.5 mL metering chamber, compared with a 4 mL fill in an LC Plus jet nebulizer.

References Introduction: Inhaled colistimethate sodium can be used for the eradication of Pseudomonas aeruginosa in patients with cystic fibrosis. The tonicity of colistimethate sodium inhalation solution has been linked to the occurrence of bronchospasm in some patients. It has been suggested that this bronchospasm can be avoided by using an isotonic solution of colistimethate sodium. [1] This can be complicated when using conventional jet nebulizers, since evaporation during nebulization tends to change the tonicity of the solution remaining in the nebulizer. However, because very little evaporation occurs within the I-neb medication chamber during nebulization, the tonicity of the loaded solution at the beginning and end of nebulization remains almost constant.

Aim: We investigated the effects of various saline concentrations upon the tonicity of two concentrations of colistimethate sodium solution (0.5 MIU/mL and 1 MIU/mL).

Methods: Diluent containing sodium chloride concentrations of: 0% (water for injection), 0.225%, 0.45%, 0.675%, and 0.9% (normal saline) were used to reconstitute vials of colistimethate sodium (Promixin ® , Profile Pharma, UK). Each vial was reconstituted with either 1 or 2 mL of diluent to make up 1MIU/mL or 0.5MIU/mL colistimethate sodium solution, respectively. The tonicity of each diluent concentration/colistimethate sodium concentration combination was made up and tested in triplicate using an Advanced Micro Osmometer (Advanced Instruments, Norwood, USA). The results were plotted on a graph along with a line of best fit, to determine the diluent concentration that gave a mean tonicity closest to that of an isotonic solution.

Results: Tonicity increased linearly with increasing concentrations of saline solution for both 1MIU/mL and 0.5MIU/mL concentrations. The tonicity of the 1MIU/mL solution was approximately 116 mOsm higher than the 0.5MIU/mL solution across the range of saline concentrations tested. The best fit line passed through the isotonic solution point (290 mOsm) closest to the 0.45% saline concentration point.

Conclusions: The diluent that produces the closest approximation of isotonicity for the mean of 0.5MIU/mL and 1MIU/mL colistimethate sodium concentrations was 0.45% saline.

References 1. Transave, Inc., Monmouth Junction, NJ, USA; 2. Activaero, GmbH, Gauting, Germany Arikace™ is an inhalation formulation of a liposomal amikacin suspension that is designed to treat chronic Pseudomonas aeruginosa infections in cystic fibrosis patients. Liposome encapsulation of amikacin reduces nonspecific binding of this cationic aminoglycoside drug to the negatively charged mucus and biofilm surfaces and allows penetration and delivery of packets of highly concentrated drug to the otherwise protected bacteria. As nebulized and delivered to the lungs, Arikace™ comprises 65% liposomal amikacin and 35% 'free' amikacin that is not entrapped by liposomes; free drug is produced by liposome leakage during nebulization. This profile provides an initial high peak concentration of amikacin followed by a sustained level as drug leaks from the liposomes. Nebulized Arikace™ with this profile was evaluated previously in human clinical studies using the LC STAR® nebulizer. Although the 0.3 µm liposomes of Arikace™ are efficiently loaded with drug, the expected effective dose in humans will require relatively large volumes of drug solution to be administered. To minimize treatment time and improve patient convenience we sought to find a nebulizer that would efficiently deposit drug and produce aerosol at a high output rate, while still producing the same level of free drug during nebulization.

Methods: Nebulization of liposomal amikacin was compared using several nebulizers, including 2 jet nebulizers: LC STAR® and LC SPRINT® (both from PARI), and 3 electronic mesh nebulizers: MicroAir® NE U22V (Omron), AeronebGo® (Nektar), and eFlow® (PARI). Additionally, meshes of different pore sizes were examined for the MicroAir® and eFlow® devices. Output rates were measured by gravimetric differences. Droplet size distribution was assessed by a laser light scattering method and by cascade impaction; mass median aerodynamic diameters (MMAD), geometric standard deviation (GSD), and fine particle fraction (FPF) (i.e., % droplet mass <5 µm). Amikacin association with the liposomes was measured using a centrifugation-filtration assay.

Results: In terms of output rate, the order of performance was eFlow® (~0.5 g/min) > LC SPRINT® > LC STAR® > AeronebGo® > MicroAir®. While the AeronebGo® and MicroAir® mesh nebulizers performed well with saline, their performance declined dramatically when nebulizing the liposomal amikacin solution, and, in fact, the MicroAir device clogged. In additional studies with the eFlow® where different meshes were compared, the 40L mesh was selected as it provided ideal aerosol properties (MMAD = 3.7 µm, GSD = 1.7, FPF >70%) with a ratio of entrapped to free amikacin of approximately 65%/35%. Importantly, from in vitro breath simulation studies it is estimated that about twice of the nominal drug dose is expected to deposit in the lungs compared to the LC STAR®.

Conclusions: For nebulization of liposomal amikacin, the eFlow® configured with a 40L mesh provided optimum aerosol characteristics with a high output rate and ideal size for distribution throughout the lung. The eFlow® reduces dose administration time not only because of its faster output rate but also because its higher deposition factor greatly reduces the amount of drug needed.

Zeman, K.L.; Bennett, W.D. CEMALB, University of North Carolina, Chapel Hill, NC, USA Many inhaled, therapeutic drugs have their site of action in the conducting airways, but may deposit elsewhere in the respiratory tract, resulting in unwanted side effects and waste. The quantity and location of particle deposition in the respiratory tract depends on both the particles' aerodynamic size and breathing pattern. In 8 subjects with normal lung function, we evaluated three methodologies for their ability to deliver radiolabeled particles (99mTc-labeled sulfur colloid) to the conducting airways: 1. Inhalation of 5 µm MMAD particles (DeVilbiss 646 jet nebulizer) using a resting tidal volume and flow (mean 0.43 L and 0.39 Lps); 2. Inhalation from the same nebulizer using a very large inspiratory capacity volume and higher flow rate (mean 2.6 L and 0.52 Lps); and 3. Inhalation of larger 9.5 µm MMAD (HEART jet nebulizer) particles with a large tidal volume and very low flow rate (mean 0.89 L and 0.08 Lps). Gamma scintigraphy gave an estimate of % conducting airway deposition (%CAD) by multiplying the percent of activity in the lungs immediately post-deposition relative to the total deposition (i.e. lung+mouth +esophagus+stomach) times the percent of activity cleared from the lungs over 24 hours. %CAD was 17%, 14%, and 31% for the three methods, respectively, significantly greater for the large particle, very slow inhalation technique when compared to the other two methods. The amount deposited in the mouth and larynx was 52%, 34 and 33%.

In addition, we evaluated protocols 1 and 3 above in a preliminary group of 3 CF patients. For protocol 1, mean inhaled volume and rate was 0.48 L and 0.44 Lps. Mean inhaled volume and flow were 0.73 L and 0.078 Lps for protocol 3. The %CAD for the two protocols 1 and 3 were 16% and 25%, respectively. The amount deposited in the mouth and larynx was 41% and 33%.

Higher therapeutic value of a medication delivered to the airways where the primary defect is associated with CF disease, and with fewer losses to the extrathoracic airways, can be obtained by using a "very large particle/slow inhaled flow" routine when compared to normal or higher flow breathing associated with typical nebulizers.

Supported by CF Foundation. Positive effects of physical conditioning in cystic fibrosis (CF) have been reported for quality of life (QoL) in addition to effectis on fitness and lung functions. The objective of this study was to identify the effects of different modes of training on QoL in CF. 58 CF-patients (age 12-40 yr., FEV1 >35%pred.) were randomized into 5 groups: Germany: unsupervised homebased training, 3 hours per week, free choice of training mode and means (UT, n=18) and controls (CON1, n=10); Switzerland: aerobic training (AT, n=11) or weight training (WT, n=11) in a fitness center, 3 x 30 min per week, or controls (CON2, n=8). Subjects in the training groups were asked to train for 6 months. At study entry and after 3 and 6 months, QoL was assessed by questionnaire and maximal physical working capacity (PWC, %pred) was determined during cycle ergometry (Godfrey protocol). Changes from baseline values were calculated and ANOVA for repeated measures was used to test for differences among groups. Weight training resulted in a significant decrease in QoL (generic dimension) at 3 and 6 months compared with all other training modes including controls. There was no significant difference between AT or UT and controls in QoL during the program. However, there was a positive association between the change in PWC and the change in QoL (r=0.32, p<0.05) in the entire sample. In conclusion, physical training may not always result in an increase in QL in CF. Possibly, the decrease in QoL in the WT group may have resulted from tiredness induced by the training. Thus, a different outcome might have been found with less intense training. Positive effects may be achieved by improving aerobic fitness.

Supported Introduction: Exercise and exercise training programs are felt to be important to improve aerobic exercise capacity, muscle strength, lung function, and quality of life in CF-patients. Recent studies in children and adults with CF have demonstrated an increase of motor performance after exercise training. Less is known about trainability of motor performance in younger children with CF. The aim of this study was to assess the effects of a training program on motor performance in preschool children with CF.

Methods: 19 preschool children with CF (10 female and 9 male patients, mean age 5.2 yrs; range 4-6 yrs, FEV1%pred. 100,1±15.6 rage from 65.8-123.1 ) participated. At baseline (T1) and at the end of the training program (T2) the "Motorik Test (MOT)" was used to assess motor performance and to evaluate training effects. Components of motor performance tested with the MOT are balance, agility and flexibility, strength, coordination, fine motor skills, reactivity, accuracy of movement). Lung function in patients with CF was measured by spirometric techniques. The duration of training was 4-6 weeks (3 times/week, 45 minutes per session) with different kinds of sport activities.

Results: At T2, strength and balance had increased significantly (p<0.05). No other components of the MOT increased, but the increases in strength and balance were enough to cause a significant (p<0.05) increase overall motor performance, as indicated by an increase of 10.7% in MQ. The MQ is the sum of all 18 test-Items of the MOT. No correlation was found between lung function and motor performance.

Discussion: To our knowledge this was the first study that examined the effects of an exercise training program on motor performance in preschool children with CF. Main findings of this study were that a training program of 4-6 week of duration improved some aspects of motor performance. The improvements may be explained by different kinds of sports activities during training when compared with activities before training. Classification of motor performance of the tested children was normal (MQ =3.1±0.8). This could be the reason that only some aspects of motor performance increased. Parents of the children were asked about habitual physical activity at home. All children were physically active and some of the children participated in organized sports. A recent study showed that with an increase in age lower motor performance in children and adolescent with CF was seen compared to healthy children. We speculate that time spent in physical activity in preschool children with CF is comparable to healthy children.

Conclusion: An exercise program offered to preschool-children with CF may lead to an improvement in motor performance. Pre-school children with CF have normal motor performance. To prevent a decline of motor performance, exercise programs should be available for preschool children.

Gruber, W. 1 ; Braumann, K.M. 2 ; Orenstein, D.M. 3 ; Huels, G. 1 1. Dept. for Sports and Exercise, Clinic Sattelduene for children and adolescent, Nebel/Amrum, Germany; 2. Institute for Sport and Exercise Medicine, University of Hamburg, Hamburg, Germany; 3. School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA; 4. Dept. of Medicine, Clinic Sattelduene for children and adolescent, Nebel/Amrum, Germany Introduction: A decline in physical activity with age is one of the main problems in patients with CF. As a consequence, lower exercise capacity and a more rapid decline in lung function may be seen in those patients who are physical inactive. Furthermore, motor performance is lower in children and adolescents with CF compared to healthy children of the same age. Little attention has been paid to younger children with CF and motor performance. The aim of the study was to compare motor performance in preschool-children with CF with healthy children of the same age.

Methods: 19 children with CF (mean age 5.2 yrs; range 4-6 yrs) and 21 healthy children (mean age 5.5 yrs; range 4-6 yrs) participated. The "Motorik Test" (MOT) was used to assess motor performance. The MOT consisted of 18 test-items which evaluated seven different components of motor performance (balance, agility and flexibility, strength, coordination, fine motor skills, reactivity, accuracy of movement). Lung function in patients with CF was measured by spirometric techniques Results: Results of the MOT showed no difference either for single testitems or for parameter "Motorische Quotient (MQ)" (p>0.05). This parameter classified motor performance of all aspects tested with the MOT. Classification of children with CF and of healthy children was normal. Mean value for parameter MQ (CF: 3.1±0.8 vs. healthy 3.29±0.7) was slight higher in patients with CF than in healthy children (p>0.05). Differences between healthy and CF for different components of motor performance were not significant (p<0.05) Discussion: The MOT is a reliable and valid test method in Germany to determine motor performance in younger children. To our knowledge this was the first study to examine motor performance in preschool-children with CF. Furthermore, we compared patients with CF and healthy children. We observed no differences between groups for motor performance. In older children with CF, lower motor performance has previously been found compared to healthy children of the same age. We do not have any information about the level of habitual activity in tested healthy children. We therefore speculate that time spent in physical activity in CF is similar to that of healthy children at this age. A recent study showed a decline in physical activity in CF when age and problems during physical activity increased. We think that motor performance is normal in children with CF up to an age of 6 yrs. When age increased, time spent in physical activity decreases (treatment, school) and therefore motor performance decreases.

Conclusion: Motor performance is normal in preschool-children with CF compared to healthy children. To prevent a decline or to maintain motor performance it is recommended to motivate CF-children from early infancy to participate in sports activities or other physical activities to maintain or to increase their physical fitness. 1 1. Respiratory Medicine, The Hospital for Sick Children, Toronto, ON, Canada; 2. Respirology, St. Michael's Hospital, Toronto, ON, Canada; 3. Respiratory Medicine, Montreal Children's Hospital, Montreal, QC, Canada Purpose: The Habitual Activity Estimation Scale (HAES) questionnaire has been shown to be a feasible tool to measure physical activity however the reliability has yet to be determined in CF populations. We therefore assessed the reliability and validity of the HAES questionnaire in paediatric and adult cystic fibrosis (CF) populations.

Methods: Fourteen (7 male, 7 female) patients aged 16.2 + 4.2 yrs with cystic fibrosis from the CF clinics at the Hospital for Sick Children and St. Michael's Hospital participated in this study. Participants were clinically stable at the time of the study (FEV1 > 70% predicted) and participating in their 'habitual' physical activity. To assess test-retest reliability, patients completed each of the HAES (Schneiderman-Walker et al., J Pediatr. 2005; 147 (3):321-6.), and a validated 3-day activity diary (Bratteby et al., Eur J Clin Nutr. 1997; 51(9) :585-91), and wore an ActiGraphTM Tri-Axial Accelerometer for two consecutive weeks. Validity was assessed by comparing the activity results of each of the three instruments over a single week time period.

Results: Intra-class correlation coefficient estimates of reliability for the HAES, diary and accelerometer were 0.72 (p<0.0001), 0.76 (p<0.0001), 0.63 (p<0.0001), respectively. Validity was estimated by comparing the results from each instrument using a one-way ANOVA with block design. There were no overall differences among the participants' activity results as estimated by the HAES, diary and accelerometer. Furthermore, there were no significant differences between activity measures among the 3 instruments when broken into morning, afternoon, or evening periods, or between measures from weekday or weekend days. While there were no significant differences among the 3 instruments when recording 'very active' intensity levels, they did detect differences during the inactive / somewhat inactive (p = 0.02) and somewhat active (p = 0.001) intensity levels.

Conclusion: The findings of this study suggest that the HAES questionnaire is a reliable and valid instrument that can be used to assess activity levels in patients with cystic fibrosis. In agreement with others, our patients found it easiest to recall and record accurately those periods in which they are participating in vigorous activity.

Acknowledgement: This research was supported by the Canadian Cystic Fibrosis Foundation. 1 1. Respiratory Medicine, Hospital for Sick Children, Toronto, ON, Canada; 2. Respiratory Medicine, Montreal Children's Hospital, Montreal, QC, Canada; 3. Respirology, St. Michael's Hospital, Toronto, ON, Canada Objective: Previous work showed a relationship between habitual physical activity (HPA) and decline of lung function (FEV1) in CF females (Sneiderman-Walker, J Peds 147:321,2005) . This study only included a small number of patients and the follow-up period was limited to 18 months. The purpose of this study was to evaluate daily activity levels and decline in FEV1 in a larger cohort over a longer period of time.

Methods: A total of 187 (n= 99 female) patients with CF (age 7-25 yrs) were studied over a six year period at the Hospital for Sick Children, St. Michael's Hospital and Montreal Children's Hospital. At each quarterly CF clinic visit, patients performed pulmonary function tests to determine FEV1 and completed the Habitual Activity Estimation Scale (HAES). Weekday total activity (WDTA), expressed as hours/day (h/day), was calculated from the sum of 'very active' (WDVA) and 'somewhat active' (WDSA) categories as previously described. A repeated measures linear regression was performed to evaluate the relationship of FEV1 and HPA over time.

Results: At baseline, mean FEV1 was 85% predicted (range 29-119% pred). Mean baseline WDTA was 5.3 h/d (WDVA = 2.3 h/d and WDSA = 3.0 h/d) similar to that previously reported. Subjects were divided into high and low activity groups based on the group median WDTA. Those in the low activity group had a significantly (p=0.004) steeper rate of decline (-1.47 % predicted per year) compared to those in the high activity group (-1.06 % predicted per year) . Unlike in the in the previous study this difference was observed in males as well as females.

Conclusions: Six year follow-up of this patient cohort has shown that higher activity levels are clearly associated with a slower rate of decline of FEV1. We are currently evaluating whether strategies aimed at increasing physical activity will have an impact on lung function decline in CF patients. Background: Knowledge of exercise principles, benefits, and techniques has not been assessed in children with cystic fibrosis (CF) or their parents. Regular exercise is a recommended CF treatment regimen that is often not followed, but one which might improve cardiovascular fitness, lung function, sputum clearance, and health-related quality of life. Further, aerobic fitness, measured as peak oxygen uptake (VO 2 ), is positively related to survival in CF. Construction of a reliable and valid measure of exercise knowledge is necessary to assess level of knowledge and study its relationship to exercise behavior. The aim of this study was to develop and test a measure of exercise knowledge for pediatric patients with CF and their parents. Methods: We constructed a 22-item exercise knowledge test (EKT) and then examined its psychometric properties. A comprehensive review of the exercise literature guided the formulation of objectives and development of items. To establish content validity, a panel of experts (CF physician, two exercise physiologists, and a clinical CF nurse) evaluated the objectives and test items for relevance, clarity, and accuracy. After feedback from the panel, we revised the EKT through an iterative process. In consultation with a reading level expert, we wrote the EKT items so that the test could be administered to children and adults. A measurement and evaluation expert also reviewed the test, and additional revisions were made. During routine clinic visits, the EKT was administered to 50 subjects with CF and 50 parents. The CF subjects included 31 boys and 19 girls, age 8 to 18 years, FEV 1 range: 26-123 % predicted. The parents had an average age of 41 years and most (n=44) were female. Results: The EKT includes 22 items, 12 true-false and 10 multiple choice. It is self-administered, yields a single summed score, and has a Flesch-Kincaid 3rd grade reading level. The means were 17.96 + 2.7 (82% correct) and 20.24 + 1.4 (92% correct) for the CF subjects and parents, respectively; 74% of the parents scored between 20 and 22. Internal consistency reliability estimates were .64 for the children and .26 for the parents, reflecting both the multidimensionality of the material and the lack of variability in the parent scores. Item difficulty ranged from 0.50 to 0.98 for the CF subjects and from 0.64 to 1.00 for the parents. In both groups, the discrimination values were positive for all items with difficulty values less than 1.00. Conclusions: We were able to develop a measure of exercise knowledge with an acceptable reading level for children with CF and their parents. This sample of CF subjects and parents showed substantial exercise knowledge. The EKT can be a useful tool to identify gaps in exercise knowledge in CF patients and their parents, develop educational programs to increase knowledge, and examine how knowledge relates to exercise behavior in children with CF.

Supported by Cystic Fibrosis Foundation Therapeutics, Inc. Purpose: Although physical activity is routinely recommended for adolescents with cystic fibrosis, past research suggests that many patients do not regularly participate in physical activity. There may be a variety of reasons, both related to disease and otherwise, for this inconsistent participation. In this study, qualitative methods were used to describe factors that facilitated or served as barriers to physical activity in adolescents with CF and compared these factors to those in a healthy peer group.

Methods: Ten subjects with CF and ten subjects without CF aged 13-17 volunteered for this study. Each subject participated in two, open-ended semi-structured interviews with 3-4 weeks between each interview. Subjects were questioned about their current and past physical activity as well as the benefits and barriers to participation. The adolescents with CF were also asked about their perceptions of physical activity on their own health and other adolescents with CF. Interviews were transcribed and coded by three investigator groups. Analysis was conducted using the grounded theory approach at two points in the study, following each interview and again at the completion of all interviews. At the first point, individual interviews were coded using line-by-line coding, and similar information was grouped into categories. In the second interview, categories were confirmed and subjects were also asked to elaborate or clarify points from the first interview. Interviews were continued until themes reached a saturation point. Upon completion of all interviews, data for each group were organized by question. Data categories appearing throughout the interviews became major themes describing facilitators and barriers to exercise. The major themes were identified by the three researcher groups and then through collaborative group analysis.

Results: The central theme for both groups was perceived importance of physical activity for health benefits. For the CF group, facilitators and barriers were both psychological and physical. Physical benefits were categorized as either relating to their general health (such as feeling energized) or as benefiting their disease (such as improving breathing). In the non-CF group, facilitators included social factors in addition to general health-related and psychological benefits. In the non-CF group, barriers were categorized as either internal or external, with the internal barriers being similar to the CF group (physical and psychological), however, the non-CF participants also indicated external barriers (such as time) not articulated by the CF group.

Conclusions: In general, we found very similar facilitators and barriers to physical activity in adolescents with CF and those without. However, the adolescents with CF strongly indicated that physical activity would have positive effects on their physical health. Further work is needed to determine ways to accentuate facilitators and remove barriers to improve physical activity levels in adolescents with CF. This project was supported by a research grant from the Cardiovascular & Pulmonary Section, APTA.

Lloyd, E. 1 ; Dodd, M. 2 1. Paediatric Cystic Fibrosis Service, Central Manchester and Manchester Children's University Hospital, Manchester, United Kingdom; 2. Adult Cystic Fibrosis Centre, Wythenshawe Hospital, University Hospital of South Manchester NHS Foundation Trust, Manchester, United Kingdom Background: Training in the skills of communication and self care are considered important components of transitional care1. Physiotherapists work to involve families and young people to encourage this independence. The CF Manager Training Programme2 (CFMTP) is a recognised tool for measuring age appropriate tasks in CF care.

Aim: The aim of the study was to survey the current levels of autonomy in all patients with cystic fibrosis in the North West of England from birth to 18 years to provide a baseline measure for encouraging autonomy in preparation for transition.

Method: Three areas of the CFMTP were selected for the survey: physiotherapy (P), inhalation therapy ( IT) and communication (C) for the 10 age groups: 0-1. 5, 1.5-3, 3-5, 6-7, 7-8, 8-9, 9-10, 11-12, 13-15,and 16+ years. Physiotherapists from all the North West CF clinics were invited to assess all the patients under their care. Data collected were age, gender and whether the patients skills fell "below", "same" or "above" the age appropriate self management tasks as described in the CFMTP. 'Not applicable' was indicated where the skills did not apply.

Results: 14/16 clinics responded. 401out of a total of 474 patients in the NW were assessed, three patients were excluded. Population by clinic ranged from 6-133. Population by age category: 15 (0-1.5 years) 24 (1.5-3years) 45 (3-5 years) 25 (6-7 years) 36 (7-8 years) 16 (8-9 years) 45 (9-10 years) 50 (11-12 years) 78 (13-15 years and 64 (16+ years). 213/398 (54%) were male. For P (n=390), IT (n=261), C (n=367), 32%, 40%, 22% = below, 59%, 51%, 65% = same, 9%, 9%, 13% = above respectively. There were no gender differences.

Conclusions: Overall scores revealed that 2/3rd of patients were gaining age appropriate independence in all areas. There was the least autonomy with IT. It was rare for a child to be setting up and cleaning equipment regularly at 7-8years, but this continued throughout the adolescent years. Overall 35% were 'below' for P and IT; this was highest after the age of 11years. Autonomy with physiotherapy appeared slow to develop. Some categories in CFMTP were different from our practice for e.g. patients are not invited to be seen alone in our clinics until 13 years v 9-10 years and this may account for the lower scores in communication below this age. This study has highlighted that some of our patients are not fully independent in the time up to transfer. We need to encourage independence from an early age with all aspects of care and integrate the process into our routine clinical practice. Refs 1 McDonagh and Viner BMJ 2006; 332:435-6, 2. Parents' Guide to the CF Manager Training Programme. Children's Hospital of Eastern Ontario

Planner, S. 1 ; Morrison, L. 2 1. Physiotherapy, Gartnavel General Hospital, Glasgow, United Kingdom; 2 

Cystic Fibrosis (CF) causes a deterioration in lung function and exercise tolerance. Assessing exercise capacity is therefore essential in order to monitor changes in disease severity.

The 6-minute walk test (6MWT) and 3 minute step test (3MST) are validated and widely used outcome measures for identifying exercise capacity in patients with CF. Exercise testing is part of annual review of all patients in the West of Scotland. Previous studies identified that the 3MST was a sensitive measure of exercise tolerance, but did not successfully challenge patients with mild or moderate lung disease as defined by FEV(sub)1(/sub)>40% (Morrison 2005) . Further analysis of the 3MST, where step height was increased to 8 inches did not show any difference in SaO(sub)2(/sub). decline or rate of perceived exertion (RPE) in patients with mild disease.

The Chester

Step Test (CST) was developed to assess aerobic fitness which is used in the UK cardiac population for exercise prescription (Sykes 1999) . It is a 10 minute progressive, sub-maximal test with a variable step height (6-12inches) , making it suitable for those with a wide range of exercise capacities. In addition, CST results extrapolate to a METS value (oxygen utilisation: 1MET = rate of oxygen utilisation at rest) as an outcome measure which can facilitate exercise prescription based on established values for physical activities.

We aim to determine whether the CST is a more sensitive measure of exercise capacity than the 6MWT or 3MST when considering decrease in SaO(sub)2(/sub) and RPE.

13 patients ( aged 16-49 mean FEV(sub)1(/sub)2.39L) with mild-moderate CF performed the CST and 6MWT and results were correlated with FEV(sub)1(/sub). Decrease in SaO(sub)2(/sub), increase in HR and RPE scores were also compared between the CST, 6MWT and recent 3MST data. Patients were exercised to a symptom limited maximum which differed from the cardiac CST where patients are exercised to 80% of maximal HR.

The CST appears to be more challenging as only 4 patients completed it, all with an FEV(sub)1(/sub)of >50% predicted, whereas in the 3MST and the 6MWT all patients completed the test.

METS levels for the CST were weakly correlated with FEV(sub)1(/sub)(r=0.456) and the 6MWT (r=0.29) . This could be explained by the narrow range of METS data available or the size of the sample population. There were significantly greater rises in HR (p= 0.01, p= 0.01), decrease in SaO(sub)2(/sub) (p=0. 03, p=0.0001) and RPE (p=0.04, p=0.013) , between the CST, 3MST and 6MWT respectively, suggesting that the CST is a more sensitive measure of cardiorespiratory effort.

Conclusions

The CST appears to be a more challenging exercise test for those with mild to moderate disease and therefore may be a better reflection of exercise capabilities. CST results are useful for prescribing exercise and monitoring programmes appropriately for this patient group. Further study could determine whether the CST is also suitable for those with severe disease.

Lee, A.L. 1 ; Button, B. 1 ; Holdsworth, M. 1 ; Holland, A. 2 1. Physiotherapy, The Alfred, Melbourne, VIC, Australia; 2. LaTrobe University, Melbourne, VIC, Australia Musculoskeletal pain is prevalent in cystic fibrosis (CF). While the chest and back are frequently reported regions of acute or chronic pain, it is unclear if pain in these areas is related to disease severity or respiratory symptoms. Manual therapy and postural advice has been shown to alleviate pain, but the effects of a combined approach of musculoskeletal physiotherapy and massage has not been studied. The aim of this study was to identify the primary regions of musculoskeletal pain, the relationship between pain severity and respiratory symptoms and to examine the effect of musculoskeletal physiotherapy techniques and soft tissue therapy, including remedial massage on pain and ease of breathing (EOB) in adults with CF. One hundred and twenty-nine adults with CF (70 with acute exacerbation, 24 post lung transplant) aged 31 ± 9 years (mean ±SD) with FEV(sub)1(/sub) of 51± 21% participated in this study. Following assessment of primary pain regions, each subject underwent a single treatment session including spinal joint / intercostal mobilisation and soft tissue therapy. Pain and EOB on a visual analogue scale were measured before and after treatment. Changes were compared using paired-samples t-tests. Pain was frequently reported in the thoracic spine region (38% of subjects), followed by the shoulder region (31%), cervical spine region (16%) and chest wall region (9%). Equivalent ratings of pain and EOB prior to treatment were reported for all regions of pain. EOB rating prior to treatment were worse in those with low BMI (r=-0.21, p=0.02) and low FEV(sub)1(/sub) (r=-0.24, p=0.01). A single treatment session was associated with reduction in pain (95%CI 1.6 to 2.1, p<0.001) and improvement in EOB (95%CI 0.4 to 0.7, p<0.001), irrespective of clinical status. Improvement in pain was equivalent for all primary pain regions. However, greater improvement in EOB was evident in subjects with shoulder pain compared to other regions (95%CI -1.1 to -0.02, p=0.04). A combination of musculoskeletal physiotherapy techniques and soft tissue therapy reduces musculoskeletal pain and improves EOB in adults with stable CF, during an acute exacerbation and post lung transplantation.

Service Cystic fibrosis is characterized by an excessive production of airway secretions responsible for bronchial obstruction and recurrent infections of the respiratory tract. The transport by ciliary activity and cough may be dependent on the unsteady rheological properties of the respiratory mucus such as thixotropy and shear-thinning properties which correspond to a decrease of viscosity with time or flow rate, respectively. We have previously shown that respiratory mucus with high shear-thinning and thixotropic properties is better transported by the cough mechanism (Zahm et al, Eur. Respir. J., 1991) . To promote mucus clearance and to finally expectorate the mucus by a cough manoeuvre, an innovative device, the Airhelp, has been developed and aimed to improve chest physiotherapy effectiveness. During a prolonged expiratory manoeuvre, the Airhelp delivers negative pressure variations (maximal amplitude: -4 kPa) at an oscillatory frequency of 13 Hz. The aim of the present work was to test the effect of the Airhelp device on the in vitro clearance of mucus by airflow. A plexiglass tube of 9 mm in diameter and 210 mm in length has been connected by one part on the Airhelp and by the other part on a 30 l thank mimicking the lung compliance. A 100 µl drop of mucus simulant at different concentrations (viscogum at 0.25% or 0.5% with actigum at 0.5% to 1.75%) or of respiratory mucus, was deposited within the plastic tube and the distance travelled by the sample under a 5sec-airflow was measured. Twelve respiratory mucus samples collected from CF patients were analyzed and the viscosity of the samples was measured before and after the Airhelp experiment. We observed a significant (p < 0.03) and negative relationship (r = -0.77) between the viscosity of mucus simulants and oscillatory airflow transport: the lower the viscosity, the higher the mucus transport by oscillatory airflow. Using CF respiratory mucus samples, we observed that without the oscillatory airflow, no transport occurred, whereas when the samples were submitted to the oscillatory airflow, a significant increase in mucus transportability was measured. In addition, the viscosity of the CF respiratory mucus samples was significantly decreased after oscillatory airflow treatment (485 ± 148 Pa.s to 85 ± 55 Pa.s, p <0.02). These results demonstrate that the mucus transport efficiency by an oscillatory airflow is dependent on mucus viscosity and that the improvement in mucus transport is related to the thixotropic and shear thinning properties of the airway mucus. We conclude that the Airhelp device may improve chest physiotherapy effectiveness.

Potter, E. 2 ; Nufer, J. 2 ; Cullina, J. 2, 4 ; Quinton, H. 3 ; Jain, M. 1, 4 ; McColley, S.A. 1, 2 1. Northwestern University Feinberg School of Medicine, Chicago, IL, USA; 2. Children's Memorial Hospital, Chicago, IL, USA; Lebanon, NH, USA; 4. Northwestern Memorial Hospital, Chicago, IL, USA Socioeconomic status (SES) significantly impacts health outcomes in cystic fibrosis (CF). The Cystic Fibrosis Foundation (CFF) is currently using median income estimated by zip code from the 2000 US Census of the Population for SES risk adjustment of data reported publicly from the CFF Patient Registry. Median income estimated by zip code is an "ecologic" variable that may misclassify patient and family income. In order to more fully assess the impact of SES on CF health outcomes, the CFF began to ask Centers to collect data on household income directly from patients and families. These data were captured for 6.5% of patients and families in 2005 and 16.1% in 2006 in the national registry. We report a method of collecting SES data from children and adults implemented at the Children's Memorial Hospital and Northwestern University Cystic Fibrosis Center, and compare directly collected income to income based on zip code.

Methods: A short survey with the questions on household composition, income, and education level from the CFF Patient Registry was developed. A letter from the Center Director and Adult Program Director was mailed to patients and families informing them of the survey's purpose and confidentiality measures, i.e. responses would be seen only by the individual entering the data and identified by CFF ID number. The survey was distributed to CF patients and families in clinic by clerical staff. Completed surveys were placed in an envelope and sealed by the patient or family. Both directly collected income and income from zip code was classified into 9 groups (<$10K, 10-20K, 20-30K, 40-50K, 50-60K, 60-70K, 70-80K, 80-90K, and >90K) . Percent concordance and Spearman's rank correlation were calculated. 2005 data were used for patients who provided data in 2005; otherwise, 2006 data were used. Each patient's data was used only once in the analysis.

Results: 202 unique patients were identified, 127 pediatric and 75 adult patients. Eighteen percent of pediatric and 39% of adult patients preferred not to answer income questions or did not complete the survey. The mean income by zip code was $60 K for patients who did not report income data and $61K for the entire group. For the 150 responses, reported income data was moderately correlated with income by zip code (rho=0.48, p<0.001). Concordance between the 9 categories was 15%. The correlation was stronger for pediatric patients (rho=0.53, p<0.001) than for adults (rho=0.27, p=0.07). Re-classifying income as "low", "medium" and "high" (<$40K, $40-70K, >$70) led to a concordance of 40%.

Conclusion: Survey methods that protect patient and family confidentiality result in good return rates of full SES data from patients and families.

Zip code derived median income shows moderate concordance with reported income. Comparing the associations of the 2 forms of income data with important CF outcomes needs to be done in a larger data set.

We acknowledge the Cystic Fibrosis Foundation, Gerald O'Connor, Kathryn Sabadosa, and participating patients and families.

The Patient Registry provided a unique opportunity to assess changes in the epidemiology of respiratory pathogens in CF. While the incidence of P. aeruginosa remained stable, the prevalence decreased, suggestive of the impact of antibiotic eradication strategies. In contrast, the increasing incidence and prevalence of S. aureus and MRSA may reflect improved microbiological surveillance and laboratory techniques. The trends noted for B. cepacia complex may reflect successful implementation of infection control strategies. Future studies are needed to better define the association between these observed trends and improving care for CF patients.

VanderWyden, A. 1, 2 ; Drumm, M.L. 1, 3 ; Schluchter, M. 2 1. Pediatrics, Case Western Reserve University, Cleveland, OH, USA; 2. Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, OH, USA; 3. Genetics, Case Western Reserve University, Cleveland, OH, USA In the Genetic Modifier Study (GMS) of lung disease of cystic fibrosis (CF), patients, homozygous for the ∆F508 mutation, were classified as having either severe or mild lung disease, as defined by the lowest or highest quartile of forced expiratory volume in one second (FEV1) and a linear model was fit from these patients' longitudinal data (Drumm 2005 , Schluchter 2006 . While these models successfully dichotomized patients for the association study, we assessed whether a single slope model could be improved. Here we have begun to compare models with one, two and three slopes, each representing different age ranges, for their use in this type of association study. As an example, interleukin-8 (IL-8), a neutrophil chemoattractant, associates with the severe lung inflammation seen in CF patients. We have observed the TT genotype of the IL-8 SNP rs4073, correlated with severe pulmonary function in the GMS sample of 808 patients, relative to the AA and AT genotypes of this SNP. In a replicate study of 268 patients, with CFTR genotypes associated with exocrine insuffiency, longitudinal linear mixed models were used to further characterize the decline in lung function over time as associated with the IL-8 SNP, rs4073. Three linear regression models of FEV1% predicted of patients were fit over age and compared with likelihood ratio tests. The three models differed with respect to the number of nodes. The first model had a single slope while the second and the third were piecewise linear models, with nodes at age 15, and ages 15 and 25, respectively. Age 15 was chosen as the first node, as by this age most CF patients have reached puberty, an event believed to coincide with changes in pulmonary function. The second node, age 25, was arbitrarily chosen as a point in the third decade, as survivor effect becomes a significant issue during this period. When compared to the single slope and two slope models, the three slope model with nodes at age 15 and age 25 better explained the pulmonary function of this study (p<0.001, -2LogLikelihood Chi-Squared tests). When the explanatory variable, IL-8 rs4073 genotype, was added, the single slope model and the three slope model were again compared. The three slope with nodes at age 15 and age 25 again resulted in a better fit of FEV1% predicted over age as correlated with rs4073 genotype (p<0.001, -2Loglikelihood Chi-Squared test). Estimates of FEV1% predicted at ages 6, 15, 25, and 30 for each rs4073 genotype (AA, AT/TA, or TT) were determined using the three slope model (p<0.001 for all estimates). Other explanatory variables, like gender and survival (p<0.001), were added to further elucidate the relation between IL-8 genotype and lung disease phenotype. We propose to use this piecewise three slope model with nodes at age 15 and age 25 to look at the pulmonary function of patients in the CF Foundation registry data to better explain the decline in FEV1% predicted over time seen in CF patients. (Supported by HL-68890 and grants from the CF Foundation)

Stephenson, A.L. 1 Purpose: The purpose of this study is to determine the annual incidence of hospitalizations for individuals with CF living in Ontario and to examine predictors of hospitalization, specifically gender.

Methods: This is a retrospective cohort study from 1993 to 2002, using newly linked clinical and administrative databases. The Canadian CF Foundation Patient Data Registry (CPDR), contains detailed clinical information on all CF patients receiving care at one of 38 accredited CF centres across Canada. The Institute for Clinical Evaluative Sciences (ICES), holds linked administrative databases containing information on all publicly reimbursed health care services delivered in Ontario. Canadian Institute for Health Information (CIHI) Discharge Abstracts Database (DAD), Ontario Health Insurance Plan (OHIP), and Homecare databases were used for this study. The CPDR was probabilistically linked (Automatch software) to the ICES administrative database using name, gender and date of birth. All individuals were then linked to the CIHI DAD, OHIP, and Homecare databases using unique numeric identifiers. With these databases, longitudinal records for each individual were created. CIHI-DAD, OHIP, and Homecare claims data were used to calculate the number of hospitalizations per year for pulmonary infections. Males and females were analyzed separately to elucidate any gender disparities. Other predictor variables include age, geographical location, number of clinic visits, socioeconomic status (income quintile) as well as clinical variables such as pulmonary function tests, nutritional status, diabetes mellitus, sputum bacteriology and pancreatic status.

Results: The CPDR contained data on 4244 individuals with CF followed within Canada between 1993-2002. Of those individuals, 1452 could be found within the ICES administrative databases using probabilistic linkage. Of those 1452 individuals matched to the administrative databases, 1170 (45.6% female) had 7244 hospitalizations. A total of 312,015 OHIP claims were made during this 10-year period which represents any encounter between an individual with CF and a physician in Ontario. Multivariable analysis of hospitalization predictors is in progress.

Zhang, Z.; Lai, H. UW-Madison, Madison, WI, USA The 2002 CFF practice guidelines recommend adjusting for genetic potential when evaluating height status in children with CF. However, calculation of adjusted height percentile is not recommended due to the complexity of method involved in such calculation. Instead, a simple method to estimate a target height based on mid parental height was recommended. However, this target height includes a 10-cm confidence interval, which spans most of the channels on the growth chart. Therefore, a child's unadjusted height may be considerably below the target height, yet remains above the lower bound of the target height range. In addition, there is a paucity of data on the associations between adjusted height to lung disease outcomes to justify whether adjustment for genetic potential is necessary.

The objective of this study is to compare two methods of incorporating genetic potential in identifying short stature, namely, the CFF's target method and the method developed by Himes et al (Pediatr 1985; 75:304-313) , which applies positive adjustments to children with short parents and negative adjustments to those with tall parents. Data from 1986-2005 CFF Registry were utilized. A total of 558 children born between 1984-1988 who had complete height data from age 2-18 years and their parental height data were analyzed. Short stature was defined by unadjusted height < 10th percentile, unadjusted height below the lower bound of CFF's target height, or Himes' adjusted height <10th percentile.

Our results showed that adjusted height percentiles are consistently lower than unadjusted height percentiles, with an average difference of 0.2 Z-score unit between ages 2-14 years and 0.3 Z-score unit between ages 15-18. Proportionately more children were classified as short stature based on Himes method compared to unadjusted height (31% vs. 26%, p = 0.0003). In contrast, proportionately fewer children were below the lower bound of CFF's target height (23%) compared to unadjusted height < 10th percentile (26%), p < 0.001. Among children with average parents, unadjusted and adjusted height percentiles were similar. However, among children with short parents (< 25th percentile), the percentage children with of short stature decreased from 68% with unadjusted height to 35% with Himes' adjusted height and 14% with CFF's target height method. Among children with tall parents (> 75th percentile), the percentage of short stature increased from 7% with unadjusted height to 32% with Himes method and 35% with CFF's target height method. Taken together, these results demonstrate that, without incorporating genetic height potential, 2-5 times more CF children who have short parents would be classified as short stature, while 5 times fewer CF children who have tall parents would be classified as short stature. Further analyses show that, compared to unadjusted height percentile, Himes adjusted height percentile is more sensitive to, and has stronger association with, percent predicted FEV-1.

In conclusion, our findings provide evidence that genetic potential should be incorporated in evaluating short statue in CF children, particularly for children with short or tall parents. The CFF's target height method and Himes method differ significantly in identifying short stature in CF children with short parents. Cohort study of individuals in the CFF Patient Registry from 1995-2005. All individuals 6 years of age or older and diagnosed before age 45 were included. The new, persistent MRSA infection cohort was defined by having at least two cultures negative for MRSA over a two-year lead-in period followed by at least two MRSA positive cultures. Individuals with only one MRSA culture (transient infection) were excluded. We developed multiple linear regression models using generalized estimating equations to assess the effect of MRSA on FEV 1 %predicted (%FEV 1 ) and rate of change of %FEV 1 . Results: During the study period 5,955 individuals cultured MRSA. Of these, 2,025(34%) cultured MRSA only once (transient infection) and were excluded from subsequent analysis. 3,116 individuals met the criteria for new and persistent MRSA infection. 16,088 met the criteria for never having MRSA resulting in a total study cohort of 19,204. Participants were followed for a median of 10 years. The median time to first isolation of MRSA in the persistent MRSA cohort was 5.5 years and the mean number of positive MRSA cultures per patient was 7.5. A comparison of baseline characteristics between the persistent MRSA and never MRSA cohorts revealed the MRSA cohort to be younger (13.1 vs 15.0 years p<0.001), equivalent in lung function (%FEV 1 79.1% vs 78.3% p=NS), and with higher likelihood of methicillin sensitive Staph aureus colonization (45% vs. 39.3% p<0.001). 5.5 years into the study period (the median time of acquisition for new MRSA), the MRSA cohort had a lower %FEV 1 (68.1% vs 71.1% p<0.001), and averaged twice as many hospital admissions per year (0.88 vs 0.44 p<0.001) and home IV courses per year (0.40 vs 0.21 p<0.001). In an unadjusted analysis, the presence of MRSA was associated with a mean %FEV 1 that was 9.8% lower than the MRSA cohort (95%CI 9.6%-10.0%). However, after adjusting for sex, age, pancreatic status, P. aeruginosa, and B. cenocepacia, the presence of MRSA was associated with a mean %FEV 1 3.3% lower (95%CI 3.1%-3.5%). Most importantly, an analysis investigating the interaction between MRSA and time, after adjusting for confounders, indicated no statistically nor clinically meaningful difference in the mean rate of lung function decline between the two groups. Conclusions: Approximately one-third of patients who culture MRSA do so only once and do not subsequently develop persistent MRSA infection.

MRSA infection occurs in a group with more severe structural lung disease as measured by FEV 1 , and frequent hospitalizations and IV antibiotics are strongly associated with MRSA acquisition. After adjusting for confounders, MRSA infection is not statistically significantly associated with a more rapid decline in %FEV 1 over time.

Simmonds, N.J. 1, 2 ; MacNeill, S. 2 ; Cullinan, P. 2, 1 ; Hodson, M.E. 1, 2 1. Department of Cystic Fibrosis, Royal Brompton Hospital, London, United Kingdom; 2. National Heart and Lung Institute, Imperial College, London, United Kingdom Introduction: Disentangling the influence different factors have on long-term survival in CF is complex. There are a myriad of influences (including environmental, healthcare-related, psychological and socioeconomic) which may be important, as there is generally a poor genotype-phenotype correlation. The aim of this case-control study was to identify these factors.

The CF database at the Royal Brompton Hospital was used to identify long-term survivors. They were classified as cases and were patients who had reached 40 years of age (without transplantation). Each case was agematched with at least one control who died of CF (or a CF-related condition) or was transplanted before reaching 30 years of age. Late diagnosis patients were excluded. Conditional logistic regression models were used to identify potential influences on survival.

Results: 78 cases and 152 controls were analysed, producing 1811 matched pairs. 97% of the 230 patients were pancreatic insufficient. Of the many factors investigated, those resulting in increased probabilities of survival included: increased body mass index (OR=1.52, 95% CI 1.35-1.72), FEV 1 (OR per 5% increase=1.62, ) at transfer to the adult clinic (after adjusting for age and sex) and the use of oral antibiotics before attending the adult clinic (OR=3.01, 1.38-6.56) after adjusting for age at first attendance. Factors associated with a reduced probability of survival included: CF diagnosis <5 years old (OR=0.46, 0.28-0.76); initial presentation of chest symptoms or malabsorption (OR=0.44, 0.27-0.71 and OR=0.38, 0.24-0.61, respectively); referral from a paediatric clinic in a deprived area (OR=0.12, 0.04-0.34); pneumothorax before transfer to the adult clinic (OR=0.03, 0.005-0.24) after adjusting for age at first attendance; and S. aureus or P. aeruginosa colonisation before 16 years of age (OR=0.36, 0.14-0.89 and 0.18, 0.06-0.58, respectively). Factors not influencing survival included: sex, H. influenzae colonisation before 16 years of age, development of diabetes before 16 years of age, major haemoptysis before transfer to the adult clinic, parents' occupation, number of siblings (CF and non-CF) and school achievements.

In a very carefully matched study we failed to identify major non-clinical determinants of long-term survival in early diagnosis CF. Findings suggest that the majority of significant factors were directly related to optimal physical parameters, the use of oral antibiotics, avoiding bacterial colonisation and a low pneumothorax rate.

Adler, A. 1, 2 ; Bilton, D. 2 ; Haworth, C. 2 ; Gunn, E. 2 ; Shine, B. 3 1. Addenbrooke's Cambridge University Hospitals, Cambridge, United Kingdom; 2. Papworth Hospital, Papworth Everard, United Kingdom; 3. Oxford Centre for Diabetes and Endocrinology, Oxford, United Kingdom Background and Aims: Although the prevalence of diabetes in patients with CF (CFRD) is high, few prospective studies exist, and even for genet-ic factors, cross-sectional studies may bias the magnitude of associations. This study sought to identify risk factors for CFRD from a predominantly white cohort of individuals with CF in the United Kingdom. Methods: 8,029 individuals aged 0 -64 were identified from the UK CF Database, a registry of patients coordinated by the UK Cystic Fibrosis Trust with data from over 44 specialists centres from England, Scotland and Wales, countries which provide medical care free of charge. Of the patients, 5,196 had no diabetes at baseline (1996-2004) , and also had at least one follow-up annual visit. Diabetes was defined as either a physician diagnosis of diabetes, a blood glucose of >11.1 mmol/l two hours following an oral glucose tolerance test, or treatment with insulin or oral hypoglycemic drugs. Follow-up was calculated as time from baseline to the first detection of new diabetes or censoring. Risk factors included clinical, genetic (functional classes I,II,III vs IV,V), and anthropometric characteristics measured at baseline. Survival analyses were limited to patients with complete data. Cox proportional hazards modeling was used to estimate the magnitude of the association between potential risk factors and incident diabetes. One-way interactions with sex were tested. Results: The median age of entry into the cohort was 12 years and BMI was 17.9 kg/m2. 54% were male, and 96% were white. 526 patients developed diabetes during 15,718 person-years of follow-up (mean 3.0 years); the incidence was 3.4% per year. Patients who developed diabetes were more likely in univariate analyses to be older, have a high body mass index (BMI), bacterial pulmonary infections, liver and pulmonary function abnormalities, poor oxygen saturation, a history of organ transplantation, supplemental feeds, and take pancreatic enzymes or bile acids. In addition, patients with Class IV,V genotypes were less likely to develop diabetes relative to patients with other genotypes, as were patients screened for CF at birth. Not associated with incident diabetes was age at diagnosis of CF and ethnicity. In multivariate analyses (n= 4,004) age (hazard ratio 1.02 per year, 95% CI 1.01 -1.03), female sex (1.67 (95% CI 1.39-2.00), % predicted forced expiratory volume in 1 second (1.019, 1.014 -1.024 per 1% decrement), dose of pancreatic enzyme replacement (1.33, 1.10 -1.62), above vs below median dosage, use of bile acid supplements (1.51, 1.20 -1.91) , and genotype class I,II, III (5.60, 2.08 -15.1) relative to class IV/V. These associations were not confounded by BMI or history of transplantation. There were no significant interactions between sex and other risk factors. Conclusions: The study confirms the high incidence of CFRD, and shows that patients with class IV/V genotypes are less likely to develop diabetes independent of other CF-related complications. Females are disproportionately at risk for diabetes which is not explained by a higher prevalence of known risk factors nor by effect modification. Cystic fibrosis children are at increased risk for hearing disorders due aminoglycoside exposure, mucopurulent upper respiratory tract infections, and increased inflammatory responses. Pure tone audiometry is the standard instrument used to determine hearing loss. However, it can be expensive and time-consuming to use audiometry for routine hearing screening. Hearing questionnaires have been used as a screening tool to assess possible hearing loss in otherwise healthy populations. What are the results of hearing questionnaires administered to pediatric CF patients? Does CF patient age or ethnicity increase the likelihood of hearing loss as determined by a hearing questionnaire? A hearing questionnaire (English and Spanish versions) was administered to pediatric CF patients over a 2 week period at our accredited CF Center. Data collected included demographic information. Inclusion criteria: age м10 years and in baseline health status. Possible hearing loss was defined as at least 3 "Yes" responses (>3) to the 14 quesions administered to the patients. Fisher's exact test was used to perform the analyses. A total of 20 CF patients were seen in the CF clinic during the study period. 14 CF patients completed the Hearing Questionnaire. Mean age was 14.3 ± 3.6 years (median 13.6 years). There were 9 males and 5 females; 10 were classified as Hispanic and 4 were classified as Caucasian. 7 patients were older than 13.6 years (the median age) and 7 patients were less than 13.6 years. 3 patients who were older than age 13.6 years had a score of >3 and 4 patients < age 13.6 years had scores >3 (p=1.0, ns). Thus, 7 of 14 CF patients (50%) were classified as having possible hearing loss as per their questionnaire scores. 6 of 10 Hispanic CF patients had hearing scores >3 and 1 of 4 Caucasian patients had hearing score >3 (p=0.56, ns) . We found that a significant proportion (50%) of our CF children may require formal hearing evaluation. We speculate that hearing questionnaires may be useful to screen for hearing loss in the CF patient population. Hispanic CF patients have increased morbidity and mortality risks compared to Caucasian CF patients who attend our accredited Cystic Fibrosis Center in Southern California. Is the increased morbidity/mortality risk due to decreased access to outpatient CF care? We performed a retrospective review of all patient visits to our CF Clinic between January 1, 2005 through December 31, 2006 . Data collected included demographic information (age, ethnicity, gender) , and the number of clinic visits during the study period. Data was analyzed by unpaired Student's t-test as well as by Chi-square. During the study period, a total of 197 CF patients (106 males:91 females; mean age 10.8 ± 5.7 years; range 3 months to 21 years) were routinely followed in our CF Clinic. These patients had a total of 1806 outpatient visits to the CF Clinic (9.2 clinic visits/patient over the 2 year study period). There were 82 Hispanic CF patients (46 males:36 females; mean age 11.0 ± 5.3 years) and 99 Caucasian CF patients (49 males:50 females; mean age 10.8 ± 6.2 years; p=0.42, ns). There was no difference in gender distribution between the 2 group (p=0.52, ns). Hispanic patients had a greater number of CF clinic visits (11.2 visits/patient) as compared to Caucasian patients (7.9 vists/patient; p=0.03) during the same period. We conclude that the increased morbidity and mortality of Hispanic CF patients in California cannot be attributed to barriers to outpatient care. In fact, Hispanic CF patients were seen more frequently in CF Clinic compared to Caucasian CF patients who attend the same clinic. We speculate that Hispanic CF patients are seen more frequently in CF Clinic due to increased severity of CF disease manifestations.

Duguépéroux, I. 1 ; Scotet, V. 1 ; Audrézet, M. 1, 2 ; Blayau, M. 3 ; Parent, P. 4 ; Journel, H. 5 ; Boisseau, P. 6 ; Férec, C. 1, 2 1. Inserm U 613, Brest, France; 2. Dep. of genetics, Brest, France; 3. Dep. of genetics, Rennes, France; 4. Unit of medical genetics, Brest, France; 5. Unit of medical genetics, Vannes, France; 6. Dep. of genetics, Nantes, France Aim: The aim of this study was to describe 15-year experience of prenatal diagnosis (PD) for CF in Brittany (western France) and to assess its impact on incidence. Method: We registered, by the genetic laboratories of our region, all the PDs performed in women living in Brittany over the period 1991-2005. We described the number of PDs made for each reason (way by which the one-in-four risk was identified: previous affected child, family testing, echogenic bowel, etc). We reported the proportion of CF-affected fetuses and of consecutive terminations, and assessed the incidence modification due to PD. Results: Over the 15-year period, a total of 253 PDs were performed in couples living in Brittany. Most of them were done in couples already having CF child(ren) (n=167, 66.0%). Extended testing in families -a practice largely proposed in Brittany -led to the identification of 18 new onein-four risk couples among the relatives of CF patients who opted for PD 38 times over the study period (15.0%). The 48 other PDs were made in couples without previous history of CF. The one-in-four risk was mainly identified following the detection of an echogenic bowel during pregnancy ultrasound examination (21 PDs were done directly following the positive ultrasound (8.3%), whereas 19 others were done for subsequent pregnancies (7.5%)). The other PDs were consecutive to the detection of an heterozygote through newborn screening (n=6, 2.4%) or for an other reasons (n=2, 0.8%). Over the study period, the number of PDs per couple varied between one and five, the mean being 1.6. Overall, a total of 88 CF fetuses were identified, among whom 77 were terminated (87.5%). The inclusion in the incidence calculation of these 77 pregnancy terminations led that to an incidence modification of 28.8% over the study period. Conclusion: This study shows that PD for CF is commonly used in Brittany and highlights the impact of family testing and of routine ultrasound examination of pregnancies in that region.

Supported Context: Hispanics with cystic fibrosis (CF) in the United States experience an 85% increased annual risk of death from CF compared to non-Hispanic patients. When adjusted for socioeconomic status this disparity persists. Studies characterizing this at-risk population do not exist.

Objective: To characterize the center-reported US Hispanic population with cystic fibrosis through a Cystic Fibrosis Foundation (CFF) patient registry analysis and to elucidate factors present within this Hispanic population that may affect outcomes.

Design: Retrospective cross sectional study of the 2004 CFF patient registry.

Patients: All 22,714 patients in the 2004 CFF patient registry were included in the analysis. There were no exclusion criteria. The Hispanic population is defined as those who were entered by their respective care center as Hispanic ethnicity regardless of race.

Main Study Measures: Prior to analysis, study measures were selected to demographically and genetically characterize the Hispanic CF population. Genotype, state of residence, mean age and mean age of diagnosis were obtained in the entire population. Maternal education (for patients <18 years of age) and insurance status data were captured as indicators of socioeconomic status. Complication rates, FEV1 percent predicted (under 18 years of age) and absolute FEV1 (over 18 years old of age) were obtained as measures of clinical status. T-tests and logistic regression analyses were used to compare measures between Hispanic and non-Hispanic groups. Odds ratios (OR) and 95% confidence intervals are reported.

Results: 1511 center-reported Hispanic patients with CF were identified. Over 50% of Hispanic patients resided in California, Florida, Texas or New York. The most common genotype was delta F508 homozygous, accounting for 32% of the population. Mean age was 12.3 years +/-9.2 years for Hispanic patients and 17.3 years +/-12.1 years for non-Hispanic patients. Hispanic patients were diagnosed at an earlier age (2.8 years vs. 3.3 years, p=0.005). They were more likely to have mothers with education at less than a high school level (OR 5.25 (4.12, 6.71) ) and have government insurance (OR 2.73 (2.45, 3.04) ). Complication patterns differed between the two groups; non-Hispanics were more likely to have complications reported that were related to depression (OR 1.94 (1.48, 2.54) ), bone health (OR 1.82 (1.41, 2.41) ) and CF related diabetes (OR 1.9 (1.6, 2. 3)) compared to Hispanics. Hispanics however, were more likely to have complications reported from CF liver disease (OR 1.31 (1.1, 1.56) ). Pulmonary function, as measured by FEV1 percent predicted in children (p<0.001) and absolute FEV1 for adults (p=0.01) were lower for Hispanic patients.

Conclusions: Differences exist between Hispanics and non-Hispanics with CF with respect to maternal education, insurance status, complications rates and pulmonary function. The lower reported rates of depression, bone complications and CF related diabetes may represent ascertainment bias. Further study is needed into the etiology of health outcome disparities in this population, and to design interventions for this high-risk population. Mastella, G.; Baldo, E.; Forneris, M.; Furnari, M.; Lucidi, V.; Manunza, D.; Marinelli, I.; Messore, B.; Neri, A.; Raia, V.; Salvatore, D.; Buzzetti, R.; CAIRO Group, Italian CF Research Foundation, Verona, Italy With our work we reviewed the international scientific literature coming from CF Registries worldwide. Our aims were: to verify what has been produced in scientific literature thanks to the material derived from FC Registries and to see which clinical problems have been tackled and which clinical questions were answered correctly and exhaustively.

A search in Medline and Embase produced 160 articles (our strategy was "Cystic fibrosis"[MESH] AND ("Registry"[MESH]) OR registr$) updated to the 04/30/07). Out of these articles 88 have been selected by two independent assessors on the basis of their pertinence (primary studies that used data taken from a formally established registry, at least national, to test some hypothesis of research).

Each article has been examined with the help of a pre-defined form that in particular evaluated skills of different registry data, selected clinical characteristics of the study populations and statistical methods.

More than one half of cases data came from USA-CFF Foundation Patient Registry, the remaining coming from Canadian, UK, French, Italian, German, and other European Registries.

The main focuses of the articles were included in the following topics: 1) Incidence/prevalence of the disease and survival (30 studies); 2) neonatal screening, growth and nutrition (18); 3) genotype/phenotype correlation/ different ethnical groups and twin/brothers (12); 4) complications and outcomes (ABPA, diabetes, nasal polyposis, pregnancy and paternity, etc) (19); 5) microbiology (9).

Our assessment scores were good/excellent for 62/88 studies for the relevance of the problem addressed; 54/88 for the usefulness of the outcomes and 46/88 for the usefulness of the implications for research from the Italian registry. Thirty-two articles gave a significant contribution to the analysis of the studied problem, but 47 of them left a partial uncertainty, while 9 left complete uncertainty.

CF Registries are a very important source of data to provide a powerful tool for clinical and research questions. However, further comparable studies have to be planned to assess registry data as a platform for improvement in clinical practice.

VandenBranden, S.L. 1 ; McMullen, A. 7 ; Konstan, M. 4 ; Morgan, W. 3 ; Wagener, J. 6 ; Schechter, M. 2 ; Watts, K. 1, 5 ; McColley, S. 1, 5 1. Children's Memorial Medical Center, Chicago, IL, USA; 2. Emory University, Atlanta, GA, USA; 3. University of Arizona, Tuscon, AZ, USA; 4. Case Western Reserve University, Cleveland, OH, USA; 5. Northwestern University Feinberg School of Medicine, Chicago, IL, USA; 6. University of Colorado, Denver, CO, USA; 7. University of Rochester, Rochester, NY, USA Background: Although life expectancy in CF has dramatically improved over the last 20 years, most of the improvement is due to improved survival during early childhood. Adolescence and young adulthood continue to be a time of worsening pulmonary disease and mortality. Study Objective: Data from a large longitudinal observational study, the Epidemiologic Study of Cystic Fibrosis (ESCF), was examined to characterize lung function decline in subjects with CF during the periods preceding and after the age of 18. Design: ESCF data from 2136 individuals with CF collected during the period of 1994-2005 were analyzed. Inclusion criteria required individuals to have a minimum of 5 encounters in each of the 3.5 year time periods before and after the age of 18. The cohort was strati-fied by disease severity based on the best FEV 1 at the 18th year (+/-6 months). One hundred thirty seven subjects had severe lung disease (FEV 1 <40% predicted), 668 had moderate-severe lung disease (FEV 1 40-<70 % predicted), 1049 had mild lung disease (FEV 1 70-<100% predicted), and 282 had very mild disease (FEV 1 >100% predicted) The study compared the annualized rate of FEV 1 decline during the adolescent period, defined as ages 14-17.5 years, and in the young adult period defined as 18.5-22 years. Results: In the entire cohort, the annualized rate of decline was less in the young adult period than in the adolescent period (p<0.001). The most dramatic improvement in slope of decline was seen in the severe group, and significant improvements were also noted in the moderate group (p<0.001). The mild group demonstrated insignificant change; the very mild group was the only group that demonstrated a more negative annualized rate of the decline in young adulthood than in adolescence (p=0.027). Conclusions: These data suggest that, overall, young adults have a slower rate of decline in FEV 1 in the period after age 18. This "stabilizing" of lung function is most prominent in those individuals with moderate-to-severe lung disease. Only those with very mild lung disease have a slightly more rapid rate of decline between the ages of 18 and 22. Overall there is less variation in the rate of FEV 1 decline across disease strata after the 18th birthday. Further analysis is underway to better understand the factors leading to these findings. 1 1. Department of Paediatrics, University of Florence, Cystic Fibrosis Centre, Florence, Italy; 2. Italian Cystic Fibrosis Microbiology Group, Florence, Italy; 3. Institute of Medical Microbiology and Hygiene, Tübingen, Germany MRSA is now a worldwide public health problem, due to the increasing rate of infection in several settings as well as in CF patients. MRSA was first recognized as being acquired from hospitalized patients (HA-MRSA), but the onset of MRSA infection outside the hospital setting, in communityassociated strains (CA-MRSA), has recently been described with increasing frequency. HA-MRSA are known to be responsible for infections in hospitalized patients, although highly virulent CA-MRSA are increasingly reported worldwide as the cause of severe outbreaks, replacing HA-MRSA strains in nosocomial infections. Little evidence is available about the characterization (SCCmec type) and epidemiology of community-acquired MRSA (CA-MRSA) or hospital-acquired MRSA (HA-MRSA), transmissibility, antibiotic susceptibility and virulence in the CF population.

The present multicenter study investigated the susceptibility pattern, epidemiology, and SCCmec type of MRSA strains isolated from nine Italian CF centers. The susceptibility pattern was determined by the disk diffusion method and Pulsed Field Gel Electrophoresis (PFGE) was performed for epidemiological purposes. The SCCmec type in order to identify hospital-and community-acquired MRSA strains (HA-MRSA and CA-MRSA respectively) was determined with molecular methods, and the presence of gene encoding Panton-Valentine leukocidin (PVL) was tested. During the study period, 181 MRSA strains were isolated from 178 out of 2362 (7.5%) patients attending these CF centers. Of the SCCmec type representing HA-MRSA, our data showed an high prevalence of SCCmec I (49.4%) while the prevalence of SCCmec II, (reported as the most represented among HA-MRSA) was only 1.1%. Furthermore a high prevalence (31.4%) was found of SCCmec IV (suggestive of CA-MRSA strains).

Epidemiological analysis showed that 31 (17.1%) out of 178 analyzed MRSA strains belong to the same PFGE clone shared among six centers, belonging to the SCCmec type IV, suggestive of CA-MRSA isolates. Twenty-four (77.4 %) out of these 31 strains were SCCmec type IV, indicative of CA-MRSA, as suggested by the good susceptibility rate to trimethoprim-sulfamethoxazole and rifampin. Surprisingly all the isolates belonging to this epidemic clone were negative for the PVL genes.

These results show that CA-MRSA is now spreading in the CF population, documenting the first epidemic of CA-MRSA in CF patients who are considered at risk for HA-MRSA acquisition due to frequent hospitalizations. These data show a peculiar picture of MRSA epidemiology indicating that further studies are required to clarify the role of CA-MRSA in global epidemiology, their pathogenicity and clinical impact on CF patients.

We thank the Fondazione per la Ricerca sulla Fibrosi Cistica-ONLUS for its grant (FFC#12 2006) . There are concerns regarding an increased risk of cancer in patients with cystic fibrosis (CF). This study aimed to review the occurrence of malignant disease in a large population of adult patients with CF.

The case notes of 372 patients attending the Leeds Adult CF Unit were reviewed. Demographics and data regarding a diagnosis of malignant disease were recorded.

A total of eight patients (3 male, 5 female) were diagnosed with a malignant condition over the 10 year period. The median (range) age of diagnosis of a malignant condition was 30.5 years (23-55 years).

Five patients were diagnosed with malignancy post-transplantation. Within this group of patients there were two cases of lymphoproliferative disease, one case of basal cell carcinoma, one case of liver cancer and one case of small cell lung cancer related to the donor lung. The median (range) time to diagnosis post-transplantation was two years (1-6 years) with a median (range) age of 26 years (23-55 years).

Three patients were diagnosed with a malignant condition without solid organ transplantation. This group consisted of one case of pancreatic carcinoma, one case of oesophageal carcinoma and one case of bowel carcinoma. The median age of diagnosis in this group of patients was 44 years (27-55 years) .

Prior to transplantation we found all malignant diagnoses were related to the gastrointestinal tract. Post-transplantation we found a wider variety of malignant diagnoses, with lymphoproliferative disease being most common.

Awareness of the increased risk of malignant disease is important and appropriate investigations should be undertaken particularly in patients with atypical symptoms. Background: Patient outcomes can be improved by implementing evidence-based guidelines for initiating therapies. In our CF center, practitioners have traditionally prescribed medications without a centralized database to screen patients for evidence-based criteria.

Goals: The primary goal of this project was to increase CF practitioner, family and patient awareness of evidence-based medication guidelines to facilitate discussions regarding therapies. The secondary goal of this quality improvement (QI) project was to determine if CF practitioner prescribing patterns would be altered by increasing awareness.

Methods: The CF steering committee at Cincinnati Children's Hospital Medical Center (CCHMC) identified oral azithromycin, inhaled tobramycin, and dornase alpha as target medications for this QI project. An evidence-based medicine committee was formed, including a parent advocate, to direct the project. A consensus was reached among CF practitioners for when to initiate therapy discussions. Dornase alpha was recommended for CF patients ages six and older. Inhaled tobramycin was recommended for CF patients ages six and older with chronic Pseudomonas (defined as 2 positive cultures in past 12 months). Oral azithromycin was recommended for CF patients ages six and older, with chronic Pseudomonas, and a weight of 25 kilograms or more. The CF database was screened to determine each patient's medication eligibilities. Eligibilities were reviewed weekly at our CF chart conference. Medication specific family and patient handouts were developed, reviewed by the parent advocate, and given to eligible families and patients on arrival to clinic. Data was collected to document the number of discussions facilitated at clinic visits, the decisions regarding medications, and the number of new treatment prescriptions that were facilitated. After seven weeks of the initiative, statistical analysis was conducted using McNemar's test for differences. A chi-square test with a two sided alpha of p <0.05 was used to determine significance.

Results: 148 of the 203 CF patients in our center were eligible for pulmonary medications using our evidence-based criteria. There were 253 individual prescribing opportunities in these 148 patients (taking each medication for each patient as a unique prescribing opportunity). At the start of the project, 64.3% (163/253) of the prescribing opportunities were fulfilled. 50% (74/148) of patients were prescribed all of their recommended medications. During the first seven weeks of the project, CF practitioners discussed initiating new pulmonary medications with 40 CF families and patients. The proportion of prescribing opportunities fulfilled increased from 64.3% (163/253) to 75.5% (191/253) (p value < .0001; 95% CI 65.3%-82.8%). The proportion of patients prescribed all of their recommended medications increased from 50% (74/148) to 65.5% (97/148) (p value < .0001; 95% CI 57.8%-80.0%).

Conclusions: Increasing awareness of evidence-based pulmonary medication recommendations with CF practitioners, families and patients facilitates therapy discussions, increases the proportion of therapies prescribed and increases the proportion of patients on all recommended therapies. PURPOSE: Our purpose was to optimize lung function (FEV1) by consistently and aggressively diagnosing and treating Pulmonary Exacerbations (PEx) in all CF patients at our Center. It was important to work on this because our families wanted standardized care and we felt our patients' FEV1's were suboptimal.

METHODS: Our Core LLC Team included two parents of CF patients and one adult patient. We met weekly, received coaching every other week, and learned and applied quality improvement techniques. Physicians began monthly meetings, and monthly meetings of the entire CF Team continued. The FAB met every other month. We identified our specific aims, and approached them systematically.

The providers, starting with criteria used by several authors, agreed upon the definition of PEx and created a Pulmonary Exacerbation Score Sheet (PExSS). We selected a score of 4 as the threshold for defining a PEx. After using the PExSS for three months, all providers reviewed the individual PExSSs and validated our original choice of 4. Run charts were produced and reviewed periodically to display each provider's percent of PExSSs used. When the use of the PExSS was not standardized after a year, it was reprinted on bright yellow paper to make it stand out.

To educate the families and patients about the early symptoms of PEx, a refrigerator magnet was designed and distributed that included all PEx criteria.

RESULTS: We standardized the use of our PExSS by all providers on all patients seen in CF clinic with ongoing measurement of its use and regular feedback. Many patients and families gained greater knowledge about what a PEx was, and the importance of early recognition and treatment. The CF nurses note that 1) many patients call with milder PEx symptoms than they did before, 2) providers treat PEx earlier, 3) many patients call with their PEx score as well as their symptoms. We hospitalized more patients for PEx after this project began, with an average of 62 patients/year in the 3 years prior, and 84 in 2006. The median FEV1 % predicted of our 6- CONCLUSIONS: Participating in the LLC changed the culture and functioning of our CF Team. Limitations to our study include: 1) FEV1 change may not have been due solely to this project, 2) the groups who's FEV1's were averaged over the various years are different. We will continue to optimize and standardize 1) prevention of PEx, 2) inpatient care, 3) care of PEx in CF providers' private offices, and 4) follow up of patients with PEx. Patient adherence to Airway Clearance Techniques (ACT) in the treatment of Cystic Fibrosis is known to be sub-optimal resulting in deterioration of lung function. The global aim of our project was to improve median FEV1. The specific aim was to improve adherence to ACT. We believed that these goals could be acheived by implementing a program which included re-education of airway clearance techniques (REACT) for our cystic fibrosis patients.

METHODS:

The initial phase of this project began with the administration of an anonymous survey to patients/families. This survey was designed to assess ACT practices, knowledge of the rationale for performing ACT and barriers that prevent patients/families from adhereing to ACT. The next step included an in-clinic questionnaire and patient/family demonstration of current aerosol therapy and ACT. Based on individual results patients were identified as: adherent with correct technique, adherent with incorrect technique or non-adherent. We defined adherent as correct performance of ACT one to two times daily as prescribed. Then all patients entered the REACT program which included a standardized tutorial on proper ACT and correction of techniques if necessary. In the non-adherent group barriers to performing ACT were discussed and problem solving techniques were used to overcome barriers. The patients in the non-adherent group were re-evaluated in clinic monthly while the adherent patients returned on an every two month schedule.

RESULTS: Analysis of our initial survey revealed that while the majority of patients/families reported performing ACT daily, the duration of treatments was less then medically prescribed. Many patients/families reported barriers which decreased adherence to therapy. To date 85% (74/87) of our patients have completed the in-clinic questionnaire and entered the REACT program. The results of these evaluations and patient demonstrations revealed: 37% (27/74) were adherent with correct technique, 3% (2/74) were adherent with incorrect technique and 60% (44/74) were non-adherent.

In the six months since the implementation of the REACT Program, patients/families are self-reporting an increased adherence to ACT and the median FEV1 for our center has increased form 84.7% (2005 CFF Registry data, ages 6-18 years) to 87.4% (CFF registry, first quarter, 2007, ages 6-17 years).

CONCLUSION: While adherence to ACT in our center was poor, we were able to improve adherence and ultimately improve the median FEV1 by implementing a program of re-education for ACT (REACT). We believe re-assessment of adherence to therapy and reviewing correct technique is necessary at every clinic visit. Continuous quality improvement (CQI) techniques have recently been employed in medicine to improve quality of care. Although there is evidence that CQI improves efficiency in hospitals, little is known about effects of CQI on improving patient outcomes. We have previously reported creation of a unique Nutritional Risk Pathway (NRP) to care for cystic fibrosis (CF) patients who are at nutritional risk. The goal of this study is to assess perceived impact and objective clinical efficiency of our NRP.

Patient data was recorded from all children attending the CF Clinic over an 18-month observation period (between Sep 05 and Mar 07). BMI percentiles were assessed and nutritional risk zones -green: acceptable (BMI > 25%ile), yellow: at risk (BMI 10-25%ile) or red: at urgent risk (BMI < 10%ile) -assigned at each visit. The visit intervals were every 4 to 12 weeks based on severity of nutritional risk. Patients that remained in the red zone for an average of 3 to 6 months with no improvement were considered for gastrostomy tube (GT) placement. The CF Dietitian updated the data weekly and e-mailed quarterly to CF physicians for review of their patients' status. Patient/family perceptions and understanding of the NRP was assessed by questionnaire.

Of the 164 children in the study, 29 (18%) were identified to be at nutritional risk and were enrolled into the NRP: 19 in red and 10 in yellow zones; 13 (45 %) male, 16 (55%) female; 18 (62%) age < 12 yrs, 11 (38%) age ≥ 12 yrs. Over the 18-month observation period, 21 (72%) showed improvement (7 moved from red to yellow, 7 from red to green and 7 from yellow to green zones). Thirteen (81%) females improved, whereas 8 (62%) males improved. In the under 12 yrs age group, 17 (94%) improved compared to 4 (36%) in the ≥ 12 yrs age group. As a result of participation in the NRP, 9 (31%) children had GT placed: six of the 9 showed improvement with four moving from the red to green zone. From the survey responders, 80% stated they understood the meaning of the CF NRP. All responders were extremely to somewhat motivated to focus on their child's nutrition when asked to return for growth monitoring or follow-up visits. Having specific goals for weight gain, calorie requirements and BMI were considered extremely helpful by 100% of families. When GT placement for their child was recommended, 40% felt they lacked all the necessary information to make a decision. Our new CF Nutrition Education booklet was not received by 40% of families. Some of the physicians who received the quarterly nutritional patient data provided positive feedback and found this information beneficial for patient care.

Nutritional CQI in our CF Center utilizing a specific NRP and monitoring of nutritional risk status resulted in improved patient outcomes. Overall perceptions of our NRP were positive from both patients and CF physicians. These results suggest that this CQI strategy improved clinical outcomes in our patient population. Patients with CF require frequent healthcare visits to optimize growth, initiate early interventions and delay progression of lung disease. The CFF recommends evaluation at a CF Center at least quarterly. As part of a LLCIV quality improvement (QI) initiative, our CF team set a goal to increase the percent of CF patients attending clinic 4 or more times a year to 90% or greater by developing strategies to monitor and improve adherence rates.

Methods:

We reviewed CF patients who attended clinic 3 times or less in 2005 to determine patterns of non-adherence in regards to age, gender, distance from Center, season, physician or insurer. A patient survey was used to assess clinic attendance barriers from a patient/family perspective. A brainstorming session identified other barriers. A fishbone diagram was drafted into 4 barrier categories: external, internal, communication, and patient/family perceptions. After reviewing all barriers, we concluded the most effective improvements could be made to internal clinic barriers.

Clinic processes were then assessed including making of appointments, reminder letters/calls, a time cycle analysis, and number of CF appointments available per patient per physician. Our Family Advisory Board participated by focusing on the importance of 4 or more clinic visits in their newsletter and by assignment of a member to the LLCIV team. We discovered a lack of a standardized process to follow-up (FU) patients who "did not keep appointments" (DNKA). A spreadsheet was created to monitor physician DNKA rates. A primary FU strategy was devised to identify patients who DNKA through daily auto-emails to the CF office assistant and utilization of a rescheduling decision tree for reappointment of these patients by our appointment center. Two secondary FU strategies were developed. The first involves the Clinical Coordinator printing a monthly report off Port CF entitled "Patients Due a Visit" then emailing this list to the CF team to contact and reappoint. The second involves the CF Social Worker tracking patients seen per quarter on a spreadsheet and emailing the spreadsheet quarterly to the CF team as well as posting on our datawall. Results: In 2004 , and 2006 , 35%, 56%, and 89% of our patients attended clinic 4 or more times, respectively. In 2004 In , and 2006 , 33%, 25%, and 23% of our patients were at nutritional risk, respectively. In 2004 and 2005, mean FEV1% predicted was 88.3% and 90.4%, respectively.

Conclusion:

The number of CF patients attending clinic quarterly at our Center increased by implementing processes to track DNKA patients and reappoint them to clinic promptly. However, some of the early improvement in clinic attendance was partially due to an increased awareness of our QI efforts by our clinic staff, CF team and patients/families. We speculate that improvements in nutritional status and lung function are at least partially related to more frequent clinic attendance by our patients. Continued monitoring of the number of DNKA patients and patients who attend clinic quarterly is required to sustain these gains in improved patient outcomes. Data When we first measured patient/family satisfaction in 2003, 31 parents and 14 children were given questionnaires to measure their satisfaction with services provided at the clinic. At that time, responses in our survey were very positive. With regards to our new model we received very positive feedback, encouraging us to continue with the model. Parents and children liked the elimination of duplication in questions and they also like having the whole team involved in the meeting and problem-solving. Final results for our second satisfaction questionnaire will not be available until June of this year as the questionnaires are being given at clinics over a 3 month period this spring. The children complete a questionnaire with the help of their parents if needed and the parents also complete a survey. Results to date are very positive for both the parents and the children. Of the 14 children who have completed the survey to date, all stated they would like to continue to meet with the group as a whole as opposed to meeting individually with team members. Of the 14 parents surveyed to date, all are either very satisfied or satisfied with the group format.

Conclusion: We have not yet completed the survey but results to date clearly show that both the patients and their families are happy with the round table model.

Results will be compiled early this summer and would be ready for the poster and/or presentation in the fall. The CFF clinical care guidelines recommend routine clinic visits for assessments, interventions, monitoring, education and counseling every 3 months or more often as indicated. Adherence to these guidelines is important in ensuring better clinical outcomes. In reviewing our CF Center data from the 2003 CFF Registry Report, 46.2% of pediatric patients (6-18 yrs.) and 47.8% of adults (18 yrs. and older) adhered to the guidelines. National goals were not provided then; instead data was given for the top ten CF Centers (80.9% and 78.5% for pediatric and adult patients respectively) as well as the national rates (55.5% and 44.8% for pediatric and adult patients respectively). Our CF Care Center acknowledged the need to improve adherence to these guidelines and at the beginning of the year 2005, embarked on a quality improvement project to address this issue. Following the PDSA (Plan-Do-Study-Act) steps, we agreed on the objective of increasing the percentage of pediatric and adult patients adhering to the guidelines by at least 50% and at the same time, to exceed the national average in 1 year or by the end of the year 2005. The physicians and social worker obtained feedback from patients and/or families during a clinic visit regarding barriers in complying with scheduled clinic visits. Suggestions were elicited to overcome these barriers. The team members then identified barriers in meeting the guidelines and implemented measures to improve compliance with clinical care guidelines. Among these measures were sharing the information about clinical care guidelines and importance of adherence through our CF Newsletter and at every clinic visit, reminder calls to patients or caregivers at least 2 days prior to the visit, appointments for the next visit given prior to discharge from the clinic, and same day calls to patients who failed to keep their appointment for rescheduling. Our Social Worker identified patients with difficulties or problems in consistently keeping appointments and helped address these problems. Progress made in this endeavor was shared with parents and caregivers during the "Parents' Night" held in the fall of 2006. Indeed, with consistent implementation of these measures, we have seen improvement in the percentage of patients adhering to guidelines. In our 2005 CF Care Center data, 68.2% of pediatric patients and 89.5% of adult patients adhered to recommended clinical care guidelines, approximating and in fact for the adults, exceeding, our target of 50% increase in the percentage (48% and 87% increase respectively for pediatric and adult patients). These data were better than the national average for 2005 (63.9% and 48.3% for pediatric and adult patients respectively) but for pediatric patients, still 22% below the national goal of 90%. Adults approximated the national goal of 90%. We are committed to continually engage in the process of quality improvement in further improving adherence to clinical care guidelines and to hold on to the gains. Weight z-scores and height z-scores were monitored prospectively for all infants, children, and adolescents at a pediatric outpatient CF clinic during an ongoing, multidisciplinary quality improvement (QI) initiative. All CF team members participated in this QI project which emphasized the importance of good nutrition and adequate growth for optimal lung health. Specific strategies for improvement included education to families at each visit with a written Nutrition Action Plan, increased emphasis on liquid supplements and enteral feeds as needed, and a consist message from all CF clinic staff in encouraging attention to nutrition. The study objective was to determine if multidisciplinary QI interventions to improve nutritional status had any effect on weight z-scores and/or height z-scores over a 3-year period. The null hypothesis was that z-scores for weight and height would either not improve or worsen. Inclusion criterion for this study was attendance at CF oupatient clinic during the first 5 months of the QI initiative (time 1) with a subsequent clinic visit 3 years later (time 2). Patients not attending clinic at both time 1 and time 2 were excluded. A cohort of 119 patients (53 male, 66 female) met the inclusion criterion. The mean time 1 to time 2 interval was 3.05 ± .11 years (range 2.70 to 3.40 years) with mean patient age at time 1 of 6.75 ± 4.55 years (range .10 to 15.80 years). Pancreatic enzyme replacement therapy was used for 112 (94.1%) of the cohort. Paired sample t-tests indicated significant improvements for weight z-score (p <.000, t (118) = 6.16) and height z-score (p <.000, t (118) = 4.49) after the 3-year intervention. At time 1, 89 children (74.9%) were underweight or less than 50th percentile while 30 children (25.2%) were normal-weight, at or more than 50th percentile for age. Mean differences in both weight and height z-scores in the initially underweight patients were greater than those for the normal-weight patients over the 3 year interval. No significant differences were found for the initially normal weight-for-age children between time 1 and time 2 for weight z-score ( p = .83) or height z-score (p =.37). Children who were not underweight at time 1 grew normally, without acceleration over the 3 yr intervention. Therefore, we conclude that a longterm multidisciplinary QI project to enhance nutritional status resulted in significantly improved weight z-scores and height z-scores for those individuals who were initially underweight for age. The most common cause of morbidity and mortality for patients with cystic fibrosis (CF) is pulmonary exacerbations. These exacerbations lead to hospital admissions and treatment with intravenous (IV) antibiotics. All CF patients at our facility are admitted to a teaching service which includes house officers and students. Standardized order forms are used to improve accuracy and timeliness of medications and therapies. The efficiency of this process was brought into question in 08/2006, when several parents complained about a delay in the receipt of first dose IV antibiotics after hospital admission. The objectives of this study were to determine what were the actual delivery times of the IV antibiotics to the patients (both retrospectively and prospectively), and areas where improvements could be made. Our predetermined target time was < 3 hours.

Two analyses were performed in order to answer the objectives. The first analysis was done in a retrospective fashion. This analysis included CF patients admitted from 5/1/2006-9/30/2006. A total of 44 patients were included. Total median time from admission to receipt of first dose of IV antibiotics was analyzed. A second collection of blinded prospective data was performed between 10/1/2006-12/31/06. Only the pulmonologist, pharmacist collecting data, and pharmacy clinical manager were aware of the data collection. A total of 15 patients were included in the prospective project. Each part of the process model were recorded and analyzed so that areas of improvement could be realized.

The median time from hospital admission to first dose of antibiotics was 7 hours (range 0-31 hours) (retrospective cohort). The prospective cohort median time from hospital admission to first dose of antibiotics was 6 hours (range 67 min-11 hrs). The median time from admission to order entry by pharmacy: 2 hrs, 24 min (range 35 min-4 hrs, 11 min). The median time from the medication order being written by medical practitioner to order entry by pharmacy: 55 minutes (range 3 min-3 hrs, 20 min). The median turn-around time by the hospital pharmacy of order entry until reaching the nursing unit: 19 min (range 3 min-50 min). The median time required by the nursing staff to administer the antibiotic: 2 hrs, 23 minutes (range 29 min-6 hrs).

This project was initiated in response to parental concerns. From the results of the retrospective and prospective analyses, it was determined that a delay is occurring in the receipt of IV antibiotics, which translates to hours in the hospital without therapy. This is both inconvenient and uncomfortable to the family, and wasteful of medical resources. The findings of this project were presented to the Pharmacy and Therapeutic Committee of the hospital, to the nursing directors, and to the parents of cystic fibrosis patients. The delay is greater than expected before this project was initiated. As a result of these findings a "fast track" medication order set was developed to expedite initiation of inpatient therapy. Four key areas included in this order set are sputum, IV access, diet, and IV antibiotics. We speculate that time to first drug delivery will decrease and patient satisfaction with improve. The pediatric Intermountain Cystic Fibrosis Center (IMCFC) provides multidisciplinary Cystic Fibrosis (CF) care to 220 pediatric patients from Utah and parts of Idaho, Wyoming, and Nevada. Historically the IMCFC has embraced early intervention. As new therapies have been added to the CF armamentarium it has become necessary to have a method of keeping track of which patients are candidates for various therapies and also which medications have been prescribed, utilized or even should be discontinued. In order to address this quality issue, the IMCFC decided to develop and implement the use of a Medication Tracker (MT). The IMCFC decided to focus on key therapies which were felt to be at risk for variability in prescriptive practice. The five therapies included were: Pulmozyme<®>, TOBI<™>, Hypertonic Saline(HS), Azithromycin and asthma medications (inhaled corticosteroids and/or leukotriene receptor antagonists). Ibuprofen was intially included and later removed from the MT.

Prior to implementing the MT, the IMCFC met to review literature, medication trackers from other centers, and develop consensus among team members. Six versions of the MT have been developed between 1/1/2007-4/30/2007 . Each MT has been tested in a PDSA cycle and revised.

Since initating the MT, 188 patients have been seen at least once. There have been 264 MT Encounters. The MT is used only after the first visit to the IMCFC and reflects changes to established CF care. There have been a total of 76 changes in prescribed therapies in 66 different patients(35% of patients seen). A minority of patients were receiving therapies that did not meet our pre-determined consensus for prescription. The most common therapies in this category were TOBI™ and Azithromycin. Primary reasons cited for prescribing were: chronic suppressive antibiotic use, Pseudomonas positive and <6 years old, and significant changes on chest imaging. Reasons patients were not prescribed clinically indicated therapies were: patient refusal, participation in clinical trials, cost, and lack of compliance with other therapies. Changes in asthma therapy included the discontinuation of medication in 17 encounters and the initiation of therapy in 8.

The use of a specific CF Patient MT allows consistent practice in selection of therapies. It can be used by all team members and helps to identify discrepancies in therapy. Assuring all patients receive information and are considered for new therapies is increasingly important as our treatment options expand. The MT also highlights the need to consider discontinuing therapies. We speculate the MT will become an increasingly valuable tool over time and improve patient care. Rationale: Recent reports document increasing prevalence of antimicrobial resistance across a range of pathogens in patients with CF. The increase in multidrug resistance of these organisms complicates the therapeutic management of these patients.

Objective: To describe patterns of antibiotic use among physicians treating CF patients.

Methods: For the original purpose of evaluating site characteristics that act as effect modifiers on the relative efficacy or cost of novel therapies, we administered a survey to acquire information on practice patterns and characteristics of physicians participating in a clinical study of an investigational drug for patients with cystic fibrosis. Chi-square and Spearman rank test were used to evaluate associations between physician characteristics and patterns of reported antibiotic use.

Results: Fifty-nine physicians, representing 59 study sites, completed the survey. Ninety percent reported a medical specialty of pulmonology and 98% reported practicing at a teaching hospital. On average, respondents reported that about half of their practice consisted of patients with mild CF (defined by FEV1 ≥ 75% of predicted normal.). Forty-two percent of respondents reported prescribing oral antibiotics and 61% of respondents reported prescribing inhaled antibiotics for maintenance therapy for at least 50% of their mild CF patients. Eight-five percent and 53% of respondents reported prescribing IV antibiotics for pulmonary exacerbation and pulmonary 'clean out', respectively, with about a third of these cases being prescribed for outpatient IV use. Physicians that prescribed IV antibiotics for pulmonary clean out were more likely to pre-scribe oral (p= .04) and inhaled antibiotics (p=.02) for maintenance therapy than physicians that did not prescribe IV antibiotics for pulmonary clean out.

Conclusions: Antibiotic use for maintenance therapy and IV antibiotic use for pulmonary exacerbation and pulmonary clean out acted as 'complements' rather than 'substitutes' as there was a group of physicians who reported less use of antibiotics. Our findings warrant further research as well as additional exploratory analyses with a larger and broader sample to examine associations between physician characteristics and patterns of antibiotic use. Understanding the relationship between physician characteristics and antibiotic use has the potential to influence the approach to management of CF patients in an environment of increasing antimicrobial resistance. Introduction: Hypertonic saline (HS) has been shown to increase mucus clearance and improve lung function in CF, and is now increasingly being used as part of routine therapy. One of the goals of our CF Center quality improvement (QI) initiative is to increase use of HS among all eligible patients.

Objectives: To assess adherence to treatment with HS, parental opinions on treatment, perception of side effects, and the effect of initiating HS on the use of other nebulized medications in pediatric CF patients started on HS therapy during 2006.

Methods: 86 patients under 18 years of age were identified as initiating HS therapy during 2006. A telephone survey of their parents using a 10-item multiple-option questionnaire was administered in April 2007 as part of our CF center QI initiative.

Results: 56 parents (65%) completed the survey: 39 (70%) patients were still using HS. Of those still reporting use of HS, 16 (41%) were using it twice a day; 20 (51%) once a day; and 6 less often than daily. Among the 17 patients who discontinued HS, 8 (47%) stopped following recommendations from their pulmonologist; 5 (29%) stopped due to side effects; 3 (18%) felt HS took too long to administer; and 3 (18%) felt HS solution was too difficult to mix. The most common side effect reported was cough in 33 patients (59%); 19 (34%) reported no side effects. The majority of patients (64%) spent 10-30 minutes per day on HS therapy. Thirteen (23%) patients discontinued another nebulized medication when they initiated HS; 11 (85%) of those stopped Pulmozyme. Reasons for stopping Pulmozyme included: time to nebulize both medications was too long (54%), perception that HS was working better for their symptoms (23%); and doctor's instructions (23%). Overall, 33 (59%) parents felt HS improved their child's symptoms; 16 (29%) saw no difference; and 2 (4%) believed HS made symptoms worse. There was a significant difference in the proportion of patients that reported improvement of their symptoms in the group using HS twice a day (15/16, 93%) when compared to those using it once a day (11/20, 55%) (p=0.022).

Discussion: The majority of patients continued to report the use of HS therapy several months after our initial intervention, with the most parents reporting improvement in their child's symptoms. Physician recommendation was the most common reason cited for discontinuation of HS therapy. Further study of physician opinions on HS therapy, particularly the timing of initiation and discontinuation, is warranted. One-half of patients reported using HS only once a day; however those reporting twice daily use were more likely to report subjective improvement of symptoms. Since treatment complexity in CF increasing, determining the effectiveness of once daily HS therapy, either alone, in place of, or in combination with other therapies such as Pulmozyme is essential. Long-term follow up of these HS-initiated patients will focus on effects on exacerbations, pulmonary function, and hospitalizations. The UAB/Children's Hospital (CHS) CF Center participated in Learning and Leadership Collaborative II. As part of the QI journey, the CHS CF QI team thoroughly evaluated the current system of care in an effort to identify areas of needed improvement.

One area assessed was clinic attendance. It was discovered that 21% of patients failed to attend clinic appointments every week. According to CFF care guidelines, patients with CF should be seen a minimum of 4 times each year at the CF center. The CFF has also found that those centers that see their patients more frequently have improved outcomes for FEV1 and nutritional status.

Aim To decrease the number of missed appointments at the CHS CF clinic.

The CHS CF center administered a QI survey to identify the reasons patients and families were missing appointments. This survey was a 7 item questionnaire that evaluated: reasons for missed appointments, if the families rescheduled missed appointments, difficulties with rescheduling, the need for reminder calls, distance traveled to the CF center, and insurance status.

Results 59 completed surveys were collected. 41% of the families reported Medicaid as their insurance provider. 74% of families live more than 50 miles away from the CF center. 27% of the families confirmed that they had missed an appointment in CF clinic over the last 12 months. The reasons for the missed appointments were widely variable; however, 15.8% of those who missed appointments reported that they did not have access to transportation to get to the appointment. Overall, families did not report difficulties in rescheduling appointments. 68% of those that missed appointments identified Medicaid as their insurance which can be correlated with lower income families and 87.5% of those that missed lived greater than 50 miles away from the CF center. 83% of families reported that a reminder call would be helpful for them.

Interventions After evaluation of the survey data, the center quickly initiated reminder calls for CF clinics. A student employed by the CF Center made these telephone calls to the patients scheduled for clinic every week. The majority of the time the student caller left a message for the family. With the implementation of this simple intervention, missed appointments have decreased from 21% to 14% which is a 30% improvement rate. Reminder calls are now a routine part of the CF Center's system of care.

Other ongoing interventions include increased involvement of primary care doctors for patients lost to follow up for more than 6 months and a system for quickly rescheduling patients.

Next Steps In addition to reminder calls, the CF team will now determine where patients who routinely miss appointments live. If there are pockets of patients in common areas of the state, the team will attempt to maximize services for those patients such as linking them with community transportation resources, community health advocates, or providing satellite clinics. This will be accomplished through geomapping of the Alabama CF population based on zip code. We designed a scoring system to uniformly identify patients experiencing a pulmonary exacerbation. The elements of this pul-monary exacerbation score (PES) have been previously described. At the time this project began the median FEV 1 % predicted of the CF patients at our Center 6-18 years old was below the national average (2004 CF Registry data). Our hypothesis was that greater uniformity of identification of pulmonary exacerbations, would improve pulmonary function in the pediatric age group at our center. This abstract describes continued improvement in the pulmonary function of this age group at our center now that the PES has become standardized in our clinical practice.

Methods: Beginning in October 2004, a PES was calculated for all patients presenting to our center age 6-18 years of age. Any patient with a PES of > 5 was defined as having a pulmonary exacerbation, and treatment with antibiotics was recommended. The patient's CF physician dictated the ultimate decision for treatment and specific course of treatment. Median FEV 1 of all patients age 6-18 years was calculated quarterly to determine the effect of the PES use on pulmonary function. In October 2006 the PES was incorporated into our CF clinic medical record and its use was standardized. The use of the PES, the adherence to the recommendation for treatment, and quarterly FEV 1 of the population has continued to be monitored.

Results: From 10/2004 to 9/2006 (the duration of the original QI project), 1067 patient visits occurred, with a PES calculated on 968 (utilization of 86.7%). 327 patients were treated for a pulmonary exacerbation of the 357 patients with a PES of > 5 (adherence of 91.6 %). 64.9% of exacerbations were treated with oral and/or inhaled antibiotics, and 35.1% were treated with intravenous antibiotics. Since standardization of the PES in 10/2006, 348 visits have occured. Use of the score has decreased slightly (83.4% utilization), however adherence to the score has increased (92.1 % of PES > 5 have been treated). Use of oral vs. IV antibiotics has been similar (69.5% oral vs. 30.5% IV). The median FEV 1 of our patient population has improved from an average of 83.7 % predicted during the 24 months before implementing the PES to an average of 88.5% predicted during the ensuing 24 months of the original QI project (a 5.7 % improvement). Since standardization of the PES in 10/2006, the average median FEV 1 of the population has continued to increase to 90.7% predicted (an 8.4% increase from the baseline period). This improvement in median FEV 1 is reflected in our CF Registry data.

Summary: Implementation of a Pulmonary Exacerbation Score, designed to encourage uniform identification of pulmonary exacerbations in patients with CF, is associated with an improvement in median FEV1 in children 6-18 years of age. Standardization of the PES has resulted in continued improvement in pulmonary function of this population at our CF Center.

This work has been supported in part by the CFF and Akron Children's Hospital. Our CF Family Council parent and patient advisory group has been involved in the development and implementation of this project. Utilizing data from the 2005 CFF patient registry, we determined that our center has had nutritional outcomes consistently below the national average for pediatric and near the average for our adult patients. As a result, we developed the global aim to improve and maintain optimal growth and nutrition in all patients at our CF Center. Our specific aims included: 1)to record growth parameters at all outpatient visits; 2)to identify patients at nutrition risk, categorizing them based on anthropometric data; 3) to develop and implement a specific nutrition plan based on nutritional category; and 4) to educate staff, patients,and families on the benefits of normal growth and nutrition.

We surveyed our patients to determine their overall attitudes toward nutrition and nutritional care received at our center. We discovered that many patients and families did not feel that nutrition was addressed at every CF visit nor that they had a nutritional plan at each encounter. This led to the development of an intake form/home-going sheet for our patients to complete that included data elements necessary for assessment of nutrition and that clearly outlined a nutrition action plan.

A standardized Nutritional Assessment Score (NAS) was devised that incorporated weight, height/length, BMI and weight for length percentiles as well as weight loss, failure to gain or maintain, and failure to remain in the expected growth channel. The NAS was developed after examining the medical literature, reviewing work done by other QI teams, and acquiring input from our pediatric gastroenterologist and endocrinologist. We then developed nutritional algorithms to address nutritional intake, malabsorption, evaluation for CFRD, and short stature. These four algorithms were used to develop a nutrition action plan for each individual.

Beginning in March 2007, a nutritional category (optimal/acceptable, concerning, at risk or failure) was assigned to each patient at each visit based on their NAS. An individualized nutrition action plan was developed with each patient.

At the time of project implementation, patients age 2-20 at our center had a median BMI %ile of 52.9% with 54.5% of our patients with a BMI <50th%ile. Our patients > 20 years had a median BMI of 22.1 with 52.5% having a BMI<50th%ile. After one month of implementation of the NAS, 60% of our patients had been categorized with 13% of our patients having individual nutrition action plan developed. To track our progress, we will monitor median BMI, % of patients with BMI <50%ile, and % of our patients in each nutritional category quarterly using PortCF. Our goal is to have all our patients assessed and a nutrition action plan implemented by the end of 2007.

We have developed a Nutritional Assessment Score and nutrition treatment algorithms for pediatric and adult patients with CF as part of a QI initiative to improve nutritional status of patients at our center.

The project has been supported, in part, by Akron Children's Hospital. We are indebted to Lori Lundquist, a parent of two of our CF patients, who has been an integral part of our nutriton QI team.

In Germany there are about 8,000 CF patients who are usually treated in special centres by a team of trained and experienced health care professionals. The result of this structured approach to patient care, including frequent clinical assessments, monitoring, and aggressive interventions is an improved mean survival and a higher quality of life of pediatric and adult CF patients. However, a recent survey showed that German reimbursement schemes cover only about 50 % of resources used within these centres.

Objective

To assess and evaluate direct costs of outpatient treatment of CF patients in Germany. A micro-costing approach was used to record resource use data directly within representative CF centres. Results of the evaluation may be used to initiate and support discussions between health care providers and insurances about adequate reimbursement schemes in Germany.

Outpatient care was evaluated in seven different centres for pediatric and adult CF patients. Patient reported outcomes, clinical patient data, resource utilisation, and physician related time consumption for treatment was recorded for every patient during one representative month in 2006. Cost data for staff, materials, medical equipment, rooms, etc. were assessed within the respective centres.

Data for 326 CF patients were collected. About half of them were children and adolescents, and half of them were adults. A comparison of the resource uses to the actual remuneration of these services indicates that only about 50 % of the costs are covered by the reimbursement system. Correlation analysis identified significant cost drivers like the age of the patient, comorbidities like pancreatic insufficiency and hepatobiliary complications, lung function (% FEV1), and the presence of certain pathogens in the lungs.

As the actual reimbursement covers only about 50 % of the resource usage costs in Germany, the existence of the CF centres and sufficient treatment for CF patients is uncertain in the future. In order to ensure the existence of the centres it would be necessary to agree to a cost-covering reimbursement at minimum. This may be based on a lump-sum payment that could be differentiated for pediatric and adult centres or according to comorbidities or clinical parameters. The data calculated in this study could be used to trigger and support discussions between health care providers and insurances about a new cost-covering reimbursement system for the out-patient treatment of CF patients in Germany. The impact of the new US Cystic Fibrosis Foundation (CFF) nutrition practice guidelines, i.e., discontinuing the use of percentage of ideal body weight (%IBW) to define "nutritional failure" and setting body mass index (BMI) below the 50th percentile to define "nutritional risk", on evaluating nutritional care practices is unclear.

Methods: Data from 14702 children reported to the 2002 CFF Patient Registry were analyzed to compare the rates of malnutrition among 113 CF centers.

Results: Nationally, eliminating %IBW < 90% as a criterion for underweight resulted in a 6.2% reduction (from 33.1% to 26.9%) in "nutritional failure" rate. Misclassification of "nutritional failure" by %IBW < 90% ranged 1.2-15.6% among 113 centers and was greater for centers having a larger proportion of tall patients. One-third of centers switched to a different tertile ranking, after correcting for misclassification by %IBW. Use of BMI < 50th percentile led to the classification of 56.8% of patients as "nutritional risk". More than half of the centers had different tertile rankings on "nutritional risk" compared to "nutritional failure". A total of 5.1% (0-12.5% among 113 centers) of patients who had height < 5th percentile but BMI ≥ 50 percentile were not considered at nutritional risk. The Cystic Fibrosis Questionnaire-Revised (CFQR) is a disease specific quality of life measure that is currently undergoing clinical validation and may prove to be a useful adjunct to provider assessments. We proposed the routine use of the CFQR in a busy adult cystic fibrosis (CF) clinic would assist in identifying patients in need of limited clinic resources. This performance improvement project focused on new and infrequently-seen patients (i.e. fewer than three outpatient clinic visits/year) to determine the frequency of patients with domain scores that fall more than one standard deviation below our clinic mean.

Methods: The CFQR is routinely self-administered to all adult CF patients in clinic at least once every year. These assessments occur after vital signs are obtained and prior to any other clinical assessments. Patients were deemed to be at baseline if there were no changes made to their pharmacologic or chest physiotherapy regimens. The mean and standard deviation of each domain of baseline CFQR assessments were determined. The frequency of rarely seen and new patients with domain scores less than one standard deviation below the mean was determined.

Results (202). Of these 84 baseline assessments, forty five were obtained from patients who were rarely seen or new to the clinic. 48.9% (n=22) of these patients had one or more domain scores falling more than one standard deviation below the mean. 35.5% (n=16) of these patients had two or more domain scores below these cutoffs. 20% (n=9) of the rarely evaluated patients had domain composite scores (i.e. the sum of the individual domains) that fell below the standard deviation cutoff. The most common outlying domains in these infrequently seen or new patients were Body (10), Weight (10), Physical (7), Health perceptions (7) and the Composite Domain (9).

Conclusions: The CFQR is easy to administer and score during routine adult CF clinical visits. As many as 35% of our infrequently evaluated patients had multiple scores that were more than one standard deviation below our clinic mean. We conclude that the CFQR is a useful adjunctive measure to provider assessments and can help focus limited personnel resources. Aminoglycoside antibiotics are known to be toxic to the inner ear. However, they continue to have a leading role in the treatment of certain infections and in the treatment of pulmonary exacerbations in patients having cystic fibrosis (CF). While there is ample evidence in the literature of ototoxicity, and there are established protocols for monitoring the effects of toxic agents on hearing, less is known about the prevalence of vestibular toxicity among patients who are exposed, and there are no commonly accepted protocols for monitoring vestibular system function.

Oscillopsia and unsteadiness when standing and walking are the two most commonly reported and disturbing symptoms associated with bilateral loss of vestibular system function. Oscillopsia is the perception that viewed stationary objects or surroundings move coincident with head movement. When it is severe, oscillopsia can prevent an individual from having clear vision with even the slightest head movement. The unsteadiness that accompanies vestibular loss can range from mild to disabling.

Because of the referral of a number of patients to the Vestibular Testing Center (VTC) at the University of Michigan in whom severe bilateral vestibular loss was evident secondary to apparent aminoglycoside toxicity and our concern about quality of life issues, we initiated a quality improvement clinical protocol with our CF patients. Vestibular system function and hearing are evaluated at the time of admission for pulmonary exacerbations and coincident with the three month follow-up visit with the pulmonologist. Our goal is to understand the incidence of ototoxicity in this patient population, and to determine how best to measure incremental changes over time. The ultimate desired outcome is to investigate the efficacy of protective agents for limiting the toxic impact of these drugs on hearing and vestibular function.

To date, we have completed vestibular testing on 30 patients with CF ranging in age from 10-56 years, some of whom were referred prior to the initiation of this monitoring program. Of the sample, 4 individuals have completely normal vestibular test results, and 2 of the 4 had no prior aminoglycoside exposure. Of the remaining patients, 10 have bilateral vestibular loss ranging from mild to severe, two have evidence of a unilateral vestibular loss, and each of the others has non-lateralizing evidence of vestibular involvement. We have also documented change in the three individuals we have seen for repeat testing. Specifically, repeat testing has shown incremental decline in vestibular function with repeated exposure to aminoglycosides. Interestingly, only 6 individuals in the group have documented hearing loss, and although most report that they do not and have never had any problems with dizziness or balance, many patients have evidence of oscillopsia or abnormal postural control test results. It is clear from our limited data that it is important to monitor vestibular system function whenever potentially toxic agents are used. While monitoring hearing is also warranted, our data suggest that a monitoring program that includes only hearing is insufficient. Background: The correlation between BMI, lung function, and longterm outcomes in cystic fibrosis patients has been well documented. Based on this data the CF Foundation (CFF) has established adult patient care guidelines for BMI of 22 or greater for women and 23 or greater for men. In our CF center implementation of nutrition therapies in CF adult patients has been difficult due to lack of acceptance of proposed interventions. Due to the importance of optimal nutrition to attain the CFF recommendation for BMI, University of Cincinnati Adult CF Center implemented a weight management quality improvement protocol to improve the nutritional status of patients below the CFF recommended BMI. Our aim is to improve to the nutritional outcomes in our center through patients' acceptance of our nutrition interventions.

Strategies: As part of the weight management protocol we developed a nutrition action plan, nutrition algorithm, and weight management education record and learning log. The action plan took a systematic approach to weight management, classifying patients into different risk categories based on BMI. Patients were categorized as mild nutritional insufficiency (BMI 20-21), moderate nutritional insufficiency , and severe nutritional insufficiency . Each category required a specific action to address weight loss or inability to gain weight. Areas of focus for the action plan include; intake assessment through computerized dietary analysis, assessment of pancreatic function, Pancreatic enzyme replacement therapy (PERT) regimen assessment, 72 hour stool collection for malabsorption, assessment of glycemic control, assessment for GERD, and enteral/parental feeding. The dietitian completed a chart review and nutrition assessment with the patients and classified them into one of the three nutritional risk categories. Eleven patients with a BMI below 19 were initially identified. Nine (82%) of the eleven patients with a BMI of 19 or less have been approached with the protocol and algorithm. Patients completed a three day food journal and oral intake was assessed via computerized dietary analysis. Oral enteral supplements were recommended 2-3 times per day. G-tube placement was discussed and an educational material on tube feeding was distributed. The protocol and algorithm guided identification and assessment of contributing factors to weight loss.

Results: Of eighty patients seen in the adult center since the beginning of October, thirty-seven of the patients (46%) have a BMI below 22 and eleven patients (14%) have a BMI below 19. Out of the nine patients approached with the protocol 55% (n= 5) completed and returned the three day food journal. Of the patients that completed their food journal 60% (n= 3) agreed to a PEG placement, one of the patients is still contemplating PEG placement while sorting through some psychosocial issues, and the other patient has had a differential diagnosis that might negate the necessity for a PEG placement.

Conclusion: We developed a systematic approach to weight management within our adult CF program. This systematic approach to weight management, which includes patient involvement and specific education, has improved the overall acceptance of nutrition interventions in our adult population. The inpatient care of adults with cystic fibrosis can be a challenging given complicated medical regimens, specialized respiratory care orders, and unique dietary requirements. A large proportion of these patients are admitted to academic centers with rotating house staff that may have never treated an adult with cystic fibrosis. To improve the quality of care of our adult patients we have implemented a standardized electronic order set, a recurring educational program, and a "pocket card" to help the house staff. The adult cystic fibrosis program at Vanderbilt cares for over 160 patients, with 7-12 inpatient admissions per month. Vanderbilt's inpatient order system is completely electronic. We have devised an order set based on our agreed standard of care for inpatients with pneumonia. The order set guides the house staff through most aspects of inpatient care including culturing, antibiotic choice and dosing, dietary consultation, nutritional recommendations, and respiratory care orders. This order set not only educates the house staff on our standard of care but also provides direct access to each order they may wish to enter. We have reiterated this educational process with a pocket sized handout that guides them through the typical issues in adults with CF as well as provide phone numbers to members of the CF team. These resources are presented in short talk given at the beginning of each month to incoming house staff. A preliminary data analysis suggests that over 90% of all adults with CF are admitted using some component of this order set. An informal survey of the house staff also concludes that the pocket card and lectures have been helpful. All three aspects of this quality improvement have been implemented since December of 2006. We plan to formally test the impact of this intervention when a meaningful number of house staff have been through the service.

* Background: Malnutrition is a common complication of cystic fibrosis (CF). The median BMI percentile in CF patients between the ages of 2 and 20 years is 44.8% but ranges between various centers from 23.6 to 65.9 percent.1 Previous research has shown a trend toward improvement of pulmonary disease and a significant improvement in weight, height, and BMI z-scores after nutritional supplementation via a gastrostomy tube2; furthermore, FEV1 is positively correlated with BMI percentile.1 Higher weight at three years of age predicts better pulmonary function at six years of age.3 The effect of gastrostomy tube feedings on pulmonary exacerbation frequency has not been well defined. Hypothesis: The incidence of pulmonary exacerbations in children will decrease after the initiation of supplemental gastrostomy feedings. Methods: A retrospective chart review of all pediatric patients seen at the CMH CF Center, and who received gastrostomy tubes between 1997 and 2007 was performed. Data was evaluated in six month time intervals for two years prior and two years after gastrostomy tube placement. We assessed weight, height and BMI percentiles, FEV1 percent predicted, number of courses of exacerbation therapy (intravenous and oral) and microbiologic data. Results: We identified 20 patients, 14 females and 6 males, with ages ranging from 11 months to 14 years of age (6.82 years mean) at time of gastrostomy tube insertion. All patients had increases in their weight-forage percentile (mean of all patients of 13.83 percentile pre to 31.23 percentile post, p<0.05), height-for-age percentile (16.78 percentile pre to 27.03 percentile post, p<0.05), and BMI percentiles (24.23 percentile pre to 45.98 percentile post, p<0.05). The mean FEV1 percent predicted declined overall from 85.5% to 78.1% (2.2 percent predicted per year) over four years in 10 patients. There was a trend of exacerbation reduction after gastrostomy tube placement from 153 to 121 total courses (p=0.06). This decline in exacerbations after gastrostomy tube placement was most dramatic in those patients growing Pseudomonas aeruginosa without other organisms, with a decline from 90 total courses to 60 total courses in 10 patients (p=0.05). Discussion: Weight, height and BMI percentile improved after initiation of gastrostomy tube feedings. Lung function (FEV1 % predicted) decreased at the rate predicted in spite of gastrostomy feedings. There is a trend toward decrease in exacerbation frequency after improved nutrition with gastrostomy tube feedings, especially in patients infected with Pseudomonas aeruginosa. Limitations of this study include the small sample size, relatively short follow-up time, and the retrospective nature of the study which did not allow assessment of adherence. Conclusion: Improved nutrition through gastrostomy tube placement may decrease the frequency of pulmonary exacerbations. Evidence of an association of improved pulmonary function with improved nutritional status has motivated aggressive nutritional intervention. Percutaneous endoscopic gastrostomies (PEGs) are employed increasingly in Cystic Fibrosis to optimize nutritional status. We reviewed our center data over the past 15 years to assess changes in our PEG practice and effects on nutritional and pulmonary outcomes. We hypothesized that our change in practice to earlier PEG intervention improved patient's nutritional and pulmonary outcomes.

We conducted a case controlled retrospective study of pediatric CF patients spanning the past 15 years divided into two periods: P1 (PEG placed before 1996) and P2 (PEG placed 1996 or later) compared with their respective age/sex matched controls. We compared PEG vs. control CF patients with respect to age, nutritional and pulmonary status. We compared weight, height and BMI directly preceding PEG placement and at approximately one-year post PEG. We compared available data for best yearly FEV1 and FVC before PEG and at 1, 5 and 10-years post-PEG. We then compared data between the two time periods.

Results: PEG patients in P2 (n=38; mean age 6.6 + 5.1 yrs.) had lower BMI ( z = -1.2 v. -0.2, p<0.001) pre-PEG placement compared to controls. By 1 year post PEG, the BMI normalized and equaled controls (z=-0.4 v. -0.1, p=0.1). PEG patients' FEV1 %ile prior to placement was lower than controls (88 v. 94, p=0.2) as was their FVC %ile (97 v. 103, p=0.2) and remained lower at 1 year post-PEG.

Following PEG placement, absolute BMI increased (0.9kg/m2, p=0.009), as did BMI Z-score (0.6, p=0.01). We did not see significant changes in pulmonary function tests (PFT).

PEG patients in P1, (n= 12; mean age 10.7 + 2.3 yrs.) also had lower BMI (z= -1.6 v. -0.3, p= 0.0001) pre-PEG than controls. At 1 year post-PEG, the PEG patients still had a significantly lower BMI (z = -0.7 v. -0.2, p=0.08). FEV1 was lower in PEG patients compared to controls in pre-, 1year, and 5-year post-PEG data (p<0.05). Variability in PFTs and small sample size limited our analysis.

Pre-PEG weight, height, BMI, FEV1 and FVC were lower in P1 patients compared to the younger P2 patients. At 1-year post-PEG, similar BMI zscores (-0.7, p=0.99) was obtained between the two periods. Variability in PFTs among the patients within each period prevented valid comparisons, although P2 tended to have better PFT values.

Conclusions: PEG placement was effective at improving nutritional status in patients with Cystic Fibrosis. Placement in younger patients with better pre-PEG nutritional and pulmonary status allowed more complete nutritional recovery. We speculate that improved height z-scores with apparently stable FEV1 is consistent with increased pulmonary growth. Demonstration of an attenuated rate of PFT decline to correlate with this earlier intervention will require a larger population and longer observation period. A multi-center longitudinal outcome study would help determine optimal timing and criteria for PEG intervention. 1 1. Pediatric Gastroenterology and Hepatology, University Medical Center Groningen, Groningen, Netherlands; 2. Biochemistery, Erasmus Medical Center, Rotterdam, Netherlands BACKGROUND: Ursodeoxycholic acid (UDCA) treatment is frequently applied for cystic fibrosis-related liver disease (CFLD). It has been hypothesized that UDCA is beneficial through its choleretic activity, i.e. its capacity to induce bile flow. The hepatic expression of the CFTR protein is restricted to the cholangiocytes lining the bile ducts. Cholangiocytic CFTR is involved in the generation of ductular bile flow. It is not known to what extent the choleretic activity of bile salt treatment depends on expression of competent CFTR protein, and whether or not UDCA differs in this respect from other bile salts. We evaluated the role of CFTR in the acute and chronic choleretic effect of bile salt treatment, through comparative studies in Cftr-null, homozygous F508del, and corresponding wild type control (WT) mice.

METHODS: Bile flow was determined after gallbladder cannulation. Bile flow during the first 30 minutes after acute interruption of the enterohepatic circulation was regarded as basal bile flow. After 30 min, taurocholic acid (TCA) or tauroursodeoxycholic acid (TUDCA) were IV administered in stepwise increasing dosages, up to 1200 nmol/min/100g BW, to Cftr-null mice and WT littermates fed a standard chow diet. Other Cftr-null, homozygous F508del mice and their respective WT littermates were fed either the standard chow, or the same diet supplemented with cholic acid (CA, 0.5wt%) for 3 weeks. Finally, Cftr-null mice and WT littermates were fed the standard chow or the same diet supplemented with ursodeoxycholic acid (UDCA, 0.5 wt%) for 3 weeks.

: Upon a regular chow diet, the basal bile flow was similar in Cftr-null and homozygous F508del mice, compared to their respective WT littermates (Cftr-null, 6.3±1.8 vs. 6.2±0.1; and, F508del(∆/∆), 5.4±2.8 vs. 8.5±10.8 µl/min.100 g BW, resp.; NS). IV administration of TCA or TUDCA to Cftr-null mice and WT littermates increased bile flow to similar extents. Dietary CA treatment increased basal bile flow significantly less in Cftr-null and homozygous F508del mice than in their respective WT littermates (Cftr-null, 13.7±3.0 vs. 17.9±1.7, p<0.05; and, F508del(∆/∆), 20.9±5.5 vs. 26.2±6.2 µl/min.100 g BW, p=0.06; resp.). Interestingly, dietary UDCA treatment increased basal bile flow more profoundly than CA treatment, and to similar levels in Cftr-null mice and WT littermates (+~500%, 29.0±5.9 vs. 31.0±4.4 µl/min/100 g BW, resp.; NS).

CONCLUSION: Upon chronic treatment, UDCA is highly choleretic in mice, independently of the presence of functional CFTR. The independence of functional CFTR is UDCA specific, since the choleretic activity of chronic CA treatment is diminished in Cftr-null and homozygous F508del mice. We speculate that this specific, Cftr-independent choleretic effect of chronic UDCA treatment could be therapeutically important for cystic fibrosis. Most cystic fibrosis (CF) patients have mild to moderate focal portal tract changes (ductal obstruction and cholangitis) and hepatosteatosis. Etiology of severe liver disease (LD)with cirrhosis and portal hypertension which occurs in 5-7% of CF patients is unknown. The C57Bl/6J congenic CF mice develop progressive CF-like LD. As a useful therapeutic model, we have shown that dietary addition of docosahexaenoic acid (DHA) significantly ameliorates inflammation but has little effect on ductal obstruction (Am J Physiol Gastrointest Liver Physiol 292: G839-G848, 2007) . We hypothesized that treatment with UDCA,a hydrophilic bile acid will ameliorate portal tract ductal obstruction and have an additive effect with DHA in preventing LD.

Methods: Thirty-day old wild-type and CF littermates are fed Peptamen and either olive oil or DHA, with and without 0.4% UDCA (8 groups,10 mice/group). Mice are killed following 90 days of treatment and bile salts, serum liver enzymes and serum and tissue lipids analyzed. H&E stained liver tissues are coded and assessed blindly by an expert hepatopathologist (JP) for evidence of hepatosteatosis, inflammation, bile duct obstruction, fibrosis and other signs of liver pathology. An arbitrary scale of 0 to 4 is used, with 0 representing no pathology and 4 being the most severe. We present interim statistical analysis performed for pathology results only for untreated mice (WT, n=7; CF, n=6); treated mice with DHA, (WT, n=7; CF, n=6), treated mice with UDCA (WT, n=7; CF, n=7) and treated mice with UDCA and DHA (WT, n=6; CF, n=6).

Results: Mean pathology scores from groups of mice are pooled to examine the effects of genotype, treatment and diet (Table) . At present,the bile,lipid and liver biochemical test data are insufficient to perform statistical analysis. UDCA-treated mice have significantly increased bile duct obstructed scores compared to the untreated mice which subanalyzed (not shown) reveals increased bile duct obstruction in the wild-type littermates and no effect in CF mice. Hepatosteatosis and inflammatory scores in the untreated mice (fed olive oil) are increased compared to the mice treated with DHA.

Summary: We anticipate completing and performing final analysis within six months. However, these preliminary data confirm that DHA reduces inflammation in C57Bl/6J mice livers and suggest that UDCA therapy has an adverse effect in WT mice and is not effective in ameliorating liver disease in CF mice. UDCA with DHA therapy does not appear to have an additive effect in ameliorating CFLD in this mouse model. Supported by Axcan Pharma Inc. BACKGROUND The CF mouse intestine has an innate response associated with small intestinal bacterial overgrowth (SIBO). A previous Affymetrix GeneChip analysis of the CF mouse intestine showed altered expression of several eicosanoid metabolic genes. Eicosanoids are biologically active arachidonic acid metabolites with a variety of pro-and anti-inflammatory actions as well as effects on mucus production, electrolyte transport, and gastrointestinal motility, all of which are of potential importance to the CF phenotype of the intestine.

METHODS Wild type (WT) and CF mouse (cftr tm1UNC ) littermates congenic on the C57Bl/6J background were fed an elemental liquid diet (Peptamen). Intestinal RNA was isolated for quantitative RT-PCR of expression levels of genes involved in eicosanoid metabolism to confirm and extend the GeneChip analysis. Various eicosanoids were measured in small intestinal lavage fluid using specific enzyme immunoassays.

RESULTS We confirmed microarray results by qRT-PCR that the following genes were differentially expressed in the CF small intestine relative to WT: Cyp4a10 (cytochrome P450 4a10; 20-HETE synthesis; 5% of WT); Ltb4dh (leukotriene B 4 dehydrogenase; metabolizes prostaglandins and leukotrienes; 22% of WT); Cyp2c40 (cytochrome P450 2c40; 16-HETE synthesis; 25% of WT); Ltc4s (leukotriene C 4 synthase; LTC 4 synthesis; 25% of WT); Hpgd (hydroxyprostaglandin dehydrogenase 15; metabolizes prostaglandins; 30% of WT); and Pla2g5 (group 5 phospholipase A 2 ; potentiates arachidonic acid generation; 500% of WT). Non-significant or less than 2-fold changes in gene expression were measured for: cyclooxygenase (Cox) genes, PGE synthase, PGI synthase, and 5-and 15-lipoxygenases. As measured by their stable metabolic products, PGE 2 and PGF 2α were significantly increased in the CF mouse intestine (600% and 1190% of WT, respectively). The following eicosanoids were significantly decreased in the CF mice: 12-HETE (20% of WT), 15-HETE (19% of WT) and 20-HETE (4% of WT). There were not significant changes in the levels of PGI 2 , PGD 2 , LTB 4 , cys-LTs, or LXA 4 . The changed levels of eicosanoids are generally consistent with the differential gene expression.

CONCLUSIONS There are significant changes in expression of eicosanoid metabolic genes and certain eicosanoids in the CF mouse small intestine. The increase in Pla2g5 and decrease in Cyp expression levels are expected to make more arachidonic acid available for eicosanoids except HETEs. Decreases in degradative genes (Hpgd, Ltb4dh) will prolong availability of prostaglandins. Altered levels of eicosanoids are expected to play important roles in the pathophysiology of the CF intestine. Elevated PGE 2 may contribute to increased mucus secretion which is characteristic of CF. PGF 2α and PGE 2 regulate intestinal motility and small intestinal transit is slower in CF patients and in CF mice. The HETEs are mostly known for their effects on vascular tone and electrolyte transport in the kidney but are less-well characterized in the intestine. Their dramatic decrease in the CF intestine may affect blood flow, electrolyte transport, intestinal motility, or other GI functions. The specific effects of the altered eicosanoid metabolism in CF remain to be explored. Supported by CFF grant DELISL05G0. For the Scandinavian Study Consortium. BACKGROUND: Progressive pulmonary disease, correlating with IgG levels, is a major cause of morbidity and mortality in CF patients. Recently, immunosuppressive effects of vitamin D and a potential role of vitamin D deficiency in the pathophysiology of autoimmune diseases have been described. We investigated whether vitamin D deficiency could contribute to the chronic proinflammatory state characterizing CF and the higher occurrence of immune-hyperreactivity-related conditions in CF patients.

METHODS: Multiple linear regression analysis, taking the confounding factor of age into account, was used to evaluate the relationship between cross-sectionally determined vitamin D variables and: lung function parameters, inflammatory parameters and immune hyperreactivity markers in 898 patients followed at the CF centers in Sweden, Norway and Denmark. Vitamin D intake was based on a 7-day precoded dietary food record, calculated in the national food databases.

FINDINGS: In the population of all patients included in the study (n=898), blood vitamin D positively correlated with FEV1 (p<0,01) as well as with FVC (p<0,01). Negative correlation was found between blood vitamin D and IgG (p<0,001), supplemented vitamin D per kg BW and IgG (p<0,001), total vitamin D intake per kg BW and IgG (p<0,05). Additionally, lower blood vitamin D levels were related to both presence of pathological OGTT (p<0,05) and occurrence of CFRDM (p<0,05). Correspondingly, food vitamin D (p<0,05), food vitamin D per kg BW (p<0,05), supplemented vitamin D (p<0,05), supplemented vitamin D per kg BW (p<0,001), total vitamin D intake (p<0,01) and total vitamin D intake per kg BW (p<0,01) all correlated negatively with HbA1C values. Most of the studied vitamin D variables were lower in patients with endomysial antibodies positivity (n=14) and higher in patients with ABPA (n=4) than in the rest of CF population studied (ns). Interestingly, in children (n=442) blood vitamin D correlated negatively with transglutaminase levels (p<0,05) and there was also a significant negative correlation between HbA1C and vitamin D intake. Surprisingly, in adult CF patients (n=456) IgA positively correlated with total vitamin D intake (p<0.05) and total vitamin D intake per kg BW (p<0.05). In the population of Stockholm CF-center patients (n=165), where it was possible to allow for the influence of both age and compliance, negative correlation between blood vitamin D and IgG (p<0.01) was found. After allowing for age, compliance showed significant correlation only with IgG (p<0.001) and ESR (p<0.05), and with none of the rest of the studied variables.

CONCLUSIONS: Vitamin D deficiency might contribute to the continuous shift towards inflammation as well as to the higher prevalence of DM and celiac disease in CF patients. New recommendations for vitamin D supplementation and monitoring of blood vitamin D levels are needed, in order to improve immunological balance and decrease the need for anti-inflammatory therapy in CF patients.

Supported by Swedish Heart Lung Foundation, Stiftelsen Frimurare-Barnhuset i Stockholm, Norwegian and Swedish Cystic Fibrosis Associations, and by an unrestricted grant from Solvay Pharma.

von Drygalski, A.; Biller, J.A. Medical College of Wisconsin, Milwaukee, WI, USA Anemia is associated with increased morbidity and mortality in other chronic diseases, but little is known about anemia in CF. Recognition and correction of anemia in CF might lead to better outcomes. Chronic inflammation, iron deficiency and malabsorption may be important mechanisms contributing to the anemia in CF patients. We hypothesized that anemia might be correlated with poor lung function. METHODS: Clinical charts and portCF.org visit logs of 218 patients of all ages (1 m to 61 yrs) with CF were reviewed for as many years as charts permitted (range 1-7 yrs). The following data were extracted: CBC, iron studies, pulmonary function tests (PFTs), vitamins A, D, and E levels, creatinine, pertinent medical history, and current medications. Anemia was defined by age-and gender-specific World Health Organization (WHO) criteria. Patients were considered anemic if low hemoglobin(Hb) was present on two separate occasions at least two months apart, or average annual Hb met WHO criteria. PFTs in 176 patients (all >6 yrs) included forced expiratory volume (FEV1) and forced vital capacity (FVC) as percent predicted of normal and were considered a reflection of patient performance. The most representative annual PFT from at least 4 annual tests was chosen for analysis since acute infection adversely influences daily performance. RESULTS: 61 of 218 patients were anemic (prevalence 28%). Prevalence of anemia increased with age from 11.5% in patients < age 18 to 58% in patients > age 40 and was significantly higher in patients with vitamin deficiencies. 90% vitamin-deficient patients vs. 59.5% non-vitamin deficient patients were anemic (p=0.02). Mean Hb was 10.1 mg/dl (range 6.6-12.9). Roughly two thirds of patients had moderate and severe anemia (Hb <11 mg/dl) with no gender difference in regard to prevalence or severity identified. Complete iron studies were available in 16/48 patients. 7 were identified as iron deficiency anemia (IDA), 4 as anemia of chronic illness(ACI), and 3 as combined; 2 others had renal failure. In 32 patients, iron studies were incomplete, but renal failure, hemoptysis, hematochezia and solid organ transplants were contributing factors in half. PFTs obtained in 176 patients ≥ 6 yrs were compared in anemic and non-anemic patients. Mean FEV1 and FVC at all ages were statistically significantly poorer in the anemic patients (p <0.005). Complete iron studies were also available in 9 non-anemic patients with impaired lung function. 8/9 patients had ferritins <35 ng/ml suggesting relative iron deficiency. CONCLUSIONS: The prevalence of anemia in CF patients is high and increases with age. In patients for whom iron studies were available, IDA was the prominent underlying mechanism, followed by ACI or a combination of both. In addition, a group of non-anemic patients with poor lung function had relative iron deficiency. The strikingly significant association between anemia and poor pulmonary function as well as the higher prevalence of anemia in vitamin deficient patients uncover anemia as significant co-morbidity in CF. Identification of iron deficiency in anemic as well as in non-anemic patients with poor lung function could be important since iron supplementation might allow Hb levels to increase as an appropriate response to hypoxia. Oral administration of DHA to a CF mouse model corrected the lipid imbalance and reversed certain pathological manifestations in tissues affected by CF. The aim of this study was to investigate the metabolism of LA through the n-6 pathway, and to determine the effect of DHA supplementation on LA metabolism in cultured cells with a CF phenotype.

Methods: Sense (WT) and antisense (CF) CFTR 16HBE cells were cultured in MEM containing 10% horse serum. Cells were supplemented with various concentrations of LA (ranging from 25 to 200 µM) and DHA (10, 20, and 40 µM) for one week. Cellular lipids were extracted from confluent monolayers, and fatty acids were methylated and analyzed by GC-MS. Metabolism through the n-6 pathway was studied by incubating the cells with 4.6 µM [ 14 C] LA for 4 hours, and quantitating its downstream metabolites by HPLC.

Results: The levels of LA and DHA were significantly decreased in CF cells compared to WT cells (LA in WT: 10.8±0.2, CF: 8.2±0.1%, p<0.01; DHA in WT: 2.0±0.2, CF: 0.6±0.1%, p<0.01, mean ± SEM for each). LA supplementation resulted in an increase in LA and AA levels in both WT and CF cells. At most LA concentrations tested, LA levels were significantly lower and AA levels were significantly higher in CF cells, indicating an increased production of AA from LA in CF cells. The study of [ 14 C] LA metabolites showed an increased production of multiple downstream fatty acids including 18:3n-6 (3.1 fold), AA (4.3 fold), and 22:5n-6 (2.7 fold) in CF cells compared to WT cells, further substantiating an increased metabolism through the n-6 pathway. Supplementing 16 HBE cells with DHA for 1 week resulted in a significant increase in LA levels and a corresponding decrease in AA levels in WT and CF cells. When supplemental DHA was added in combination with LA, DHA inhibited the LA-derived increase in AA levels in CF cells and normalized them to the WT values. The formation of [ 14 C] LA metabolites was inhibited after 24-hour supplementation with DHA. This inhibitory effect of DHA was greatest on AA production, and it was more prominent in CF than in WT cells (WT: 3.7 fold, CF: 12.4 fold decrease in AA production).

Conclusions: Decreased LA levels in CF are at least in part related to increased LA metabolism through the n-6 pathway. The fact that DHA supplementation decreased the flux from LA to fatty acid metabolites in the n-6 pathway, especially to AA, may partially explain DHA's mechanism of therapeutic benefit. These data would also suggest that DHA combined with LA in the diet may normalize fatty acid metabolism in CF patients. Background: Decreased levels of linoleic acid (LA) and docosahexaenoic acid (DHA) have been found in the blood and tissues of CF patients. Studies on cultured CF cells and cftr-/-mice have indicated that decreased LA levels in CF might be related to its increased metabolism through the n-6 pathway. Other potential mechanisms of the observed decreased LA and DHA levels in CF could be related to altered cellular uptake of these two essential fatty acids. The goal of this study was to investigate the uptake of LA and DHA into cultured CF cells, and their distribution among the major lipid classes.

Methods: Sense (WT) and antisense (CF) CFTR 16HBE cells were cultured in MEM containing 10% horse serum. The uptake of LA and DHA into the cells was studied by incubating the cells with 0.45 µM [ 14 C] LA or [ 14 14C] DHA for 1 hour and 4 hours, followed by measurement of radioactivity in the supernatant and cell lysate. Lipid fractions from the cells (total phospholipids, triglycerides, cholesteryl esters, and free fatty acids) were separated by thin layer chromatography (TLC), and the fatty acid composition was determined by GC-MS after methylation. Incorporation of LA and DHA into cellular lipid fractions was determined by incubating the cells with 4.5 µM [ 14 C] LA or 9.1 µM [ 14 C] DHA for 1 hour and 4 hours, followed by TLC separation of lipids and measurement of radioactivity associated with the various fractions. (All data are mean±SEM).

Results: The levels of LA and DHA were significantly decreased in CF cells compared to WT cells (LA in WT: 10.8±0.2, CF: 8.2±0.1%, p<0.01; DHA in WT: 2.0±0.2, CF: 0.6±0.1%, p<0.01). Fatty acid composition analysis of lipid fractions indicated that LA levels were significantly decreased in the phospholipid and triglyceride fractions of CF cells (80.9% and 42.3% of the WT values, respectively), and DHA levels were significantly decreased in the phospholipid fraction of CF cells (21.2% of the WT values). The uptake of LA was significantly higher into CF cells compared to WT cells at 1 hour (WT: 14.5±0.1, CF: 35.5±0.9 dpm/µg protein, p<0.01) and 4 hours (WT: 34.7±2.9, CF: 100.8±5.0 dpm/µg protein, p<0.01). The uptake of DHA was also significantly higher in CF cells compared to WT cells at 1 hour (WT: 46.3±8.4, CF: 131.2±8.7 dpm/µg protein, p<0.01) and 4 hours (WT: 140.7±12.9, CF: 276.8±12.0 dpm/µg protein, p<0.01). Incorporation of LA and DHA was increased in the phospholipid, cholesteryl ester, and free fatty acid fractions of CF cells at 4 hours. DHA, but not LA, incorporation was significantly increased in the triglyceride fraction of CF cells.

Conclusions: Low LA and DHA levels in CF cells are associated with a decrease in cellular lipid fractions, the largest of which is in total phospholipids. There is an increased uptake of LA and DHA into CF cells, and into most lipid fractions of these cells. The decreased levels of LA and DHA in CF cells, despite their increased cellular uptake and esterification in lipid classes, indicate that there is increased mobilization of these fatty acids in cultured CF cells.

Perez, A.; Issler, A.; Davis, P.B. Pediatrics, CWRU, Cleveland, OH, USA Even though lung disease continues to be the primary cause of death in Cystic Fibrosis (CF), the considerable increase in life expectancy of the CF patient over the last decades, has made management of the diverse gastrointestinal and nutritional complications in the adult CF increasingly important. Malnutrition and fatty acid deficiency has been a hallmark of CF and its impact in long-term lung health has been established. Data obtained by us using Affymetrix MG-U74Av2 arrays from ileum and liver of pristine 8-week-old male mice bearing the Y122X CF mutation and their normal littermates, indicates that there is a profound disturbance in genes associated with lipid metabolism, many of which are regulated by the peroxisome proliferator-activated receptor-gamma (PPARγ), a ligand-regulated transcription factor that regulates lipid metabolism and possess anti-inflammatory properties. We hypothesize that lack of CFTR function results in altered expression of PPARγ regulated genes in the gut and liver, leading to the fatty acid metabolism abnormalities observed in CF.

PPARγ mRNA and protein levels were measured in ileum, colon, liver, and fat of 8 week old female mice bearing the CF mutation S489X and its WT littermates (C57Bl/6J background), and FABPCFTR mice (gut corrected, mixed background) after the administration of ethanol or Pioglitazone (a PPARγ agonist, 30 mg/kg) by gavage every 24h for 2 days, sacrificed 24h after last dose. Highest PPARγ mRNA levels were found in fat, followed by colon, liver, and lowest in ileum. At basal state, mRNA levels were higher only in FABP ileum vs. WT and FABP colon vs. CF. After treatment, higher expression was seen in CF and FABP ileums and CF liver vs. controls and WT treated mice. However, there were striking differences at the protein level, especially in the colon. At basal state, PPARγ was localized, mainly at the surface of the epithelia in WT, in crypts in CF, and mainly in crypts but extending towards the surface in FABP mice. Upon treatment, PPARγ location in WT colon moved almost exclusively to the crypts, and in the CF and FABP mice, although expression remained mainly in crypts, PPARγ could now be seen at the surface. Changes in PPARγ protein were also seen in ileum and especially liver, between WT and CF, at basal and upon treatment, but changes were subtle and will require more sensitive methods to further examine them. We also examined mRNA and protein levels for several genes involved in fat processing that could be related directly or indirectly to PPARγ. mRNA and protein levels for AQP1 and cubilin, and protein levels for megalin, ApoA-1, and ApoB-48 were reduced in the ileum of CF mice compared to WT.

Our data indicates a PPARγ dysfunction in the gastrointestinal system of CF mice, which could be responsible for the gastrointestinal manifestations observed in CF. Understanding the association between PPARγ, CFTR, and fat processing may help us to develop new therapies directed towards improving the nutritional status of CF patients that, in the long run, will influence their lung health. CF is characterized by low linoleic acid (LA) and docosahexaenoic acid (DHA) levels. Although this could be due to a primary abnormality in fatty acid biosynthesis, this could also be explained by alterations in phosphatidylcholine (PC) formation. PC are formed by two pathways. The most utilized is the de novo conversion of choline to PC via the CDP-choline pathway, which preferentially forms PC containing saturated and monounsaturated fatty acids. In the other pathway phosphatidylethanolamine (PE) is converted to PC through three methylation steps utilizing the enzyme phosphatidylethanolamine N-methyltransferase (PEMT). This pathway preferentially forms PC containing polyunsaturated fatty acids. Thus decreased PEMT activity would be predicted to result in decreased DHA and LA levels. This is supported by results seen in PEMT knockout mice (Watkins et al, J Nutrition, 2003) showing decreased DHA and LA levels, and variable to increased levels of arachidonic acid (AA). Furthermore, Innis et al have shown that CF patients have decreased choline levels making them more dependent on the PEMT pathway. We hypothesized that alterations in fatty acid levels in CF may be due at least in part to decreased PEMT activity.

Methods: Cftr -/mice (CF) and WT littermates were fed with either a diet containing choline (control) or, in order to mimic choline status in humans, an otherwise matched choline deficient (CD) liquid diet for 7 days. Choline levels were measured using a kit. Livers were perfused and microsomes prepared. To measure PEMT activity, microsomes were incubated with radioactive S-adenosylmethionine in the presence of dimethyl PE. The PC product was extracted and radioactivity measured. Fatty acids from liver and pancreas were analyzed by GC/MS.

Results: There was no difference in choline levels in serum or liver comparing WT and CF mice. There was no difference in PEMT activity in livers from CF mice compared to WT on control diet. CF mice on CD diet had decreased PEMT activity in the liver compared to WT mice on CD diet (WT:26145±1893; CF:16515±1240 cpm/mg protein/min; p<0.001 Conclusion: Induction of choline deficiency in CF mice to mimic the situation in CF patients, leads to decreased PEMT activity. This resulted in low LA levels compared to WT mice in both the liver and pancreas and indicates that the fatty acid abnormality in CF is at least in part due to defective PC metabolism through the PEMT pathway. (4995).

Urolithiasis appears to be more common in adult cystic fibrosis (CF) patients than in the general population. A variety of urinary findings have been observed which could account for this, including low volumes, raised oxalate and low citrate, but not all the reports are consistent. Most CF patients have pancreatic insufficiency and a plausible explanation for an elevated oxalate excretion in these patients is malabsorbed fat in the intestinal tract sequestering calcium and permitting more oxalate to be absorbed from their food. If this were the mechanism for the elevated stone risk in CF patients then we would expect to observe differences in stone incidence and oxalate excretion between CF patients with exocrine pancreatic sufficiency (PS) and pancreatic insufficiency (PI).

Methods Diagnosis of CF is by clinical and genetic criteria and PS or PI status is based on faecal elastase measurement. 39 patients and 14 healthy volunteers (HV) provided 24h urines for analysis of volume, pH, Ca, Mg, Na, K, PO 4 , oxalate, citrate, creatinine, urea. Supersaturations were calculated using Equil2.

There is a history of renal stone disease in 20 out of 307 CF patients attending our unit. 286 patients have PI and include 19 of the stone formers. The remaining stone former is amongst the 21 PS patients. No significant differences in urine chemistry or derived supersaturations were found between the CF patients without a history of stones that were PS (n=10) or PI (n=11). Nor were any such differences found when stone forming CF patients were included (PS, n=11; PI, n=18) or when the healthy volunteers were included in the PS group (PS, n=25; PI, n=18).

Amongst the CF patients, only citrate and creatinine excretions were significantly different between stone formers (n=8) and non stone formers (n=21) (medians (mmol/24h), citrate, 1.4 vs. 2.3, p=0.019; creatinine, 9.7 vs. 12.4, p=0.032) . The difference in creatinine excretion can be accounted for by the difference in body mass of these two groups (median (kg) 54.8 vs. 62.8, p=0.045) Compared to the HV, the CF patients with a history of stones had significantly lower citrate excretion and higher calcium oxalate supersaturation (medians, citrate (mmol/24h), 2.9 vs. 1.4, p=0.002; 4.4 vs. 8.1, p=0.037) Conclusions There is no evidence to suggest that the prevalence of stones is different in the PS and PI groups nor could we detect any difference in excretion of oxalate or any other urine variable related to pancreatic sufficiency. Comparing those with and without stones, the only consistent finding was a lower citrate excretion by stone formers. Between the HV and stone formers this was also associated with a significantly reduced calcium oxalate supersaturation. Our evidence does not support the hypothesis that fat malabsorption by the majority of CF patients contributes to an elevated risk of renal stone disease. We undertook assessment of the use of a hypopharyngeal sensor for detection of laryngopharyngeal reflux in infants and children. Gastric reflux in the airway, or supraesophageal reflux, commonly takes a gaseous form that cannot easily be measured using conventional technology. The miniaturized pH sensor at the tip of the Dx-pH Probe is the only sensor able to measure pH in the airway. Fifteen infants and children referred to the Pediatric Pulmonary Center for assessment of chronic cough, hoarse voice, or uncontrolled asthma unresponsive to usual medical intervention underwent 6-48 hour to placement of the Dx-pH probe in supplementation of radiologic assessment of gastresophageal reflux. The probe was placed transnasally and visualzed in the hypopharynx with the red light at the tip. Drops in pH <5 were correlated with food and symptom diaries. Positive results were found in 10/15 patients, leading to surgical intervention in three patients. One patient with cystic fibrosis age 3 months with failure to thrive underwent study for symptoms of cough. Upper GI demonstrated no gross gastroesophageal reflux or anatomic obstruction.Gastric emptying scan was mildly prolonged. Outpatient supraesophageal pH probe demonstrated drops to pH<4 associated with respiratory symptoms. Patient underwent Nissen and G tube placement with resolution at six week follow-up.The patient returned at age six months with onset cough and underwent repeat study. There were no drops in pH and the cough responded to oral steroid burst.

The device was tolerated in 13/15 patients with early removal in one young adult with endstage neuromuscular disease and in an infant with "colic" due to a hysterical parental reaction.As a result of our experience we have incorporated the use of supraesophageal pH monitoring into our practice, with particular attention to infants with persistent asthma and children with histories suggestive of gastric asthma. We have now incorporated it into our Vivometrics LifeShirt, an ambulatory physiologic monitoring device which will allow us to examine supraesophageal pH and its relationship to sleep apnea, cough, tachycardia, tachypnea, and hypoxemia. Aim: Monitoring and adjusting dose requirements of pancreatic enzyme replacement therapy (PERT) are an integral part of the dietetic assessment of patients with CF. We investigated characteristics of enzyme use in our adult clinic population and determine the extent to which inappropriate enzyme use contributed to poor nutritional and clinical state. Method: Information was collected using an annonymous self-administered questionnaire developed to measure patient practice, knowledge and beliefs on PERT. Exclusion criteria included pancreatic sufficiency, <1500 units lipase/kg/d, and FEV1 <30%. Results: 49 out of potential clinic population of 72 patients completed the questionnaire (16-54y, 55% male, FEV1 31-125%). 67% of participants reported to never miss enzymes with meals; this was considerably lower for snacks (35%). Those patients who omit enzymes with meals also missed enzymes with snacks (r2=30%, p<0.001). A more appropriate use of PERT was observed in patients with lower as opposed to higher BMI. Despite intensive dietetic input 29% of patients missed PERT with foods containing fat and 20% took PERT inappropriately with food and drink that did not contain fat. The results identified 5 potentially better practices for measuring PERT behaviour and knowledge: 1) taking PERT with all meals, 2) taking PERT with majority of snacks 3) swallowing capsules intact rather than splitting them open, 4) carrying enzymes around with them and 5) the ability to titrate PERT in accordance to the fat content of food. Patients were scored on the basis of the above criteria to differentiate between prudent enzyme users and those who are compromising therapy. In conjunction with their nutritional status score (BMI and gastrointestinal symptoms) risk for intervention can be assessed. Discussion:Underweight patients have more optimal enzyme use, suggesting greater dietetic involvement in these patients. Schall et al 2006 also found this to be the case in children. The findings emphasised the need for targeted and effective input in patients where problems are less obvious. The questionnaire has been a useful research tool, and has been adapted as a screening tool for dietitians to gain a subjective perspective of patient's enzyme management and identify patients who need support. The combination of patient's PERT usage and their nutritional status could help capture and identify risk objectively and quickly and allows resources to be allocated most effectively. Schall .2), nutritional failure, measured as BMI<20, (p>0.99), or lung transplant status (p>0.3) between patients with and without colonic wall redundancy. There was no significant difference in CFTR function, per sweat chloride values, between groups (p>0.9).

Conclusion: We report a previously undescribed colonic wall abnormality seen on CT in patients with cystic fibrosis: Proximal colonic wall redundancy was found in 39% of adults with cystic fibrosis, but not in children, appeared to be chronic, and was associated with non-∆F508 CFTR gene mutations, particularly the G542X mutation. Recognition of the CT appearances will prevent erroneous diagnosis of acute colonic disease in this patient population, and stimulate investigation of a biologic basis for this finding as a recognizable feature of adult cystic fibrosis. Intro: Infants and children with cystic fibrosis, both pancreatic sufficient and insufficient, are at risk for developing hyponatremia due to an excess loss of salt though skin during sweating. These children should receive daily supplemental sodium. Children without CF require 2-4 mEq/kg of sodium daily; and one could argue children with CF need more. A historical recommendation is 1/8 tsp table salt daily for infants; this contains 11 to 13mEq of sodium based on the density of the table salt. Parents may measure the salt inaccurately, and it may be difficult logistically to distribute the salt throughout the day. If supplementation is given as sodium chloride solutions distributed through pharmacies, as suggested in a recent Consensus Conference, co-pays or full payments may be required. Our aim is to provide a simple, inexpensive, and more accessible method for accurately giving sodium supplementation. Do pre-packaged salt packets contain a precise amount of sodium and thus could be recommended to fulfill the daily requirement of sodium in patients with CF?

Method: Pre-packaged table salt packets were collected with permission from four national fast food restaurants and one commercial salt producer. Contents (sodium chloride) of ten packets from each source were weighed individually. Measurements were made using a scientific scale with the ability to measure up to 1/10,000 of a gram. Weights of sodium chloride from individual packets were converted to mEq sodium. Means, ranges and standard deviations were calculated for the data using Microsoft Office Excel 2003™ program.

See Table 1 Conclusions: There is some variability in the content of pre-packaged salt packets, both between and within brands, though some brands have less variance than others. Measurements showed that packages contained mean values of approximately 11 to 13 mEq of sodium. There is a range in the sodium dose recommendations, and spot urine sodium can be used to determine if patients are sodium depleted. Pre-packaged salt packets lack the precision of pharmacological dosing but are inexpensive, practical, and supply a reasonably predictable dose of sodium. Background: Ten to forty percent of people with CF have vitamin D deficiency secondary to pancreatic insufficiency. Obtaining optimal vitamin D levels (25-OHD) has been a challenge. For vitamin D deficient individuals, current supplementation strategies (50,000 IU weekly for 2 months) have been mostly unsuccessful.

Objectives: To identify individuals at risk for vitamin D deficiency and to evaluate the efficacy of a 2 week repletion course of high dose ergocalciferol in children and young adults with CF.

Methods: As part of a quality improvement initiative, a prospective cohort study was performed from January to April 2007. Phase I included querying our CF practice database to identify patients who were due for annual routine blood tests or who had recently documented levels of 25-OHD less than 30 ng/ml. In phase II, 50,000 IU of daily ergocalciferol was prescribed for a 14 day period as part of routine inpatient or outpatient care. During phase III, a post treatment 25-OHD level was obtained. Baseline subject characteristics were obtained at entry and included age, gender, pubertal status, pancreatic status, and FEV 1 %. Post 25-OHD levels were classified as sub therapeutic (<30 ng/ml), therapeutic (30-50), high therapeutic (50-100), or potentially toxic (>100). A paired t test was performed to evaluate pre and post intervention differences. The impact of age, gender, and lung function on the response to vitamin D supplementation were also analyzed. All values are expressed as mean ± standard deviation.

Results: Eighteen individuals with CF participated in the study. The mean age was 17 years with a range of 6 to 25 years. 66.7% were male, and 100% were pancreatic insufficient requiring pancreatic enzyme replacement. FEV 1 % was 64.9 ± 26.4%. All 18 participants had 25-OHD levels less than 30 ng/dL pre-treatment. 25-OHD levels increased from 20.5 ± 6.2 to 57.8 ± 21.1 (p<0.001). 17 of the 18 participants (94.4%) became therapeutic over the two week interval. An increase of 37.3 ± 22.0 (ng/ml) or 214% was seen in the two week period. No correlation was seen on the extent of increase in 25-OHD and baseline lung function. Pre and peri pubertal individuals had a greater increase in 25-OHD levels than post pubertal individuals (55.5 ± 21.7 vs 28.3 ± 16.3, p<0.05) although the magnitude of change did not reach significance. The degree of response also appeared related to age (r=-0.47). No impact of gender was seen. Seven individuals achieved normal therapeutic values while 10 were in the high therapeutic range. No participants had toxic levels.

Conclusions: The results of this study demonstrate that high dosing of ergocalciferol over a 14 day period is an effective strategy in achieving therapeutic levels of vitamin D. It is unclear whether a pubertal effect on the degree of response exists or if this response is merely age related. Further research is needed to evaluate whether this strategy is able to maintain therapeutic levels after completion of the intervention. Additional studies that monitor compliance and evaluate responders and non-responders are needed. Foundation consensus committee established guidelines for optimizing bone health in individuals with CF, including screening for vitamin D insufficiency and a treatment protocol for achieving and maintaining normal serum vitamin D concentrations. However, since the guidelines were published, there has been additional research assessing various vitamin D treatment regimens and their effect on serum vitamin D concentrations. In addition, vitamin D assay variability has recently been well documented.

The objective of this study was to evaluate variability of vitamin D surveillance methods as well as treatment protocols used to treat low vitamin D concentrations at CF centers.

Methods: A survey was created via Survey Monkey, and was sent to the CF Nutrition listserv, which includes over 200 CF healthcare providers. SPSS version 13.0 was used for statistical analysis.

Results: Fifty-nine listserv members responded to the survey. The majority (72%) of respondents measured 25 OH vitamin D concentrations, and 75% considered patients to be vitamin D deficient when serum concentrations were below 30 ng/ml. However, 27% were monitoring only 1, 25 OH vitamin D or a combination of both concentrations. There was considerable variability in vitamin D supplementation protocols for treating low 25 OH vitamin D concentrations. Only 39% were following the consensus guidelines of 50,000 units of ergocalciferol once/week for ages 5 through adults. One third indicated that they followed consensus guidelines for patients less than age 5; however nearly half of the remaining two thirds were not sure what is prescribed or have not observed low vitamin D concentrations in this age group. Fifty-two percent of respondents were aware of vitamin D assay variability. Despite this knowledge, the majority were unaware of the type of vitamin D assay used to measure serum vitamin D concentrations at their institution.

Conclusions: There is considerable variability in the measurement and treatment of vitamin D concentrations across CF Centers despite guidelines provided via a CFF consensus report. The consensus guidelines for measuring vitamin D concentrations and treatment of vitamin D insufficiency should be revisited based on recent studies and expert opinion. Children with cystic fibrosis (CF) who suffer from sub-optimal growth are offered overnight enteral feeding via gastrostomy tube. This study aimed to assess the nutritional impact of overnight feeding over a 2 year period using a retrospective study design. Data, including height, weight, FEV1, pancreatic function were collated for all patients with CF who had a gastrostomy tube placed in the past 4 years. Data were collected at baseline and then every 6 months for 2 years post-gastrostomy insertion. Data for height, weight and body mass index (BMI) were converted into Z scores using Centre for Disease Control 2000 reference ranges and the LMS method (1, 2) . A total of 21 patients (7.7% of the clinic population) were identified as having a gastrostomy tube. Two were no longer receiving feeds and there were incomplete data for 6 subjects. On placement of the tubes, there were substantial deficits in nutritional status with mean Z scores for height -0.9 (SD 1.02), weight -1.43 (SD 0.71) and BMI -1.1 (SD 0.58), with compromised lung function (mean FEV1 69%, range 44-86). Using a repeated measures ANOVA with FEV1 as a co-variate, there were no significant differences in weight, height or BMI from baseline to 2 years (at 2 years post placement: mean Z scores for height -1.1 (SD 0.59), weight -1.36 (SD 0.56) and BMI -0.86 (SD 0.75). FEV1 remained stable over time. These results indicate that gastrostomy feeding can potentially halt the decline in nutritional status that is a feature of CF. However, patient expectations that over night enteral feeding will lead to an increase in nutritional status need to be sensitively managed. Background : Cystic Fibrosis (CF) is the most common life threatening autosomal recessive disorder in Caucasians.Following a landmark paper by Corrie et al a high fat,high calorie diet has promoted a normal growth pattern.Improved nutritional status together with prevention and early treatment of respiratory infections has contributed to improved survival.

Objective: The aim of this study was to assess the relationship between lean body mass and disease severity in children with CF.

Methods: The nutritional status and body composition of 108 children with CF was measured using a Harpenden stadiometer, calibrated electronic scales and a GE Lunar Prodigy densitometer. The following indices were calculated; body mass index (BMI), fat mass (expressed as logarithm % Total Body Fat (ln%TBF) and fat free mass (FFM). Body composition data were expressed as Z-scores using Dutch reference values. Correlations were performed between indices of nutrition (BMI, Height, FFM, TBF) Z scores and markers of disease severity (percent predicted FEV 1 and Shwachman Kulczycki (SK) scores).The nutritional status of these children was also compared with 154 healthy controls. Statistical analyses were performed using Microsoft Excel 2000.

Results: There was a weak but significant correlation between SK scores and height Z score and FFM (r =0.41, p<0.0001 & r =0.54, p= <0.0001 respectively). There was also a weak but significant correlation between % predicted FEV 1 and BMI and FFM (r=0.39, p <0.0001, r= 0.49, p<0.0001). Height and weight Z scores were significantly lower in children with CF (-0.42 and -0.59 respectively) than in control subjects (0.25 and 0.41 respectively) with p <0.0001 in both groups.

Conclusion: This study demonstrates that there is an association between indices of nutrition and disease severity in children with cystic fibrosis. Muscle mass (FFM), assessed by Dual Energy X-ray absorptiometry (DXA) correlates with lung function and with SK scores. The strongest correlation between markers of disease severity (SK scores and % predicted FEV 1 ) was with FFM. FFM may be expected to be associated with respiratory function tests.However,SK scores are a more general assessment of well being and less obviously directly related to muscle mass. FFM and growth may reflect long term nutrition and health in contrast to BMI and ln%TBF which may reflect acute deterioration or short term nutritional intervention. Therefore FFM may prove to be a useful tool to assess quality of nutrition and may be predictive of respiratory and general decline. Body composition assessment has an important role in chronic conditions such as cystic fibrosis, in order to identify nutritional depletion and evaluate nutritional interventions. It is important that reliable and accurate methods are used. The usefulness of non-invasive methods, such as skinfold thickness measurements (SFT) to measure change in body composition has not been evaluated in CF. This study aimed to compare changes in fat-free mass (FFM) and fat mass measured using SFT and dual-energy X-ray absorptiometry (DXA) in adults with CF.

Methods: 57 adults with CF (60% male, 88% pancreatic insufficient, mean age at baseline 30.3 (SD 7.9) years, mean FEV 1 at baseline 63.2 (SD 21.3) % predicted) were studied. They underwent body composition assessment using DXA and SFT, at baseline and a mean of 3.6 (SD 0.4) years later. Durnin and Womersley equations were used to estimate FFM and fat mass from SFT. Estimates of change in FFM and fat mass obtained using SFT were compared with DXA using paired t-tests, univariate analysis and Bland and Altman analysis.

Results: At baseline, mean FFM was 49.5 (SD 9.0) kg; mean fat mass was 10.8 (SD 5.8) kg and mean BMI was 21.2 (SD 2.4) kg/m 2 . Mean FFM, fat mass and BMI were not significantly different at follow-up, indicating no change in nutritional status for the population overall. Individual change in FFM ranged from -4.3 to +5.9 kg, while change in fat mass ranged from -4.2 to +5.8kg. The table shows mean change (∆) in FFM and fat mass by each method, correlations (R 2 ) between SFT and DXA, and the 95% limits of agreements (LOA) between SFT and DXA for change in FFM and fat mass. Mean change in FFM and fat mass estimated using SFT did not differ significantly from DXA. Mean bias between the methods was small, and overall correlations between SFT and DXA were strong for changes in both FFM, suggesting good agreement for the overall population. However the 95% LOA between the two methods were wide for change in both FFM and fat mass. This suggests that SFT will not accurately predict changes in body composition in all CF patients.

Conclusion: SFT measurements do not accurately estimate change in FFM or fat mass over time in individual adult CF patients compared with DXA. Body composition changes measured using SFT tended to more closely reflect changes detected by DXA in male patients compared with females. Caution should be exercised when interpreting the results of serial measurements of body composition using SFT in individuals with CF.

analysed were significantly correlated with AGE-CML or sRAGE levels (age, gender, BMI, presence of CF-related diabetes mellitus (CFRD) or HbA1c level).

Conclusions: Serum levels of AGEs are elevated in adults with CF. Elevation is not restricted to those with CFRD. The levels of sRAGE reflect ability to respond to this pathway and are associated with poorer lung function. Modification of the diet in CF may reduce this mediator and may have the potential to modify lung and renal injury. Mouse models of CF generated by expression of improperly processed CFTR (∆F508) or lacking CFTR expression (KO) demonstrate selective induction of SULT1E1 in the liver (Falany et al., Biochem. J. 364:115, 2002) . SULT1E1 is responsible for the sulfation and inactivation of β-estradiol (E2) at physiological concentrations. The increase in SULT1E1 expression is specific to the hepatocyte, whereas CFTR is selectively expressed in cholangiocytes. The induction of SULT1E1 activity is associated with changes in levels of E2regulated proteins in CFTR (-/-) liver (Li and Falany, J Cystic Fibrosis 6:23, 2006) . Increased SULT1E1 activity is correlated with low body weight and decreased IGF-1 message levels in livers of CFTR(-/-). The mechanism for the induction of SULT1E1 in hepatocytes by cholanigiocytes lacking CFTR expression, and the mechanism for the inhibition of IGF-1 expression by increasing SULT1E1 expression was investigated in human cells. Human MMNK-1 cholangiocytes in which CFTR expression was inhibited by short interfering RNA (siRNA) induced SULT1E1 expression in human HepG2 hepatocytes when cocultured in a permeable membrane separated system. SULT1A1 expression was not altered in this model. The data suggests that loss of CFTR function in cholangiocytes is capable of selectively modulating SULT1E1 expression in hepatocytes. To investigate the ability of increased SULT1E1 activity to modulate expression of IGF-1, HepG2 cells were stably transformed with SULT1E1/pcDNA3 to activity levels intermediate to the levels observed in CFTR(-/-) mice. The increased SULT1E1 activity was associated with decreased IGF-1 message expression in the HepG2 cells. Since a major pathway for IGF-1 regulation involves growth hormone (GH) stimulation of STAT5b phosphorylation, the effects of E2 and increased SULT1E1 activity on GH stimulated STAT5b phosphorylation were examined. STAT5b activation was identified using immunoblot analysis with a rabbit anti-tyrosine phosphorylated-STAT5b antibody. Treatment of control HepG2 with 10 nM E2 prior to the addition of GH increases STAT5b phosphorylation. In HepG2 cells expressing increased SULT1E1 activity, the stimulation of STAT5b phosphorylation by 10 nM E2 was significantly decreased. E2 had an apparent rapid action on STAT5b phosphorylation that is not attenuated by the estrogen receptor antagonist, ICI 182,780. E2 was effective at increasing STAT5b phosphorylation when applied to HepG2 cells 15-30 min before GH. Increased STAT5b phosphorylation was observed in control HepG2 cells at 1 nM E2 although 10-fold higher E2 concentrations were required in the SULT1E1-HepG2 cells consistent with the increased E2 sulfation. No differences were observed in total STAT5b in any of the studies. This study demonstrates that human cholangiocytes with low levels of functional CFTR are capable of inducing SULT1E1 expression in human hepatocytes and the increase in E2 sulfation inhibits the GH stimulation of IGF-1 expression via a decrease in STAT5b phosphorylation. Supported by grants from the Cystic Fibrosis Research, Inc. and NIH (GM38953).

Williams, J.E. 1 ; Benden, C. 2 ; Jaffe, A. 2 ; Suri, R. 2 ; Wells, J.C. 1 ; Fewtrell, M.S. 2 1. MRC Childhood Nutrition Unit, Institute of Child Health, London, United Kingdom; 2. Portex Respiratory Medicine Unit, Great Ormond Street Hospital, London, United Kingdom Background: Patients with cystic fibrosis (CF) are at high risk of malnutrition. To our knowledge, no study so far has employed a reference fourcomponent model(sup)1(/sup)(4CM) to assess body composition (BC) in CF children, which allows accurate evaluation of both fat mass (FM) and the components of fat-free mass (FFM; mineral, protein, water) and most studies have been cross sectional.

Methods: 53 CF subjects, aged 8-12 yrs, were compared with age-and sex-matched healthy controls for assessment of BC using a reference 4CM. Comparison between groups was performed using paired t-tests and general linear models to adjust for age, height and pubertal status. In addition FM index (FMI; FM/height2) and FFMI (FFM/height2) standard deviation scores (sds) were calculated using measurements performed in a reference population of 415 healthy children aged 4 to 19 yrs. Repeat measurements were made in 31 children with CF (15 boys) after 2 yrs and the change in FMI sds and FFMI sds investigated.

Results: For the initial measurement at baseline, boys with CF (n=26) were significantly shorter (mean (se) height sds: -0.7 (0.20, p<0.01) compared to UK 1990 reference data; girls with CF were lighter, BMI sds (-0.6 (0.2), p<0.05). At baseline, comparison of CF children with pair-matched controls indicated that there was no difference in the boys but CF girls had less FM (CF minus control) (-2.6 (0.8)kg, p<0.01) and mineral mass (-0.1 (0.0)kg, p<0.05) after adjustment for age, height and puberty. FMI sds adjusted for age and puberty was also significantly lower in girls with CF (-0.7 (0.3), p<0.01) whereas FFMI sds was not. For comparison with the large reference population there was no difference in the boys at either time point for FMI sds and FFMI sds (paired t-test). For girls however, there was a significant difference at both baseline FMI sds (-0.8 (1.0), p<0.01), FFMI sds (-0.8 (1.1), p<0.05) and 2yrs FMI sds (-0.9 (1.3), p<0.05) and FFMI sds (-0.7 (1.0), p<0.05). Paired t-test analysis between baseline and 2 yr followup measurements indicated that FMI sds and FFMI sds were not significantly different at each point in time in either boys and girls.

Conclusion: At baseline, the BC of CF boys appears to be within normal range, but the CF girls already have lower FM and mineral than their healthy pair matched controls, even when adjustment for size and puberty is made. Follow-up measures in 31 CF children indicated that these descriptions of nutritional status of CF boys and girls remained consistent over the following 2 years, with no evidence of any further deterioration in girls. However, nutritional surveillance is important in both sexes in CF at around puberty to address potential deterioration and this study suggests that the 4CM is a useful tool for this purpose.

1Fuller Background: Although lung function (LF) in children with cystic fibrosis (CF) has been reported to be related to their nutritional status, particularly body fat, the latter has usually been assumed from anthropometric measurements and not measured using an appropriate reference method.

Habal, H. 1 ; AL-Turkmani, M. 1 ; Freedman, S. 2 ; Laposata, M. 1 1. Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA; 2. Medicine, BIDMC and Harvard Medical School, Boston, MA, USA BACKGROUND: Increased dietary intake of fatty acids is recommended for CF patients. It is unknown if this results in increased incorporation of all fatty acids-saturated, monounsaturated, and polyunsaturated-into cells with a CF phenotype to a greater extent than into non-CF cells. Docosahexaenoic acid (DHA) supplementation of cftr-/-mice has been shown to reverse the pathologic CF phenotype.

AIM: To investigate whether the uptake of multiple predominant fatty acids and the release of membrane-bound arachidonic acid (AA) are altered in cultured epithelial cells with a CF phenotype; in addition, to determine whether fatty acid uptake is selectively inhibited by DHA supplementation.

METHODS: Sense (WT) and antisense (CF) CFTR 16HBE cells were cultured in MEM containing 10% horse serum. AA release from membrane phospholipids (PLs) was studied by incubating the cells with 2 µM [ 14 C] AA for 15 min, and then with serum free medium containing 0.2% fatty acid free human serum albumin in the presence or absence of 100 µM adenosine for 15 min, followed by measurement of radioactivity in the supernatant. The uptake of fatty acids into the cells was studied by incubating the cells for different time points with 0.04 µM [ 14 C] linoleic acid (LA), [ 14 C] AA, [ 14 C] DHA, [ 14 C] palmitic acid or [ 14 C] oleic acid, in the presence and absence of 10 µM or 20 µM DHA, followed by measurement of radioactivity in the cell lysate.

RESULTS: Basal and adenosine-stimulated AA release from membrane PLs was significantly higher in CF cells compared to WT cells (basal WT:113.7±1.1, CF:202.4±12.8 dpm/10 6 cells, p<0.01; stimulated WT:207.8±11.4, CF:468.7±49.1 dpm/10 6 cells, p<0.01; mean±SEM for each). The uptake of LA, AA, DHA, palmitic acid and oleic acid were significantly higher into CF cells compared to WT cells. DHA significantly decreased the elevated uptake of these fatty acids into CF cells, but not into WT cells. Data are shown in the table below.

CONCLUSION: In this cell culture model, these data demonstrate that AA uptake and release are increased in CF cells, which may suggest elevated mobilization of this fatty acid by CF cells. Our data also show that CFTR dysfunction leads to increased uptake of all tested fatty acids into the cells. Downregulation of this increased uptake may be one mechanism by which DHA exerts its therapeutic benefits. Low BMI in CF correlates directly with worsening lung function and is an independent risk factor for mortality. The causes for this decreased BMI are incompletely understood but could be related to levels of the orexigenic peptide ghrelin, and to levels of leptin, an anorexigen. Studies assessing plasma levels of leptin in CF have shown contradictory results, while the data on ghrelin in CF are scarce. We assessed both leptin and ghrelin plasma levels in mild (FEV1 > 75% predicted), moderate (FEV1 45% to 74% predicted), and severe (FEV1 < 45 % predicted) adult CF patients. We compared these levels in CF subjects to 20 age matched controls with normal BMI (range 19.5 to 24.5 kg/m2).

We conclude that leptin levels are decreased, while ghrelin levels are increased only in CF patients with severe disease and low BMI. Plasma levels of ghrelin and leptin did not differ from normals in those with mild and moderate disease. Gastroesophageal reflux (GER) is common in CF before and after lung transplant (LTx). Laparascopic fundoplication (LapF) is used when patients fail on conservative medical therapy. Aims: To review longer term clinical outcomes of LapF in patients with CF in terms of lung function, Body Mass Index (BMI), reflux symptoms, patient satisfaction with surgery and complications. Method: Retrospective review of patient records and patient questionnaire to gather relevant information. Results:Thirty-two patients with CF underwent Lap.F recently; 15 transplanted (LTxGroup), 8 female (mean age 40 [21-67]) years, and 17 not transplanted (CF Group), 14 female (35 ). The median time between LTx & LapF was 654 (414; 1133) days, range 145-1838. Endoscopy was undertaken in 27 patients prior to surgery and all underwent 24hr pH monitoring. Complete (Nissen) LapF was undertaken in 50% patients and a partial (Toupet) in the other 50%. Total median time in the operating theatre was 143(81-232) minutes. The average length of stay in hospital (in days) in the LTxGroup: pre-op. was 7.81; post-op. was 7.82 days; and in the CF Group was: 5.2 and 12.5 respectively. Lung function (FEV1 and FVC in percent predicted) and BMI were measured at minus(-)100, plus (+)100, +200 and +300 days from LapF. Patient satisfaction with LapF was a mean of 95% in the LTx Group and 85% in the CF Group and improvement in reflux symptoms of 90% in both groups. Nocturnal cough reduced markedly in the CF Group and there was an overall improvement in quality of life in both groups. Complications consisted of 2 port site hernias, 1 failed LapF with re-operation 20 months later; 2 esophageal dilatations and 1 patient with lower esophageal sphincter hypotonia. There were 3 deaths on Day 19, 253 and 382 post fundoplication. One was in a non-transplanted patient with recurrent episodes of hemoptysis. She succumbed to a catastrophic hemoptysis episode not related to LapF. The other 2 were in transplanted patients. Conclusions: In this relatively small series of patients there was a small but significant drop in FEV1 in the LTx Group 300 days after LapF, but not in the CF Group who remained stable throughout the review period. There was a significant drop in BMI in both groups after LapF, but the CF Group returned to pre LapF values at +300 days. Patient satisfaction with the procedure was high. Fundoplication is an expensibe complicated procedure not without risk. A long term prospective study is warranted. Historically this phenomenon has been attributed to intestinal malabsorption and cachexia as a consequence of chronic lung infection. However, CF mice exhibit similar growth and weight deficits but are not afflicted with malabsorption or lung infection, suggesting other mechanisms are participating as well. We hypothesize that small size is in part a result of increased β-oxidation in adipose stores. To test this we utilized a mouse model to study lipid metabolism. Our study includes female CF mice with a null mutation in the Cftr gene. These mice are congenic on the C57BL/6J background as well as age matched with their controls (WT) to 50-70 days. Liver, adipose and skeletal muscle (calf) were dissected from WT and CF mice in a fed state. Tissues were immediately flash frozen in liquid nitrogen and used for RNA and protein lysate harvest. Quantitative PCR (qPCR) was performed using primers for the lipogenic isoform of acetyl-CoA carboxylase 1α (ACC1α) in the liver and adipocytes, whereas the non-lipogenic isoform ACC2β was assessed in the skeletal muscle. Primers for acyl CoA oxidase (AOX), carnitine palmitoyltransferase 1 (CPT1) and long acyl-CoA dehydrogenase (LCAD) were used as gene markers for β-oxidation steps in the peroxisome, carnitine shuttle and mitochondrion, respectively. Western Blots were performed to detect ACC (active) and phospho-ACC (inactive) forms. The results show that there is a decrease in the total amount of ACC protein in the fat as well as the skeletal muscle of the CF mice and the ratio of ACC:phospho-ACC is substantially lower in the CF tissues. Phosphorylation and/or decreased expression of ACC results in decreased expression of malonyl CoA, leading to a cascade of events that increase β-oxidation. In addition, increased expression of AOX and LCAD, two downstream β-oxidation genes, is also observed. Hyperactivity of the β-oxidation pathway could account for the decreased adipose stores seen in CF mice compared to the WT. Such patterns of expression are reminiscent of starvation. Caloric intake and intestinal absorption of dietary lipid were measured and CF animals consumed approximately 110% the calories of WT animals (per gram body weight) and lipid absorption was not significantly different (94% for CF, 97% for WT), indicating the CF animals have sufficient access to nutrition. We will continue to investigate fatty acid synthesis and degradation pathways during both the fed and fasted states to delineate the mechanism. This research was supported by grant HL 68883. effect of gender on DHA levels in P and RBC with lower values for male CF patients.

Acknowledgments: This work was supported by the Belgian CF Association (ABLM), the French CF Association (VLM) and by a grant from FRS (UCL).

Data are expressed as means ± SD. * and indicate a significant difference with controls and CF PS patients respectively, P<0.022. Background: Deficit of antioxidant systems and increased oxidative stress have been demonstrated in CF patients with respect to healthy controls. The origin of this imbalance is known to be multifactorial.

Objective: The aim of our study was to assess changes in antioxidant systems and oxidative stress parameters in CF pediatric patients in presence or absence of pancreatic insufficiency Material and methods: we recruited 70 patients (37 with pancreatic insufficiency PI and 33 with pancreatic sufficiency PS) attending the Cystic Fibrosis Centre of Turin (2005) (2006) . During the annual routine control visit were performed antropometric measurements, pulmonary function tests (FEV1), and were collected fasting blood samples and last 3 days food record. Vitamin E, A, C, selenium, glutathione (GSH, GSHPx, GSSG), dehydroascorbic acid, PCR were assessed. Data were analysed using the two sample t-test with equal variances.

Results: There are not statistically significant differences between two groups for sex, age and FEV1. 33/37 of PI patients and 8/33 of PS patients took oral vitamin supplementation. Serum levels of alpha-tocopherol and beta-carotene resulted within the local reference range either in PI or PS patients showing no significative differences. The mean serum levels were respectively suboptimal and optimal for vitamin E and A (Biesalsky '97). PI and PS patients did not show significant intergroup differences of water soluble and enzymatic antioxidants levels as well. Serum levels of vitamin C, selenium, erythrocyte reduced glutathione (GSH) and glutathione peroxidase activity (GSHPx) resulted within the local reference range. Plasma vitamin C concentrations resulted slightly inferior to the optimal values (Biesalsky '97) in both groups. Oxidative stress parameters (erythrocyte glutathione disulfide-GSSG, dehydroascorbic acid) and inflammatory parameters (PCR) did not differ significantly, presenting values within the local reference range, in both groups. Nutritional data revealed significantly reduced Zscore for weight (p<0.05) and height (p<0.01) in PI patients vs PS patients. Our data did not evidence significant differences in total energy (kcal/kg) and macronutrients (% of total kcal) intake in both PI and PS patients. Dietary intake of antioxidant vitamins resulted slightly superior the RDA (LARN-1996) presenting no significant intergroup differences.

Conclusions: This study shows that pancreatic insufficiency do not influence oxidant/antioxidant balance in CF patients at least in paediatric age. Until innovative supplementation guidelines will be proposed for application, nutritional education and monitoring and correct dietary habits are useful to optimize antioxidant status in CF patients. There are more than 1400 CF mutations that are responsible for the spectrum of disease severity in cystic fibrosis (CF). The most common, F508del mutation, is associated with pancreatic insufficiency (PI), and a classic phenotype. . Yet even within a group of F508del homozygotes, there is a small minority of patients (2.3%) who are not taking pancreatic supplements, and are therefore presumably pancreatic sufficient (PS) (data from 2002 * CFF Registry). This observation prompted us to look more closely at enzyme use in CF patients with non-F508del homozygous genotype.

We reviewed enzyme use in CF patients who were fully genotyped and whose enzyme use was documented in the 2002 * CF Registry. Heterozygotes with one or two non-F508del mutations were included in this analysis. Using the assumption that the milder mutation is dominant, if the patient was PS and heterozygous with one F508del mutation, the second mutation was classified as PS; if the person was PI, the second mutation was classified as PI. We also reviewed the pancreatic status and genotype of patients participating in the Wisconsin Newborn screening study. In this Randomized Control Trial (RCT), pancreatic status was determined by 72hour fecal fat analysis. Enzyme use was extracted from the database for the year 2002.

There were 14494 patients in the 2002 CFF Registry with a total of 27 different mutations. Twenty-one of these mutations were associated with PI. This finding correlated well with the per cent of patients who were on enzyme supplements (range of per cent patients on enzymes by mutation 100-88%). However when we looked at the six mutations associated with PS, the distribution of patients on enzymes vs. not, was much closer. See table below.

Possible reasons for enzyme supplement use by patients with PS mutations include a) treatment for recurrent pancreatitis, b)the fact that this is a cross-sectional analysis and patient may be on enzymes temporarily, c) patient may truly be PI, d) overuse of enzymes due to unreliable test to measure pancreatic status in patient, e) patients who are on enzymes are older, and have changed from PS to PI phenotype. This analysis suggests a need for better assessment of pancreatic status in patients with CF. * This analysis was originally done by request for the ECFS Conference at Garda on CFTR Genetics, and focused on the suggested "33 main mutations", and will be updated with the 2005 CFF Registry data. (Supported by NIH grants DK072126, DK34108, MO1 RR03186, MO1 RR00058)

Number of patients in RCT with each mutation: 7,3,1,1,3,0 respectively.

Dutta, A. 1 ; Woo, K. 1 ; Fitz, J.G. 2 ; Feranchak, A.P. 1 1. Department of Pediatrics, UT Southwestern Medical Center/Children's Medical Center, Dallas, TX, USA; 2. Medicine, UT Southwestern, Dallas, TX, USA Extracellular ATP is an important signaling molecule contributing to bile formation by the liver through binding biliary epithelial cell (cholangiocyte) membrane P2 receptors and stimulating Cl-efflux. Importantly, ATP appears to work through Cl-channels unrelated to CFTR, suggesting it may be a potential way to modulate bile flow in CF liver disease. However, the signaling pathways linking receptor binding to channel activation in cholangiocytes are unknown. Consequently, the aim of the present study was to identify the pathways responsible for ATP-stimulated increases in [Ca2+]i and membrane Clpermeability in a human biliary epithelial cell model. Studies were performed in Mz-cha-1 human biliary cells; [Ca2+]i was measured by Fura-2 and membrane Cl-currents by patch-clamp techniques. Results: Exposure of cells to ATP (100 µM) resulted in a rapid increase in [Ca2+]i to 1054.23 ± 244.90 nM (n=6) which was abolished by prior depletion of intracellular Ca2+ by thapsigargin (37.94 ± 14.50 nM; n= 7), but unaffected by removal of extracellular Ca2+ (EGTA, 819.00 ± 211.94 nM; n= 7). In parallel studies, ATP (50 µM) increased current density from -1.99 ± 05 pA/pF to -13.20 ± 2.00 pA/pF (n= 19) . Currents reversed at 0 mV (ECl-), were outwardly rectifying, and were inhibited by the Cl-channel inhibitor NPPB (-1.81 ± 1.71 pA/pF, n= 6), but not the CFTR inhibitor CFTRinh172. ATP failed to activate currents after depletion of intracellular Ca2+ stores by BAPTA-AM (-0.70 ± 0.72 pA/pF, n=3). The P2Y receptor antagonist suramin inhibited ATP-stimulated increases in [Ca2+]i (50.04 ± 8.17 nM; n= 7) and Cl-channel activity (-2.25 ± 0.80 pA/pF; n= 6). However, the P2X receptor antagonist brilliant blue G did not affect the magnitude of ATP-stimulated Cl-currents (-14.63 ± 1.81 pA/pF, n=4). Together these findings suggest that ATP activates Cl-currents primarily through P2Y, but not P2X, receptors in human biliary cells. Since P2Y are G-proteincoupled receptors and modulate [Ca2+]i by phospholipase C (PLC) generation of inositol 1,4,5-trisphosphate (IP3), a series of experiments were designed to directly assess the effects of IP3 receptor inhibition or stimulation on ATPstimulated Cl-currents. First, the PLC inhibitor, U73122 blocked ATP-stimulated Cl-currents (0.86 ± 0.60 pA/pF; n= 3). Second, the specific IP3 receptor blocker 2-APB, inhibited both the ATP-stimulated increase in [Ca2+]i (34.36 ± 7.33 nM, n= 4) as well as membrane Cl-currents (-0.13 ± 0.55 pA/pF, n= 6). Lastly, intracellular dialysis with purified IP3 during whole-cell patch clamp activated Cl-currents with identical properties to those activated by ATP (-9.40 ± 1.23 pA/pF, n=4). Together these studies demonstrate that extracellular ATP stimulates Ca2+-activated Cl-channels in cholangiocytes primarily through P2Y receptor binding and PLC/IP3-dependent release of intracellular Ca2+ stores. Thus, the P2Y-IP3 receptor signaling complex represents a potential target to increase cholangiocyte secretion, and hence augment bile flow, in the treatment of the cholestatic liver disease associated with CF.

Supported by grants from the NIH NIDDK and the CFF to A. Feranchak. Background: Supplemental PEP therapy is used to treat CF patients with EPI to assist with digestion, food absorption. This open-label, multipledose, single-treatment, multicenter study evaluated efficacy and safety of EUR-1008 for malabsorption treatment in young patients with CF and EPI.

EUR-1008 is a novel, zero-overfill, highly-stable formulation of porcinederived pancreatic enzymes, and was specifically developed for use in very young children allowing the product to be sprinkled on food. Methods: Eligible patients were < 7 years with CF and EPI, in acceptable nutritional status, needing PEPs and clinically stable. Patients consumed a standard CF recommended diet, could not take drugs affecting gastric pH or motility. The trial involved a 7-day dose stabilization period and a 7-day treatment period. Patients switched from their baseline enzyme treatment without washout. The optimal dose of EUR-1008, determined during dose stabilization, was used during treatment. Primary endpoint compared malabsorption of fat after EUR-1008 vs. previous treatment. Clinical symptoms were also evaluated, as was physicians' and parents' judgment on the control of malabsorption signs and symptoms. Safety measures included AEs, physical exams, vital signs and changes in clinical laboratory findings, including cholesterol levels and fat-soluble vitamins. Results: Nineteen patients (12 male) enrolled and completed all study phases. Mean age was 3.9 years (range 1-6). Percentage of responders (patients without steatorrhea (<30% fecal fat content) and without signs and symptoms of malabsorption after 1 and 2 weeks of treatment) was 52.6 at baseline, 68.4 (P=0.375 vs. base) after stabilization and 57.9 (P =0.999 vs. base) at study end. Frequency and oily stools showed a significant decrease vs. baseline at study end. The mean number of stools per day was 1.82 stools at screening and 1.45 stools during treatment (P<0.001). The mean proportion of stool samples with visible oil or grease at screening was 11.10% and 4.73% during treatment (P=0.001). A statistically significant improvement was seen in the incidence of moderate bloating. Physicians characterized 37% of patients as "improved" in the control of EPI signs and symptoms vs. previous therapies and 63% as "unchanged" at the end of the treatment phase, while parent assessments were 47% "improved" and 53% "unchanged." Vitamin K absorption (measured through PIVKA) improved with EUR-1008 treatment over prior PEP treatment measured at baseline. EUR-1008 was safe and well tolerated, with minimal AEs. No AEs led to discontinuation of study drug interruption. No incidences of uric acid toxicity or fibrosing colonopathy following EUR-1008 treatment. No trends were seen in lab parameters, physical examination, or vital signs. Conclusion: EUR-1008 is effective, safe, well-tolerated in the treatment of EPI in young CF patients. Treatment controlled malabsorption and clinical symptoms of EPI in young patients with CF. Safety and efficacy of EUR-1008 in this trial is consistent with the pivotal phase III trial. There is an association between the presence of CF and alterations in fatty acid metabolism, consisting of a decrease in linoleate (18:2n-6). More recently, a deficiency of docosahexaenoate (22:6n-3, DHA) has arisen as a prominent fatty acid abnormality in CF.

An established breeding colony of exon 10 CFTR knockout (cftr-/-, CF, n=14) and wild type (WT, n=7) mice was used for this study. All mice were weaned at 23 days of age and then raised on Peptamen and water until 30 days of age, and then continued for 10 days with 15 mL/day of Peptamen to homogenize the nutritional status. After this period, some mice (n=3 per group) were subjected to DHA treatment (40 mg/day) for one week. After sacrifice, pancreatic tissue was collected, homogenized and extracted. Lipid classes were separated by sequential elution in aminopropyl columns. The eluted fractions were transmethylated and injected into a Hewlett-Packard 5890A gas chromatograph with a flame ionization detector for quantitative analysis.

The vast majority of fatty acids were mainly distributed in phospholipids, except those with 18-carbons. WT and CF mice were studied A gender gap in the survival of cystic fibrosis (CF) is well-documented. However, little is known whether there is a gender différence in pubertal growth pattern in CF children. Delayed and attenuated pubertal growth are commonly observed in adolescents with CF. However, characterizing height velocity (HV) pattern is very challenging because of errors associated with calculating HV, e.g. interpolation and extrapolation between 2 heights measured in irregular intervals, as well as difficulties in determining the location and the magnitude of peak height velocity (PHV). In this study, we applied a new statistical method, namely, the semi-parametric shape-invariant model, to characterize the HV pattern of CF children. This method is based on the assumption that all individuals' growth curves have a common shape. Therefore, this method attempts to shift and stretch (squeeze) individual's growth data until they can be modeled by the common shape curve. The semi-parametric shape-invariant model has the advantage of fitting measured heights directly, thereby eliminating the errors associated with calculating HV.

The study population included 996 CF children who had height measurements between ages 2 to 18 years documented in the 1986-2002 CFF Registry. Because fitting each individual height curve is computing-intensive, 100 boys and 100 girls were randomly chosen for this preliminary analysis. Our results showed that girls exhibited a typical growth deceleration during late preadolescence, followed by accelerated HV and a single PHV during adolescence (Figure) . Unexpectedly, CF boys exhibited a notable HV peak during late preadolescence in addition to the typical adolescent PHV (Figure) . Overall, the mean magnitude and age of PHV was 7.4 cm at 14.6 years for boys and 6.6 cm at 11.9 years for girls. When compared to PHV of non-CF children based on Tanner's reference (Pediatr. 1985, 107:317) , CF boys showed a greater delay in the age of PHV while CF girls showed a greater attenuation in the magnitude of PHV. Further analyses are being performed to characterize the HV patterns of all 996 children to confirm these results. (Supported by NIH-DK02891).

Anelli, M.; Foresti, R.; Peloso, L.; Ortenzi, G. Medical Affairs, Eurand, Pessano con Bo, Italy Background: Given their inherent instability, the dose units of all pancreatic enzyme preparations (PEPs) have always been "overfilled" to compensate for enzyme degradation over time and assure at least 90% of the label lipase content at the end of their shelf life. This overfill issue was first examined by Whitehead who reported overfills ranging from 14 to 39% after analyzing several commercially available PEPs (Whitehead AM. Pharm Weekbl Sci. 1988) . The USP presently allows for PEP overfills up to 65%. Thus, the actual active lipase content of a PEP capsule labelled at 10,000 IU might vary from 9,000 to 16,500 units according to its "freshness." As a result, patients taking PEPs receive a product with variable potency, which can lead to efficacy issues, increased pill burden, unnecessary drug interactions, and product switching. The FDA has also noted the potential safety risk posed by overfilling and, in a recently published Guidance, it imposed a zero overfill requirement for all PEPs to be marketed in the USA after April 2008.

Methods: Over a period of two years, we ran a series of tests similar to those conducted by Whitehead to evaluate overfill on 16 commercial PEPs currently available in the U.S. and Europe.

Results: No finished product was formulated to 100% of the labelclaimed lipase enzyme activity. Overfill on commercially available PEPs ranged from 7% to 47%, with a median value of 32%. The PEPs analyzed were 10 to 12 months old on average; therefore, the actual amount of overfill is likely underestimated.

Conclusion: Our findings confirm those reported earlier by Whitehead and demonstrate that none of the currently available PEPs complies with the zero overfill requirement of the FDA Guidance. These findings highlight a potential cause of suboptimal therapy with currently available PEPs.

References: -Whitehead AM. Study to compare the enzyme activity, acid resistance and dissolution characteristics of currently available pancreatic enzyme preparations. Pharm Weekbl Sci. 1988 Feb 19; 10(1): 197-201. -Food and Drug Administration. Guidance for Industry: Exocrine Pancreatic Insufficiency Drug Products -Submitting NDAs. April 2006.

Heubi, J. 1 ; Boas, S.R. 2 ; Blake, K. 3 ; Nasr, S.Z. 4 ; Woo, M.S. 5 ; Graff, G.R. 6 ; Hardy, K.A. 7 ; Amaro-Galvez, R. 8 ; Latino, M. 9 ; Lee, C. 10 1. Children 's Hospital Medical Center, Cincinnati, OH, USA; 2. Chicago CF Care, Glenview, IL, USA; 3. Nemours, Jacksonville, FL, USA; 4. Michigan U, Ann Arbor, MI, USA; 5. Children's H, Los Angeles, CA, USA; 6. Penn State H, Hershey, PA, USA; 7. Children's H, Oakland, CA, USA; 8. Texas U, Tyler, TX, USA; 9. Eurand, Milan, Italy; 10. Eurand, Vandalia, OH, USA Background: Supplemental PEP therapy helps prevent maldigestion and malabsorption in CF patients. EUR-1008 is a novel, zero-overfill, highly-stable formulation of porcine-derived pancreatic enzymes developed to treat EPI in CF patients. This phase III, randomized, doubleblind, placebo-controlled, two-treatment, crossover, multicenter trial evaluated the efficacy and safety of EUR-1008 to treat EPI in CF patients.

Methods: Patients with confirmed CF and EPI, age ≥7 years, good nutritional status and weight ≤ 70 kg were enrolled. No drugs affecting gastric pH or motility were allowed. After open-label dose titration, patients were randomized to receive EUR-1008 or placebo over a week-long period. Following another open-label normalization, all patients were crossed over to the alternative treatment arm. Primary endpoint was change in coefficient of fat absorption (CFA) following oral administration of EUR-1008 vs. placebo. Secondary endpoints were change in coefficient of nitrogen absorption (CNA), cholesterol, fat soluble vitamins, body weight, BMI and EPI symptoms following oral administration of EUR-1008 vs. placebo. Safety endpoints included AEs, clinical laboratory parameters, physical exams and vital signs.

Results: Thirty-four patients were enrolled (17 female); 32 were evaluated. Mean age was 15.4 years (range 8-23). EUR-1008 treatment was associated with statistically significant (P<0.001) increases in both CFA and CNA vs. placebo. EUR-1008 treatment was associated with decreases in PIVKA II (an indicator of vitamin K status). EPI symptoms consistently improved during EUR-1008 treatment, including a statistically significant reduction in stool frequency and fewer soft and watery stools. Treatment with EUR-1008 improved the signs and symptoms of malabsorption even in the small subset of patients with CFA values greater than 80% under placebo treatment. There were no notable changes in serum lipids, vitamin A and E values, mean weight or BMI in patients who received EUR-1008. EUR-1008 was safe and well-tolerated. There were no unexpected or significant differences in the frequency or type of AEs between EUR-1008 and placebo, and no trends were seen in lab parameters or vital signs. No patient discontinued from the study due to an AE. Two serious AEs were considered unrelated to the study drug and both resolved.

Conclusion: In this randomized, double blind, placebo-controlled study, EUR-1008 was safe, well tolerated and effective in the treatment of CF patients with EPI, with clinically and statistically significant improvements in CFA, CNA, and EPI signs and symptoms in the absence of any concomitant treatment affecting gastrointestinal motility and pH. The therapeutic benefit of EUR-1008 was seen even in patients with very high CFA values (>80%). The relative infrequency of cystic fibrosis related diabetes (CFRD), until recently, means little is known about hypoglycaemia ('hypos') associated with its treatment. Insulin is frequently taken by those who have CFRD, for whom the risk of hypoglycaemia may be high due to dramatic changes in insulin sensitivity that can occur with pulmonary exacerbations and a decreased ability to secrete glucagon. A number of studies involving people with type 1 diabetes mellitus (T1DM) have reported on their experiences of hypoglycaemia, but whether results from such investigations apply to CFRD is unclear. An exploratory investigation was conducted to compare the type, frequency and severity of hypoglycaemic symptoms experienced by patients who had CFRD with patients who had T1DM.

Method A cross sectional study was conducted, involving adults (18-60 years) with T1DM or CFRD, recruited from two hospitals in England. A questionnaire, sent to 265 patients with T1DM and 145 with CFRD, included investigator-developed items about knowledge and frequency of hypoglycaemia and the Edinburgh Hypoglycaemia Scale (EHS), a standardised measure that assesses autonomic (e.g. sweating, hunger) and neuroglycopenic (e.g. confusion, poor co-ordination) symptoms associated with hypoglycaemia. Questionnaire data were entered into SPSS for analysis. Comparisons were conducted using t-tests, Mann Whitney tests or chi square tests as appropriate.

Results Questionnaires were returned by 60 patients with T1DM and 52 with CFRD. Five of those with T1DM were over 60 years of age and, therefore, excluded from analysis. Almost all participants had experienced a 'hypo' (98%). Comparisons between the two groups are reported in table 1.

Conclusion Patients with CFRD reported fewer and tended to have less severe episodes of hypoglycaemia compared to those with T1DM. Fewer neuroglycopenic symptoms in the former could be related to their shorter duration of diabetes or to how CFRD was managed; patients with CFRD were given shorter acting insulin, making them less prone to 'hypos' during the night, and had a high sugar diet to ensure weight maintenance. Differences between the groups could also be related to specific characteristics of these two forms of diabetes. Table 1 : Data from questionnaire insulin resistance. The purpose of this study was to investigate the relationship between PEM and the risk of developing glucose intolerance and/or CFRD and the effect on lung function in children ≤21 years of age whose history suggests PEM. Patients whose diagnosis of CF was suggested by failure to thrive(FTT) and whose growth parameters remain <10% for weight and height for age were identified as having incurred PEM. A retrospective analysis of our CF Registry database was performed. 173 CF patients ≤21 years old were identified. 32 of 173 patients(18.5%) had CFRD. 103(59.5%) of these plotted <10% for weight and height (PEM)and 23(22.3%) were diagnosed with CFRD. Further review of the 103 patients showed 35 whose diagnosis of CF was suggested by FTT and 10 (28.6%) of those 35 had CFRD. Statistical analysis did not show a clinical significance, but a strong relationship is implied. Pulmonary function, FEV1%, was significantly lower in the PEM/CFRD patients (mean FEV1 63.5% ± 24.36%) than the PEM only group, (mean FEV1 88.17% ± 22.14%)(p=0.0001). This review provides further support that patients who have PEM are at increased risk of developing CFRD. A significant risk was not established; however a strong association is suggested. Assessment of a larger cohort of patients may show a stronger relationship. Findings from this study would suggest that further research into the physiological relationship between PEM and CFRD should be initiated. CFRD screening should be considered at a younger age than the current CFF recommended age of >14 years. Diagnosing CFRD/glucose intolerance early may contribute to improve growth parameters, preserve lung function, and improve survival. Based on this review, our center is initiating CFRD screening for all patients ≥ 10 years, and considers screening for all PEM patients ≥6 years. Background: Patients with CF are thought to have a rapid postprandial rise in plasma glucose which may be followed by a delayed and prolonged insulin response and hypoglycemia. We sought to estimate the prevalance, cumulative one year incidence, and factors associated with hypoglycemia during oral glucose tolerance testing in non-diabetic adults with CF. We also compared the prevalence of hypoglycemia to a geographically matched cohort of young adults without CF. Methods: We performed a prospective study of 161 individuals aged over 16 years with CF followed in clinic providing population-based, specialized care in Cambridgeshire, England. We excluded patients with previously-diagnosed diabetes or taking anti-diabetic medications. We evaluated 108 individuals who underwent 75 g 2 hour oral glucose tolerance testing (OGTT) in 2004. Of these, 65 individuals had neither new diabetes nor hypoglycemia in 2004 and were retested using OGTT in 2005. We testing the association between biochemical and anthropometric factors measured in 2004 with the prevalence of hypoglycemia in 2004, and incident hypoglycemia in 2005. Biochemical hypoglycemia was defined as a blood glucose between <4.0 mmol/l (72 mg/dl), but greater or equal to 2.5 mmol/l (45 mg/dl) and severe hypoglycemia as a value lower than 2.5 mmol/l on either the fasting or post glucose challenge value. For comparison, we used data from the Cambridgeshire-based Young Ely Study of 183 individuals age 30 -48 years who underwent 75 g OGTT testing in 1994 -1995 . Results: In 2004 , in patients with CF, the prevalence of fasting hypoglycemia was 5.5%. No patients had fasting severe hypoglycemia. Following glucose challenge, 20.2% and 2.8% of patients had glucose values consistent with hypoglycemia and severe hypoglycemia respectively. There were no diferrences in age, hepatic, pulmonary or renal function, but patients with hypoglycemia at two hours were five times more like to be male (odds ratio 5.1, 95% CI 1.7 -15.6) and more likely to have a higher BMI. The prevalence of a 2 hours value of < 4.0 mmol/l in patients with CF at 22.9% (25/109) was not different than in controls 20.7% (38/183). Of patients who did not have hypoglycemic readings in 2004, at one year, 1 patient developed fasting hypoglycemia and 14 individuals developed post-challenge hypoglycemic for a cumulative incidence 21.5%. Of these 6.2 % (4) had values in the range of severe hypoglycemia. There were no differences between age, sex, BMI, liver, renal, or pulmonary function in those who did and did not develop hypoglycemia. Conclusions: Hypoglycemia appears common in patients with CF but is also common in a healthy population. Male sex and BMI were associated with post load hypoglycemia in patients with CF in cross-sectional analyses only. Further research needs to assess whether these low blood glucose values are associated with symptoms.

It is well recognised that cystic fibrosis related diabetes (CFRD) is a poor prognostic indicator in CF, and therefore early recognition is imperative to allow effective treatment in an attempt to prevent the associated decline in pulmonary function and nutritional status. Although the oral glucose tolerance test (OGTT) is the accepted method of detecting diabetes mellitus in non CF individuals, glucose intolerance and lack of insulin can be variable in CF patients such that the reliability of OGTT in the diagnosis of CFRD has been questioned. Under these circumstances, other methods of monitoring glucose intolerance, such as serial post prandial glucose monitoring (SGM), may be more appropriate. To look at this further, we prospectively compared OGTT with SGM in a group of adult CF patients.

Fourteen patients with no previous history of CFRD ( 127], 9 DF508 homozygous, 9 male) admitted for acute pulmonary exacerbations were evaluated. All were exocrine pancreatic insufficient, and 1 was on long term oral steroids. Eleven patients received 30mg/day oral prednisolone in addition to IV antibiotic therapy for the duration of their admission. OGTT was performed in the standard fashion (ingestion of 1.75g glucose/kg body weight [maximum 75 g] within 5 minutes, after an overnight fast). SGM was performed 2 hours post prandially and before bedtime over the period of the admission. The local ethics committee approved the study and informed consent was obtained from all patients.

Results SGM revealed elevated postprandial blood glucose values in all patients. There were 83 episodes (mean 6 per patient [0 to 18]) with glucose >11.1 mmol/l (frank diabetes, WHO criteria) and 86 episodes (mean 6 per patient [1 to 13] ) with glucose 7.8 to 11.1 mmol/l (impaired glucose tolerance, WHO criteria). However, OGTT revealed impaired glucose tolerance in only one patient (2 hour glucose value 9.8 mmol/l). In the remaining patients, OGTT revealed normal 2 hour glucose values (mean 5.0 mmol/l [range 3.8 to 6.9]). Fasting plasma glucose values were also normal (mean 4.5 mmol/l [3.5 to 5.9]) in all patients.

We have shown a high prevalence of hyperglycaemia in our adult patients admitted for pulmonary exacerbations that was only detected using serial glucose monitoring. This study suggests that SGM may be more sensitive at detecting early CFRD than the OGTT. Further work needs to be carried out to look at the best methods of diagnosing CFRD, to facilitate early treatment and thereby improve the prognosis.

The diagnosis of cystic fibrosis related diabetes (CFRD) is associated with a decline in pulmonary function and nutritional state and is a poor prognostic indicator: the risk of its development increases with age. Furthermore, the development of CFRD is due to a gradual loss of insulin production, and the disease may therefore be preceded by a period of glucose intolerance. Despite this, the incidence of hyperglycaemia in CF patients is unknown. To study this further, we monitored the blood glucose profiles of adult CF patients admitted for the treatment of a pulmonary exacerbation.

We looked at 60 consecutive CF patient admissions for IV antibiotic therapy to our large adult unit. Of these, 28 had established CFRD and were excluded. The remaining 32 patients (mean age 23 years [range 17 to 54], 22 male) formed the study population. None were on long term oral steroids, but 28 received 30mg/day prednisolone as part of their routine exacerbation treatment. All patients underwent blood glucose measurement 2 hours after each meal and before bedtime as part of a standard assessment protocol in our clinic.

Thirty patients (94%, 26/28 on steroids) demonstrated post prandial hyperglycaemia: 14 exhibited values between 7.8 and 11.1 mmol/l (impaired glucose tolerance, WHO criteria) and 16 >11.1 mmol/l (frank diabetes, WHO criteria). There were no significant differences in age, genotype, FEV1, FVC, BMI, or duration of hospital stay between the two groups. Of the hyperglycaemic patients, 28 (93%) had abnormal glucose levels detected before bed time, significantly more than 2 hours post breakfast (23%, p<0.001), post lunch (60%, p<0.002), but not post dinner (80%, p=0.16).

This study has shown a high prevalence of hyperglycaemia in adult CF patients treated for a pulmonary exacerbation who were not otherwise known to suffer from CFRD. Nearly all our patients demonstrated glucose intolerance later in the day, when the recommended screening test for CFRD (the fasting oral glucose tolerance challenge) cannot be used. Post prandial glucose monitoring of such patients during times of stress may therefore be advisable, allowing the introduction of insulin treatment at an earlier stage in the disease process.

Whittaker, L.A.; Tilluckdharry, L.; Christian, R. Medicine, University of Vermont, Burlington, VT, USA OBJECTIVE: Bone disease is a recognized complication of cystic fibrosis (CF). There are currently no guidelines for bone mineral density (BMD) testing or tools for fracture risk assessment in CF adults. We piloted a survey designed to identify CF-specific risk factors for low bone density and fracture.

METHODS: A 45 question survey was completed by patients prior to measurement of BMD at the hip, spine and radius by densitometry (DXA). The survey was comprised of questions on calcium and vitamin D intake, exercise, family history, medications, reproductive health, ultraviolet exposure, CF related diabetes (CFRD), weight loss, severity of CF lung disease and quality of life. Z scores (BMD adjusted for age and gender) were used so that male and female data could be analyzed collectively. Patient perceived health (PPH) was rated by four questions and totaled for the PPH score, which ranged from 4-12 with the highest score indicating the best health. Functional status (FS) was rated by three questions that reflect the degree to which CF interferes with school, work or play. The total FS score ranged from 3-9, with the highest score representing the least interference with life. Questions on PPH and FS were derived from a validated clinical assessment tool, the CF questionnaire (CFQ) (1) . RESULTS: All CF adult patients (n=46, age 18 and older) were offered participation in the study. DXA of total hip and lumbar spine was performed on 39 patients (85%); 30 (65%) of these patients also completed the survey. 51% of patients had low bone density on DXA (Z score <-1). 62 % of male CF adults had low bone density vs 39 % of females. Fracture was reported by 57% of patients who completed the survey (64% female and 50% male). Lower PPH score was associated with lower Z score at the hip and spine. Lower FS score was associated with lower Z score at the hip and spine. Low body mass index was associated with low Z scores at both hip and spine. Patients with BMI below 22 kg/m2 were most likely to have osteopenia.

CONCLUSIONS: Osteopenia and history of fracture were common in our adult CF population regardless of gender. BMI, PPH and FS predicted bone density. Risk stratification by a CF-specific bone health survey may guide bone density screening strategies for CF adults. Prospective studies in CF adults are needed to determine whether a CF-specific survey can predict fracture risk as well as bone density.

References : As life expectancy in cystic fibrosis (CF) increase, osteoporosis has become more prevalent. Vitamin D status is one of many factors that contribute to optimal bone health. Previous data has shown, the majority of CF patients are deficient in Vitamin D despite being prescribed daily CF specific vitamin therapy. Previous research has also shown that the current recommendations for Vitamin D supplementation and repletion provided by the Cystic Fibrosis Foundation Bone Health Consensus (2002) are not sufficient to achieve optimal Vitamin D levels in the majority of patients. Moreover, the recommendations fail to sustain levels in the optimal range for bone health.

Purpose: We demonstrate a successful supplementation algorithm for individualized high dose Vitamin D with continued high dose maintenance therapy which achieves and sustains appropriate levels.

Methods: All patients followed at the pediatric center are prescribed routine CF vitamin supplementation. Vitamin D levels are obtained as part of routine clinical care via mass spectroscopy. If serum vitamin D level is <30 ng/dL, Vitamin D therapy with ergocalciferol is initiated one time per week. Patients < 5 years of age receive 12000 IU/ dose and > 5 years of age receive 50000 IU/dose. At the next routine quarterly visit, Vitamin D levels are re-assessed via mass spectroscopy to determine efficacy of once weekly supplementation. If levels are > 30ng/dL, patients continue high dose supplementation 1 time a week as maintenance therapy. However, if levels remain <30ng/dL, the ergocalciferol is increased by 1 time a week in 3 month increments until adequate levels are obtained. Once adequate serum Vitamin D levels are obtained, patients remain on the level of supplementation needed to achieve levels >30ng/dL as maintenance therapy. All patients are instructed to take vitamin D and their CF vitamins with enzymes and food to optimize absorption.

Results: After implementation of the algorithm, 70.5% of patients, with varying levels of supplementation, had vitamin D levels > 30ng/dL (mean=38.4 ng/dL). Of the patients that achieved goal Vitamin D levels, 58% were able to achieve and sustain adequate vitamin D levels without additional vitamin D supplementation while on standard CF vitamins (mean vitamin D level = 35.9 ng/dL). The remaining patients required chronic maintenance vitamin D: 17% required vitamin D supplementation once per week with a resulting mean Vitamin D level of 45.2ng/dL, 15.9% required maintenance vitamin D supplementation twice per week with a resulting mean vitamin D level of 39.9 ng/dL, and, 3.6% required vitamin D supplementation 3 or more times per week with a resulting mean vitamin D level of 42.3 ng/dL. Another 4% of patients take vitamin D supplements other than ergocalciferol and were not included in this study.

Conclusions: Individualized vitamin D repletion and maintenance therapy as outlined in our algorithm is a successful mechanism to obtain optimal Vitamin D levels in patients with CF. inhaled tobramycin (29%). Sixty-two percent reported using at least 5 different types of oral medications on that day including pancreatic enzymes (85%), oral antibiotics (33%), anti-reflux medication (49%), azithromycin (47%), anti-histamine (29%), and a decongestant (10%). Twenty percent reported taking at least one medication for pain. Forty-nine percent reported performing airway clearance at least twice during the day with 35% using a vest, 18% performing standard chest physiotherapy, and 15% using a flutter or acapella device. There was no difference in the reported number of medications or the reported time needed to complete therapies based on respondent age. Although respondents with severe lung disease (FEV1 <40% predicted) reported a higher median number of therapies (8 vs. 7, p=0.05), the reported time for completing therapies was not associated with FEV1 (p=0.4).

Conclusion: Daily treatment burden and complexity in all adults with CF is high, and is only marginally increased in those with worse pulmonary function. Given this already high load of daily therapies, efforts to assess the treatment burden of new CF therapies are warranted. The impact of this high treatment burden on overall health related quality of life for adults with CF needs to be assessed.

Objective: The Cystic Fibrosis Foundation (CFF) has a network of over 115 accredited care centers throughout the United States. Data is collected from these care centers to help better understand the disease and describe changes in survival, standards of care, and health outcomes. When the CFF was founded in 1955 few children survived to school age. Now the predicted survival extends beyond the mid 30's (CFF, 2005) . With this change has come the need for age appropriate care in many areas including reproduction. The Adult Cystic Fibrosis Center at Morristown Memorial Hospital presently follows 68 adults. Of those, 3 women(4%) with CF have had children affected with cystic fibrosis. Each woman was diagnosed after her child. The CF data registry does not have any means of identifying if a patient has an affected child. As the average age of survival of CF patients has increased, so has the likelihood of CF patients having CF offspring. What is the estimated incidence of CF patients having CF affected children attending accredited CF centers in the United States?

Method: Questionnaires were sent to care centers and satellite centers via standard mail or facsimile based on information from the CFF directory. One month after the last questionnaire was mailed, the CF Foundation sent out a reminder email to the nurse coordinators. Questions asked were as follows: 1) Name of Center 2) Do you presently care for any patients with CF who have given birth to, or fathered a CF affected child?

3) How many patients at your center have had CF affected children? 4) Was the parent diagnosed before or after pregnancy was confirmed? 5) If parent's diagnosis came after conception, was this after child was diagnosed? 6) Was the parent's diagnosis a result of prenatal testing?

Results: A total of 144 questionnaires were mailed or faxed. This netted a result of 50 responses. After email reminder, an additional 68 questionnaires were returned for a total of 118 responses, an 82% response rate. Name of center was checked to remove duplicate information. The number of parents with cystic fibrosis who have had children with CF numbered 62. Of those, 27 (44%) were diagnosed before pregnancy; 35 (56%) were diagnosed after pregnancy. Of those diagnosed after pregnancy, the majority (86%) were diagnosed as a result of their child's diagnosis. Only 3 (8%) were diagnosed as a result of prenatal testing.

Conclusion: Based on the responses from accredited CF care centers, 62 patients with CF have children also affected with this disorder. Prior to conceiving their child, 44% knew that they had CF. Having CF did not appear to be a deterrent to delivering CF offspring. This study did not look at the age of the parent or child; they may have conceived their child before genetic screening was common for spouses. As the median survival of this population increases, the incidence of CF adults with CF children may also increase. The CFF may choose to add a question to the CF data registry, identifying CF patients with CF offspring. The potential physical and psychosocial impact can then be tracked.

Anderson, A.; Popli, K.; Stewart, J.; Heslop, K.; Gascoigne, A.; Bourke, S. Adult Cystic Fibrosis Centre and Fertility Centre, Royal Victoria Infirmary, Newcastle upon Tyne, United Kingdom As the outcome of patients with CF improves fertility issues become increasingly important. Almost all men with CF are infertile because of azoospermia due to congenital absence of the vas deferens. These men can attain genetic paternity by sperm aspiration from the epididymis or testis with intracytoplasmic sperm injection (ICSI), but few men with CF seek fertility treatment

To assess their knowledge and attitude towards fertility issues a questionnaire was sent to 71 men with CF: 50(70%) responded; 13(19%) declined to give information and 37(53%) replied in full. Overall 75% thought that nearly all men with CF were infertile, 25% did not know how many were affected, and 66% could describe the exact nature of infertility. Only 50% knew that treatment for infertility was available: 33% of these thought it was rarely successful and 33% were unsure of success rates. Although 75% had thought of having children at some stage, only 15% had sought investigation of infertility. Two men had children, one by ICSI and one by donor insemination. Overall 67% feared that the child could have CF and 53% worried that their health would affect their ability to function as a parent. In terms of receiving advice 57% thought that this should be given by a fertility specialist, 27% by a CF physician and 16% by a primary care physician. With regard to fertility treatment 44% thought that decisions should be made by the patients and their partners, 26% by the patient alone, and 30% by a doctor. The reluctance of men with CF to seek fertility treatment appears to be due to many factors including lack of knowledge of available treatments, fears about the child inheriting CF, and concerns that their health would adversely affect their role as a parent.

In addiition to information provided by the CF team advice from a specialist in fertility treatment may help to improve the provision of reproductive counselling to these patients.

Nash, E.F. 1,2 ; Coonar, A.S. 3 ; Stephenson, A.L. 1, 4 ; Delgado, D.H. 5 ; Singer, L.G. 2, 4 ; Tullis, E. 1, 4 ; Chaparro-Mutis, C. Studies have shown varying degrees of right ventricular (RV) dysfunction in adult CF patients with severe lung disease. Left ventricular (LV) dysfunction is much less common with conflicting reports as to its prevalence. Pulmonary hypertension (PH) is commonly seen, the severity correlating with FEV 1 . One group reported that PH was more prevalent in B. cepaciacolonized patients, albeit in a small sample. The aim of this study was to determine the true prevalence of PH and cardiac dysfunction in CF patients with severe lung disease.

We carried out a retrospective cohort study of all adult CF patients referred to the Lung Transplant Program at Toronto General Hospital from 1987-2007. Adult patients referred with non-CF bronchiectasis were included for comparison. Demographic data, lung function, 6-minute walk, micro-biology, Doppler echocardiography and resting and stress radionuclide ventriculography (MUGA) results were analyzed by Students t-test (parametric) or Mann Whitney U Test (non-parametric).

Results 244 adult CF patients (61% non-B. cepacia, 39% B. cepacia) and 25 non-CF bronchiectasis patients were included. Data from these 3 groups are summarized in the table. We found higher 6 minute walk and BMI values and a higher percentage of males in the B. cepacia compared to non-B. cepacia group. PH was present in 53% of the CF patients in whom PASP could be obtained. B. cepacia patients had lower first pass RVEF compared to non-B. cepacia patients. 51% of the B. cepacia patients had an abnormal LVEF response to exercise (defined as <5% increase on exercise) compared to 42% of the non-B. cepacia patients.

We found a high prevalence of PH as well as RV and LV dysfunction in CF patients with severe lung disease. B. cepacia-colonized patients had worse RV function compared to non-B. cepacia-colonized patients on first pass radionuclide ventriculography.

*Non-CF bronchiectasis group older than CF groups (p<0.001) B. cepacia CF group higher than non-B. cepacia CF group (p<0.01) B. cepacia CF group lower than non-B. cepacia CF group (p<.001)

Perobelli, S. 1 ; Dorazio, C. 1 ; Assael, B. 1 ; Tamanini, A. 2 ; Castellani, C. 1 1. Cystic Fibrosis Center, Verona, Italy; 2. Molecular Biology Laboratory, Verona, Italy CF Neonatal Screeening (NBS) can identify neonates with elevated IRT, normal or borderline sweat chloride concentrations, and one CF mutation plus an additional CFTR sequence variant, or, alternatively, two CFTR sequence variants. The combination of hypertrypsinogenemia and two CFTR gene defects is consistent with the presence of a CFTR-related disorder (CFTR-RD). Although these neonates are usually healthy at diagnosis, the long-term phenotypical consequences may be highly variable. Such variability makes impossible to predict the clinical outcome, and provide satisfactory genetic counselling for the family. The diagnosis of such a vague condition could potentially cause considerable family stress. This study aimed at assessing how parents of infants diagnosed with a CFTR-RD through NBS perceive their children's health.

A questionnaire for parents was designed, consisting of three sections. The first section asked about information given to the family in the neonatal period, and about emotional reactions to the CFTR-RD communication of diagnosis. The second section investigated the present perception of the child's health, and the parents level of anxiety/concern. The third section focused on the impact of the diagnosis on family planning, on relationships inside the couple and with the child, and on emotional distress; this section was concerned also with the use of carrier testing in relatives. Parents were also asked to fill in a standardized form on child behaviour (Achenbach Child Behaviour Check-List). The questionnaire was distributed to two populations, the parents of 18 children diagnosed with CFTR-RD through NBS (group A), and the parents of 18 children diagnosed with CF through NBS

As the life expectancy of children with CF increases new issues arise that have not been dealt with in the past. In the past,issues related to the support and care of a young adult with CF were rarely raised because only small numbers of children with CF lived into adulthood. However, with an increasing young adult population, young adults without family support have surfaced in significant numbers and their needs should be identified and understood by the CF Center Care Team. These young adults face the same range of issues others with CF face related to health insurance, government benefits, housing, education and employment. However, they do not have a family to help them wiht support, health insurance benefits and care needs.

METHOD In October 1998, the CF Legal Information Hotline ("Hotline") began to provide free and confidential information to people with CF, family members and healthcare providers in the U.S. The Hotline is staffed by attorneys, Beth Sufian and James Passamano(BS and JP) and is accessed by dialing a toll free number or by e-mail. Information about the Hotline is available at CF Care Centers, in CF publications and on the Web. Information regarding the content of each call is recorded on an intake form and includes caller name, address, phone number, age of person with CF, CF Center attended, and information related to the caller's question, attempts made to access legal information prior to calling the Hotline, relation of access to information to access to care. Calls and e-mails are tracked by seven major categories which are divided into subcategories. If a caller has more than one question, the call is tracked using the caller's primary issue.

RESULTS From November 2006 to April 2007 the CF Legal Information Hotline has received 42 calls from young adults with CF who do not have family support and have issues related to obtaining health insur-ance, government benefits, educational oppornties or employment opportunities. As well as issues related to lack of support for care. The Hotline has also received 33 calls from CF Care Centers team members with questions related to the needs of young adults without family support.

Over an eight year period the CF Hotline received _______calls. The categories for the calls were: Social Security benefits (1, 324) , Health Insurance (1, 163) , Educational (434), Employment (422), Long Term Disability benefits (190), Family Law (59) and miscellaneous legal issues (95).

DISCUSSION People with CF face obstacles to care if they are unaware of their legal rights especially in the areas of health insurance and government benefits. Healthcare providers are busy addressing the medical needs of patients and are often not well equipped to provide detailed information about the legal rights of their patients. The Hotline provides important information to both patients and healthcare providers that can optimize both patient care and outcomes. Access to legal information can result in individuals obtaining health insurance, Medicaid, Medicare or other state coverage that in turn results in better access to treatment. Legal information can also lead to important modifications in the educational or employment environment for the individual with CF. These modifications can be critical in allowing those individuals access to important care and treatments. Providing information that can help individuals obtain Social Security or Long Term Disability benefits can result in a monthly benefit and health insurance that then allows the individual to devote more time to their healthcare. Many people with CF and healthcare providers lack information about the legal rights of individuals with CF. Understanding the legal rights of people with CF can reduce obstacles to obtaining care, optimize education and employment, and increase access to retirement benefits. Further work needs to be done to identify ways to make individuals with CF and healthcare providers aware of the Hotline.

The CF Hotline provides a unique service to CF patients and healthcare providers by providing information regarding the legal rights of people with CF. Identification of ways to assist young adults without family support should be identified and CF center XCare teams shoudl be given the tools to allow them to help their patients access government assistance and other programs that can help make sure that the young adults have acces to care and support they need to live with CF. The DVD gives healthcare staff, patients and families a rare glimpse of ten couples with Cystic Fibrosis (CF) interacting at a CF Care Center-sponsored 'couples support group'. They discuss the impact this forum has in their daily lives for gainingo Information about how CF affects their lives as individuals, and as couples, and how the 'non-CF partner' learns from others. o Skills at networking and personal coping strategies. o Reflections on 'family life and CF', and o Observations on ways CF care centers can join them in greater health partnership.

Lastly, this DVD provides a 'starter kit' for ways to establish such a group.

One of the realities of a life-threatening illness can be that of social isolation, and of separation from others who share the same disease. Having the opportunity of participate in a support group can significantly reduce that sense of isolation and give the participants an ability to share experiences and give support to others with this disease.

Cystic fibrosis as an illness has growing ranks of 'survivors' who live longer, fuller lives, thanks to the advances made on multiple-research fronts. Historically, adults with cystic fibrosis have not had much occasion to share person-to-person experiences within the structure of their CF care centers. Justifiably, with the potential for cross-infection, the Cystic Fibrosis Foundation has urged caution for the CF population in having face-to-face interactions. In 2004, our CF Care Center recognized this need and sponsoredwith guidance from staff facilitators-an on-going adult support group. Prior to the formation of the group pre-and on-site group IC protocols were developed. At the pre-group stage, the physician initially certifies that the patient is free from aggressive pathogens. Each time the patient RSVP's permission to attend a group session, the physician must 'recertify' this status. On-site protocols emphasize a number of IC precautions to minimize crossinfection. The Center recognized that by developing such a support group, 'patient and family centered care' could be enhanced. A goal in the group's formation was to create 'mutually beneficial partnerships among patients, families, and healthcare providers'.

Group Structure/Successes to Date: The group is limited to 20 participants. Two CF care center staff act as facilitators. This is a support, not a therapy group. Compliance with therapies is reported to have increased as one outcome of this forum. Formal topics are purposefully not predetermined; the group's success is predicated on the personal responsibility of each participant to bring topics/issues from their daily life experiences. The group has gained identity and trust and has evolved over three years into two separate groups: an individual CF group and a couples CF group. Ownership of these groups by those who attend is apparent. Both the care center and the couples support group are forums for learning 'how to live with CF'.

Future goals will call for using identified quality of life tools to measure psychosocial and emotional growth in individuals through their participation over time in the group.

Sawicki, G. 1 ; Asher, D. 1 ; Dill, E. 2 ; Sellers, D. 2 ; Robinson, W. 2 1. Children's Hospital, Boston, MA, USA; 2. Education Development Center, Newton, MA, USA Background: Advance care planning (ACP) is an important tool to promote alignment of the patient's care with his or her needs, goals and values, and is particularly useful if the patient's ability to make decision becomes compromised. The purpose of ACP is to develop a treatment plan promoting a high quality of life for those near the end of life. Such planning is especially appropriate in persons with life limiting disease. Minimal research exists on the advance care planning process of adults with CF.<p>Methods: In the fifth survey wave of the Project on Adult Care in Cystic Fibrosis (PAC-CF), an ongoing longitudinal panel study of CF adults from 10 CF centers in the US, participants were asked to report on their opinions and experiences with advance care planning. The survey asked about opinions on advance care planning documents, communication about advance care planning with family and clinicians and whether a directive had been completed. Clinical data were obtained from the participating centers. Preliminary bivariate analyses were conducted to examine factors associated with completing an advance care directive.<p>Results: 233/303 (77%) surveys were completed. The mean age was 34 ± 13 years, 63% were female, and the mean FEV1 (% predicted) was 69 ± 28. 77% reported that they had spoken to someone about the care they would want if they became too ill to make their own decisions in the future, 65% had thought about whom they would want as their healthcare proxy, and 60% reported having specific wishes about the types of medical treatment they want or would not want if they became too ill to make decisions for themselves in the future. However, only 33% reported completing a healthcare proxy form, living will or any other type of written instructions concerning advance care directives. 80% of respondents reported feeling comfortable talking to their clinician about ACP. 28% of subjects said that their CF clinicians have asked them about ACP, and only 11% reported that they have discussed ACP with their CF clinician. The following characteristics were significantly related (p<.05) to completion of an advance care directive: female gender (38% vs. 25%), age (mean 37 years vs. 33 years), having a college degree (41% vs. 23%), being married currently or in the past (38% vs. 25%), having children (41% vs. 28%), having diabetes (48% vs. 31%), having specific wishes about treatment decisions (46% vs. 15%), and reporting that a clinician had discussed ACP with them (59% vs. 24%).<p>Discussion: Though the majority of CF adults report thinking about, communicating with family and deciding on their advance care wishes, only a minority report completing any legal documentation supporting their decision. Additionally, very few CF adults report being asked about ACP by their clinicians and even fewer report discussing ACP with them. Formulating specific wishes and discussing ACP with a clinician are strongly associated with completing an advance care directive, suggesting that if clinicians were more active in talking to their patients about ACP, patients are much more likely to complete advance care directives. Efforts to improve clinician communication with CF adults around the issues of advance care planning are needed. Objectives: Caregiver involvement in children's CF treatments fosters better adherence and improved health outcomes. Several studies have suggested that parents of children with CF report elevated symptoms of depression (Quittner et al., 1998) , however, there is no data on the effects of caregiver depression on adherence. Adherence to treatment is difficult to measure in patients with CF, and prior research has relied primarily on self-report, which is likely to be inflated. The purpose of this study was to examine the effects of caregiver depression on adherence to enzymes using both parent-reported and electronically monitored enzyme adherence over a three-month period. In addition, we examined the relationship between depression, rates of enzyme adherence and short-term changes in weight.

Methods: As part of a larger intervention study at 3 CF Centers, 89 children with CF ages 1 to 11 and their parents were recruited. Parents completed a standardized measure of depression (CESD) at enrollment and were provided with MEMS caps that recorded the date and time of each bottle opening. Three months later, the electronic data from the pill caps was downloaded and parents completed a structured interview reporting their child's adherence to all components of the CF treatment regimen. Only the section addressing enzyme adherence was included in this analysis. Standard health outcomes, such as weight and height, were also assessed at enrollment and 3 months later.

Results: Preliminary analyses indicated that caregivers reported elevated levels of depression, with 30% scoring in the clinical range. Rates of adherence to enzymes were poor (43% at home and 48% at school). Caregiver depression was negatively associated with adherence, with depressed caregivers demonstrating lower rates of adherence (11 percentage points). Adherence to enzymes was associated with changes in weight, with a 100% adherence translating into 5 percentile points of weight gain. Final analyses will also include parent-reported adherence to enzymes.

Conclusions: A substantial number of parents scored in the clinical range on a depression screening measure. In addition, depression was associated with poorer adherence. Rates of adherence to enzymes were surprisingly low, both at home and at school. Furthermore, poor adherence was associated with a decrease in weight three months later. Caregiver depression appears to be under-diagnosed and these results suggest that screening and intervention may be warranted. Adherence to enzymes should also be targeted in clinical interventions.

Funding was provided by NIH grant #HL69736 of non-compliance with medical treatment and greater behavioral and emotional distress. Adherence is particularly problematic during adolescence, however, few family-based interventions have been developed to target adherence behaviors in this population. The goal of this study was to compare the effects of Behavioral-Family Systems Therapy (BFST), an empirically-supported treatment, to both Family Education (FE) and Standard Care (SC). Methods: As part of a larger intervention study, 117 adolescents with CF ages 10 to 17 and their parents completed the Conflict Behavior Questionnaire (CBQ) and the Parent-Adolescent Relationship Questionnaire (PARQ) developed by Robin and Foster (1989) . Adolescents completed these measures for each parent and both parents, if available, completed them with respect to their teen. The Problem-Solving, Communication, and Beliefs subscales from the PARQ were administered. Participants were then randomly assigned to one of three treatment arms: BFST, FE or SC. Families in the BFST group received ten 90-minute sessions over a 6 month period, including family problem-solving, communication skills training, and cognitive restructuring. Families assigned to the FE group received ten 90minute psychoeducational sessions over 6 months aimed at increasing knowledge about CF and its management. Those in the SC group received their usual care at the CF Center. The CBQ and PARQ were completed pre, post and 6, 12 and 18 months following the intervention.

Results: Adolescents reported that the BFST intervention, relative to the other two, improved their communication with their primary caregivers (p < .03) and caregivers improved their communication with their adolescents (p < .02). In addition, caregivers reported improved problem-solving with their adolescents (p < .04). On the CBQ, BFST reduced both adolescent and primary caregiver conflict and FE reduced caregiver conflict. No significant improvements were found for those in SC.

Conclusions: Preliminary findings suggest that BFST is effective in improving communication skills and reducing conflict in adolescents with CF and their caregivers. Future analyses will evaluate the effectiveness of BFST and FE on adherence and other family functioning measures over the course of the study.

Funding was provided by NIH (HL #47064) Background: Cases of diabetes among people with cystic fibrosis (CF) have increased as life expectancy for these patients improves, yet the impact of this additional illness on daily functioning is under-researched. To explore this issue, a study was conducted comparing the views of patients with either cystic fibrosis related diabetes (CFRD) or type 1 diabetes mellitus (T1DM) about being diagnosed and living with diabetes.

Methods: Qualitative research was used because the study was concerned with understanding diabetes from patients' perspectives. Purposive sampling was employed to achieve maximum variation in terms of duration and type of diabetes experienced by participants. Data were collected via semi-structured interviews, all of which were taped and transcribed verbatim. Recruitment continued until no new ideas or insights emerged from additional participants. A framework approach to analysis was adopted. This involved coding and summarising interview data into charts to explore and develop main themes. Initial analysis was undertaken by two researchers (ST and CD) and amended after comments from the remaining authors.

Results: Interviews, carried out with 11 CFRD and 12 similarly aged T1DM patients, lasted for an average of 40 minutes (range 20-60 minutes). The following themes were derived from the data collected: Evolving vs fracturing identity; Diabetes in context; Self-management motivators. For patients with CFRD, diagnosis represented a progression in their health status, which called on them to adapt existing treatment regimens to accommodate this additional condition. In contrast, interviewees with T1DM had to re-evaluate their previous sense of self as 'healthy' and adjust to manag-ing a long-term complaint. These individuals were more likely to talk about diabetes in relation to a range of competing commitments (e.g. work or family related) and to describe feeling psychologically low due to diabetes compared to patients with CFRD, the latter depicting demands from their primary illness (CF) as a major obstacle to caring for diabetes. Participants with CFRD recalled feeling lucky when told they would not face strict dietary restraints, which they associated with other forms of diabetes, and seemed less concerned about diabetic complications than those with T1DM. For interviewees with T1DM, a desire to reduce future health risks motivated their self-management efforts, whilst those with CFRD were driven by the negative effect poor control of diabetes had on their chest and weight.

Conclusions: Both sets of interviewees found diabetes time consuming and, on occasions, frustrating to accommodate into daily life. Findings from the study act as a reminder that patients manage their condition in the real world, against a plethora of other demands on their time and energy. In the narratives of participants with CFRD, these demands were often related to the existence of their primary illness.

ing the course of development. The purpose of this study is to examine rates of medication adherence across a wide age span using an objective measure of adherence and to compare adherence to health outcomes. METHODS: Patients with CF age 6 years or older who are prescribed azithromycin, colistin, hypertonic saline, pulmozyme, and/or TOBI for at least one year are eligible. Recruitment is ongoing. The previous year's medication refill records were requested from participant-identified pharmacies. A Medication Possession Ratio (MPR) was calculated for each drug and was defined as the sum of all days of medication supply received during the 12 months divided by the number of days the medication was prescribed. Values were truncated to 100% and averaged across all medications to obtain a Composite Score. Medical records were abstracted to identify the prescribed drug regimen and health outcomes (e.g., lung function, exacerbations, BMI) over the same time period. RESULTS: Thus far, 100 patients have joined the study (80% participation rate), resulting in 213 pharmacy records requested (86 unique pharmacies). Of the first 48 participants with complete pharmacy data, the mean age was 21.5 years (SD=11.9; Range= 7-69) and 56% were female. The table presents MPR data for each drug and Composite Score overall and stratified by child, adolescent, and adult participants. The complexity of the drug regimen and age were significantly correlated with the Composite Score (rho= -.39 and rho= -.29 respectively; p<.05). Future analyses will compare the Composite Score and each drug's MPR with health outcomes. CON-CLUSIONS: Participants had suboptimal medication adherence similar to that reported in the literature. As MPR provides the maximal possible level of adherence, actual adherence may be even lower. Poor adherence spanned the age groups and was associated with regimen complexity. Children had the highest adherence, but also the least complicated drug regimen. These results suggest that obtaining pharmacy records is a viable means to objectively assessing medication adherence. Moreover, results suggest that interventions targeting medication adherence may be a strategy for improving health outcomes, particularly for adolescents and adults.

* fewer than 5 participants within this cell 564 (CF) whose exchange capacities for oxygen and carbon dioxide are already diminished. Lack of REM and Delta sleep have significant implications for the daytime functioning of pediatric patients with CF, potentially leading to impaired memory and attention which would reduce their ability to complete typical activities, such as school work and disease-specific tasks related to self care and adherence. Based on these clinic findings, the authors are beginning a prospective study on the effects of sleep on neuropsychological functioning, well-being and compliance in pediatric patients with CF. Future research should examine the relationship between decreased slow wave, REM sleep and psychosocial outcomes, including attention, memory, and health-related quality of life. There are diagnostic challenges related to non-classic CF (Knowles and Durie, 2002) particularly patients presenting in adolescence or adulthood with obstructive azoospermia, chronic sinopulmonary disease and chronic or acute, recurrent pancreatitis. Practitioners are often unable to provide clarity to patients about diagnosis, disease course, and prognosis. Potential harms and benefits of relaying a clear or unclear diagnosis may be psychological rather than physical and have long-term implications for mental and social well-being. The aim of the current study is to assess the psychological impact of diagnostic information for adults presenting with a CF phenotype, be it confirmatory (CF or most likely not CF) or unclear. Method: We administered self report measures pertaining to psychological state, cognitive appraisal and uncertainty in a welldefined cohort of adult patients. Presenting symptoms of patients include chronic sinopulmonary disease, obstructive azoospermia, pancreatitis, and those presenting with more than more CF phenotype. Patients completed the self-report measures on two occasions: at the time of diagnostic testing and 6 months after being counseled and notified of the diagnostic results. At the time of the initial assessment, the subjects were unaware of the test results and had not been seen by a CF physician. Results: We provide interim observations on 70 patients (sinopulmonary (n=34), obstructive azoospermia (n=20), pancreatitis (n=4), and multiple phenotypes (n=6) who completed the initial assessment. At the time of diagnostic testing, the level of depression, hostility, anxiety and interpersonal sensitivity were elevated in each group (T score > 65); male patients reported significantly greater depression, anxiety and interpersonal sensitivity than female patients. Compared to published normative data of patients with mixed chronic illnesses, patients presenting with sinopulmonary or pancreatic disorders reported significantly more uncertainty(p < .05). To date, 26 of these patients have completed the measures 6 months after receiving the diagnostic information. Preliminary review of results shows that the level of depression, hostility, anxiety and interpersonal sensitivity remained elevated while the degree of uncertainty decreased. Among the 26 patients who completed assessment at 6 months, 17 patients were told that it was unlikely that they had CF, 7 were told that they had mild CF, while 2 were given an unclear diagnosis. Those who were told that they did not have CF reported more uncertainty than those who were told they had CF (p < .05). Detailed analyses related to outcome diagnosis will be conducted at the time of study completion (October 2007) . When this study is complete, the information will assist in our understanding of the psychological impact of a genetic diagnosis at an older age, identify issues confronting those individuals and help to establish appropriate paradigms for delivering complex information about genetic diseases.

Supported by grants from NIDDK [SCOR] and the Canadian CF Foundation. (1). In accordance with these guidelines, we report our experience in using our Nutrition Action Plan (NAP) in patients with suboptimal nutrition.

Objective: To determine if implementing our NAP would improve the BMI of these patients.

Methods: Our NAP is a written contract between the patient, caregiver and CF team. For clarity of nutritional goals, the NAP is based on the colors of a traffic light. The red zone indicates a BMI less than the 25th percentile and poor nutritional status. The yellow zone indicates a BMI between the 25th and 49th percentile and fair nutritional status. The green zone indicates a BMI greater or equal to the 50th percentile and desired nutritional status. Within each color zone, specific recommendations to improve nutritional status are listed. The ultimate goal for each patient is reaching the green zone. The caregiver, patient and nutritionist review the current nutritional status and a mutually accepted weight gain goal is established. The goal is recorded on the NAP which is then signed by the nutritionist and patient. This contract establishes a commitment to achieve the goal by the next clinic visit. A reward system is implemented in conjunction with the NAP. For patients reaching their goal, a previously agreed upon prize is given. Patients achieving some weight gain but falling short of their goal receive a gift such as candy or small toys that serve as an incentive for continued weight gain. Patients chosen for inclusion into this study demonstrated a BMI less than the 50th percentile; those on growth hormone, appetite stimulants or chronic systemic steroids were excluded.

Results: Twenty-one patients, 12 male and 9 female, with a mean age of 11.6 years (range 3 to 18 years) were included in this study. Fifteen patients demonstrated an improvement in their BMI with use of the NAP: 7 improved by 1-10%; 2 improved by 11-20%; 3 improved by 21-30%; 1 improved by 31-40%; 2 improved by 41-50%. Six participants had a negative change in BMI which we attribute to family upheaval, missed appointments with the CF Team, exacerbations of disease and non-adherence.

Conclusion: We intervened in 21 patients with a BMI less than the 50th percentile using a NAP and a reward system. The results of this pilot study suggest that our NAP and reward system may be of benefit in achieving positive gains in BMI.

Reference: (1) Previously it was rare for women with severe CF to undertake pregnancy and doctors tended to advise against pregnancy if the forced expiratory volume in one second (FEV1) was less than 60%. The attitudes of doctors and patients are changing and many women living their lives with CF nowadays chose to undertake pregnancy despite the severity of their health problems.

Over the last 15 years 102 females (age range 16-56 years) have attended the Newcastle Adult CF Centre. There have been a total of 25 pregnancies involving 20 women with 24 live births (1 ectopic). A further 3 women had termination of unwanted pregnancy. Mean age at the time of pregnancy was 21 years. We compared an age-matched cohort of never pregnant women to see if there were differences in key variables making successful pregnancy more likely.

The mean FEV1 was 61% (range 30-112%) and 58% (range 10-91%) of predicted in the pregnant and non-pregnant group respectively (p=0.36); Rates of chronic Pseudomonas aeuriginosa infection were 14 (70%) and 12 (57%) respectively (p=0.52). 2 cases of Burkholderia cenocepacia were identified in the non pregnant group. There were no cases in the pregnant group (p=0.21); 7 (35%) had diabetes with 5 patients having CF related diabetes prior to pregnancy and 2 patients developing gestational diabetes; this compared with 9 (21%) of non pregnant women (p=0.75); There was impaired nutritional status within both groups with a mean BMI of 18.9 pre pregnancy (range 14-27) and 20.1 in the non pregnant group (range 14-31) (p=0.28); 5 (25%) and 3 (14%) patients had had previous gastrostomy feeding (p=0.45). Pancreatic insufficiency was present in 17 (85%) and 19 (90%) patients in respective groups (p=0.66).

No child had CF but one child had complex congenital heart disease. There were no maternal deaths but 2 patients subsequently underwent lung transplantation and 2 patients died leaving young children (ages 5 and 7 years). 2 patients in the non-pregnant group have also died.

Conclusion

The profile of women with CF undertaking pregnancy is now similar to the overall CF population. The woman's decision to undertake pregnancy is often not directly related to the severity of her disease. Amongst those who successfully completed pregnancy maternal and fetal outcomes are generally favorable. Objectives: Health-related quality of life (HRQoL) instruments provide important information on disease progression and are increasingly used as patient-reported outcomes (PROs) for behavioral and pharmacological trials (Goss & Quittner, in press) . Recent advances in medical care have improved the long-term stability of pulmonary functioning, which has made it more difficult to detect small improvements in FEV1% predicted. New outcome measures are needed; the Cystic Fibrosis Questionnaire-Revised (CFQ-R; Quittner et al., 2004) , a disease-specific HRQoL measure, has shown promise in this regard. The purpose of this study was to examine the longitudinal relationships between FEV1% predicted, weight percentiles, and CFQ-R scores for adolescents with CF.

Methods: As part of a larger study evaluating adherence interventions, 117 adolescents with CF, ages 10 to 17, and their parents were enrolled at six centers. Parents reported on their adolescents using the CFQ-Parent version, and adolescents completed the CFQ-Teen/Adult version as a selfreport. Measures were completed pre and post treatment and at 6, 12, and 24 month follow-up visits. Spirometry and weight data were collected at these same time points.

Results: Empirical bayes estimates were used to examine change over time. Preliminary results indicated that adolescents' scores on the CFQ-R Respiratory scale were significantly and positively correlated with changes in pulmonary functioning. Parents' scores on the CFQ-R Vitality scale were also correlated with changes in pulmonary functioning, with higher CFQ-R Vitality scores associated with better lung function. In addition, parents' scores on the CFQ-R Weight scale were significantly and positively correlated with changes in adolescents' weight.

Conclusions: These results support the validity and sensitivity of the CFQ-R to changes in both pulmonary functioning and weight for adolescents with CF. Additional analyses will be conducted to determine the relative stability of HRQoL scores over time and to examine other CFQ-R scales that may reflect positive changes in adolescent functioning (e.g., reduced Treatment Burden Background: Development of CF specific patient symptom tools as outcome measures is a critical step toward the development of potential new therapies for CF. We have developed a novel instrument to assess respiratory symptoms in CF patients.

Methods: We conducted 25 in-depth qualitative interviews using the Day Reconstruction Method and 9 cognitive debriefing interviews (3 adults, 4 youth, and 2 parents) at two CF programs, the University of Washington and Children's Hospital and Regional Medical Center. Interviews were conducted until no new symptoms were mentioned by interviewees (i.e., data saturation). The interviews were audio-recorded, transcribed, coded and analyzed for themes.

Results: Six pulmonary symptoms were identified in the interviews: cough, sputum production, wheeze, chest tightness, Difficulty breathing/shortness of breath and fever. In addition, the most commonly cited activity and emotional impacts identified in the interviews were also included on the questionnaire. These emotions included: frustration, sadness/depression, irritability, worry, difficulty sleeping, and activities included time spent sitting or lying down, reduction of usual activities, and missing school or work. In all, eight symptom items were selected for inclusion on the self-administered questionnaire (see table) . As a result of the cognitive debriefing interviews, we changed initial language and adjusted the response options on some of the items in order to create greater distinctions between them. No important issues were felt to be omitted by patients who were interviewed.

Conclusions: Using qualitative inductive methodology, we have created a novel patient reported outcome measure for CF. Further study will be required to assess validation of the instrument. or more signs and symptoms occurred for more than two days, antibiotics were ordered with physician input and phone follow-up in 3 days. If symptoms occurred less than two days, chest clearance was increased with phone follow-up in 24 hours. The initial algorithm (IA) was piloted by 2 nurses on 25 patient calls on the EMR and reviewed for adherence. Subsequently, the CF phone note was revised, expanding the symptoms to include GI issues, last visit date and last hospitalization. The revised algorithm (RA) was used for a one month trial period with the entire nursing staff. A total of 22 patient calls relating to pulmonary exacerbation were again reviewed for adherence to the algorithm.

RESULTS: The review of 25 phone notes following the IA found 14 (56%) of the phone notes to be adherent to the assessment and treatment plan on the algorithm. Several exceptions included calls received for GI issues, medications, complaints of general pediatric symptoms, and lab results. The phone follow-up within 24-72 hrs was also difficult to accomplish secondary to nursing responsibilities and difficulty locating the patient. The review of calls following the RA found 19 (86%) notes were adherent to the algorithm.

DISCUSSION: The revised algorithm and CF phone note for pulmonary exacerbation improved consistency in nursing assessment of signs and symptoms, evaluation of the current plan of care, and triggered appropriate interventions based on symptoms and length of illness. This improvement was particularly pronounced among the part-time RNs who found it easier to obtain a more accurate history of patient's illnesses without asking for assistance from other CF team members. Since the algorithm and CF phone note are newly implemented, tracking for consistency with the CF phone note will continue, revisions will be made as needed, and phone notes for conditions other than pulmonary exacerbation may be developed. METHODS: In 2000, Helen DeVos Children's Hospital integrated Research Coordinators (RC) from the Spectrum Health Research Department into the CF Care team. This model incorporates two key aspects: 1. a shared resource of experienced, centrally educated and trained research coordinators (RC) with three RC's having a primary CF focus 2. Incorporation of RC's into Helen DeVos CF Care Centers' multidisciplinary team for weekly contact with patients. Using the CF Registry consent as a starting point of research discussion with the patients and families, the RC's present new and upcoming opportunities and answer questions about research. With attendance at local and national CF meetings, RC's have expanded their CF knowledge, networked with other RC's and explored new research trials. Our site also integrated education about various aspects of research into both the clinical staff and CF Family Education meetings. A review of the research database was completed to assess our CF Center's research program.

RESULTS: Over the 7 years reviewed, participation in clinical research trials has seen significant growth and patient participation has exceeded 250 patients enrolled into clinical studies.

CONCLUSION: Utilizing a team of dedicated RC's from the Research Department has proven to be effective in advancing our CF Center's clinical research program.

DISCUSSION: To meet the growing demand of patients needed to participate in clinical trials, CF centers will need to provide research education to both the patients and families and the clinical team, improve on skills needed to identify and enroll our patients into clinical trials and include the RC's into the multi-disciplinary CF patient care team. Centralized research represents an excellent model for CF Centers that want to excel in clinical research. Integration of RC's into the CF care team, along with education initiatives for both clinical staff and patients and families has led to successful participation and enrollment in clinical trials. Further studies need to be done to better identify barriers that patients and families have with participation in research. The goal of "Cystic Fibrosis transition care" is to ensure that youth with Cystic Fibrosis (CF) are adequately prepared to participate in the management of their health condition into adulthood and as they graduate to the adult health care system. In Canada, it is unknown if, or to what extent, CF transition care is practiced in individual clinics. We therefore surveyed all Canadian pediatric CF clinic coordinators (n=27)using a brief standardized questionnaire; 23 surveys (85%) were completed and returned for analysis.

FINDINGS: All 23 responding clinics transfer their pediatric patients to a distinct adult CF clinic, however almost half of these share one or more team members between the two clinics. Transition care is recognized and practiced by 74% of responding clinics: 22% follow a formal transition program, and 52% follow an "informal program". "Informal" practices vary widely from one clinic to another. Formal programs were created by 5 individual clinics, and share the common properties of being based on current research, having set goals, and utilizing a tool to document the process. The average age of transfer to the adult clinic is 18 years. Most clinics have "rare exceptions only" to delaying the age of transfer, the most commonly cited reason being "very ill/palliative patient". A small number of clinics also cited "intellectually challenged patient" and "reluctant patient" as reasons to delay transfer. Only 30% of pediatric clinics hold a formal "transition" or "graduation" clinic, which allows adult team members to be introduced to youth before the actual transfer of care takes place.

CONCLUSIONS: Although sharing team members between pediatric and adult clinics may ease the change from pediatric to adult care, it can also create barriers to the transition process. Therefore, it is encouraging to learn that the majority of Canadian pediatric CF clinics view adult care as distinct from pediatric care, and that there is a consistent age of transfer across the country. Transition preparedness is also recognized as an important component of care, as evidenced by the number of clinics following a formal or informal transition process. It remains undetermined, however, whether in general there is adequate preparation for adult care. With only 22% of clinics following a formal transition process (and 26% not following any type of transition program), further assessment seems warranted and may reveal inadequacies requiring remediation.

FUTURE DIRECTION: To strengthen CF transition care in Canada, a "patient readiness to transition" questionnaire has been developed and will be administered to Canadian CF patients transferring to adult CF clinics in 2007. Analysis of this questionnaire will provide insight into how well Canadian CF clinics are preparing their youth for transfer to adult life, and may provide directions for further improvement.

Nearly all females with CF now survive into adulthood, and advice regarding pregnancy and contraception is becoming increasingly important as part of their sexual health education. We set out to identify the level of knowledge of these issues in patients attending our large adult CF unit.

Using a structured questionnaire, we surveyed 49 consecutive CF females (age range 17 to 42 years) attending routine CF clinics. We asked about knowledge and usage of contraception, and issues relating to pregnancy and its possible effect in CF, ensuring the opportunity for a one to one discussion with the individual on completion of the questionnaire.

Only 29 (60%) claimed to have had previous advice relating to contraception and 23 (47%) pregnancy education. Of these, 23 (81%) stated that this was from adult CF nurse specialists, 17 (57%) from adult CF physicians and the remainder from other sources (general medical practitioners, written literature, and the internet). Only 7 patients (16%) recalled that they had been given information in the paediatric sector before transition. The majority of all advice was given verbally. Of the 29 (60%) who were using some form of contraception (10 [33%] condoms, 9 [30%] combined pill, 3 [11%] depot injection, 2 [7%] a coil, and 6 [19%] the mini pill); 10 (33%) had experienced problems, and in some cases patients had been misinformed about the reliability of their contraceptive choice.

As regards pregnancy, 2 (4%) had undergone previous termination and 8 (16%) already had children. Thirty six (73%) had considered becoming pregnant but 46 (93%) said they would discuss this with a member of the CF team beforehand: 35 (71%) were aware this might impact on their health. Of these, 32 (91%) believed a reduction in lung capacity to be the biggest possible problem. Other problems identified were alteration of medication, tiredness, and weight loss or gain.

Overall, 29 (60%) felt they did not have enough information regarding pregnancy in CF and 25 (51%) that insufficient contraceptive advice was given, expressing a wish for further knowledge.

Conclusion This survey has shown that a significant number of adult females with CF require further education about contraception and pregnancy. In some cases the advice they had already been given may not have been appropriate for patients with CF. Furthermore, few patients appeared to have had effective counselling in the paediatric sector, despite the risk of pregnancy. We are working with patients and paediatric colleagues to improve the education in this important area for our women with CF. The state of Minnesota added cystic fibrosis to the newborn screening panel on March 1, 2006. One-year follow-up to date reveals 286 infants have screened positive, of the screen positive, 23 have been diagnosed with cystic fibrosis, 30 infants are pending sweat chloride tests and evaluation. Of this population 8 infants are being followed by the Minnesota CF Center. The University has a comprehensive, interdisciplinary team available to provide care to infants, children and adults with cystic fibrosis. In response to newborn screening for cystic fibrosis, the center has implemented a prophylactic care program to include clinical care through early intervention and education within the first two weeks of life. Benefits of collaborative care have been previously identified in the literature, showing improved quality of care with increased patient satisfaction, lower mortality, and improved outcomes. Infants who are identified by newborn screening and receive earlier treatment have the potential for improved physical health and development. The inclusion of comprehensive genetic consultation as an integral component of CF education, allows for informed reproductive decisions. The newborn screening intervention plan is supervised by a nurse practitioner and consists of initial evaluation, initial therapy implementation, and a teaching/consultation schedule, which covers a minimum of four appointments scheduled over 2 weeks. Follow-up visits are scheduled within 24 hours of referral and then at one week, 10 days and 2 weeks post referral. The schedule is adjusted based on infant acuity and family needs. Initial evaluation consists of the confirmatory sweat chloride testing and/or genetic mutation analysis, serum laboratories, chest x-ray, stool for fecal elastase, and nasopharyngeal culture. General, supportive teaching with evidence from the literature is provided. Preliminary results to date demonstrate that our infants have had no pulmonary exacerbations, no pulmonary hospitalizations with pulmonary functions obtained in over half of the infants. Nasopharyngeal cultures obtained at the time of ascertainment demonstrate newborn screening patients have positive cultures for Staphylococcus aureus, Haemophilus influenzae, Escherichia coli, Streptococcus pneumoniae, Klebsiella pneumoniae, Acinetobacter junii and Alcaligenes faecalis.

One infant was culture positive for Pseudomonas aeruginosa at the first out-patient clinic visit. This demonstrates colonization of organisms prior to clinic exposure. Through Minnesota's early intervention and education plan, the families will be more adherent to prescribed care, which will result in improved CF infant outcomes as monitored by number of exacerbations, pulmonary functions and potential complications. A detailed outline of medical intervention, education, and outcomes will be presented.

CF-Pediatric Centre

We developed a Malabsorption Blood Test (MBT) using 2 odd-chain length fatty acids, pentadecanoic acid (PA) and heptadecanoic acid (HA), to assess fat absorption in subjects with CF. PA is a free fatty acid, while HA requires hydrolysis of triheptadecanoic acid (THA) by pancreatic enzymes for absorption. Objective: To determine the MBT reproducibility in healthy subjects and subjects with CF. Methods: Subjects with CF ingested a liquid test meal including 2 fats (PA and THA) on 3 occasions at least 5 days apart. For each subject, a standard dose of PERT (80,000 lipase units) or the subject's usual dose (if higher) was given with each test meal. Serum was analyzed for PA and HA levels at baseline and then hourly for 8 hours. A non-compartmental pharmacokinetic analysis was performed using PA and HA concentration-time data from subjects with CF. C max was directly observed from the individual subject profiles, and AUC was calculated using the linear trapezoid rule. Summary statistics (mean±s.d.) were calculated for all parameters. Reproducibility of the MBT was also assessed in healthy adult controls tested on 3 occasions. Between subject (BSV, %) and within subject variability (WSV, %) was calculated for PA, HA and the HA/PA ratio. Results: BSV (%) and WSV (%) of PA and HA absorption are presented in the Table for 9 subjects with CF (age 17.8±3.9 y, 4 females) and 6 healthy adult subjects (age 25.8±3.8 y, 4 females). As expected, the MBT had greater reproducibility in healthy controls than in subjects with CF. In healthy controls, the reproducibility for PA and HA absorption was comparable; there is less WSV than BSV, and the HA/PA ratio has the best reproducibility. In subjects with CF, PA and the HA/PA ratio show less WSV than HA. Conclusion: Wider variability in fat absorption in subjects with CF reflects the complex interactions of the biliary, pancreatic and intestinal factors

Background: The international, multi-center Gene Modifier Study of cystic fibrosis liver disease (CFLD) initially enrolled 133 patients with severe liver disease (i.e. cirrhosis and portal hypertension). For the initial population

Despite severe liver disease with portal hypertension, many of these patients had normal liver biochemical function tests and/or INR (24-68%), depending on the test. The mean age of enrollment into the study was 20 years, with the mean age of CFLD diagnosis of 11 years. Candidate gene testing revealed an increased prevalence of the alpha-1 antitrypsin Z allele in CFLD patients, particularly in females, as compared to CF patients (>15 yrs.) without CFLD, and an association with TGFβ1 variants (-509 and codon 10) in males

Smoothing reference centile curves: the LMS method and penalized likelihood

HEPB) AND INACTIVATED HEPATITIS A (HEPA) VACCINES IN CYSTIC FIBROSIS (CF) PATIENTS

Chronic liver disease (CLD) remains the second mortality cause in CF. We evaluated the immunogenicity of HepA-and HepB-vaccines in CF-patients because it is described lower in patients with CLD. Patients and methods: Blood samples of 153 CF-patients aged 0.2-57.8y (mean 13.5y) were tested twice (time interval 1y) by chemiluminescent microparticle immunoassay to asses the presence of HBs-Ab and HA-IgG, and their vaccination status was recorded. Seronegative patients for HBs-Ab and/or HA-IgG were vaccinated in between tests with Engerix®, Havrix® or Twinrix®. Results: 115/120 (95.8%) patients tested positive for HBs-Ab. 59/115 CLD (9 steatosis and 10 cirrhosis). 3/115 tested positive at first but negative

At present data are incomplete for HBs-Ab and HA-IgG for 33 and 70 patients. Conclusions: 1. Immunogenicity of HepA-and HepB-vaccines is comparable in CF-patients and healthy subjects. 2. CF-patients are at risk for CLD and seroconversion must be checked after vaccination. 3. Vaccination records are important as numbers of antibodies may decline beneath the detection limit in time

Methods: LF was assessed in 53 children with CF, aged 8-12 yrs, using forced expiratory volume (FEV(SUB)1(/SUB)% predicted). Body composition was measured using a reference four-component model(sup)1(/sup) (4CM) allowing accurate evaluation of both fat mass (FM) and the components of fat-free mass

Dual-energy Xray absorptiometry (DXA; Lunar Prodigy, G.E. Medical Systems) was used to measure fat, lean (non mineral) and bone mass and skin-fold thicknesses (bicep, tricep, sub-scapular) were measured. Strength of relationship was assessed using Pearson's correlation coefficient (r) and significant body composition components were fitted into a regression model. Difference in FEV(SUB)1(/SUB) between the sexes was assessed with an independent t test. Results: Boys with CF did not differ from the reference population

Conclusion: Our results confirm that body fat, but not FFM or bone mass, are related to the severity of impaired LF in children with CF. It is likely that the significant association between FEV(SUB)1(/SUB)and fat in girls but not boys reflects the poor body composition of the girls with CF. Longitudinal follow-up of these children should indicate whether this sex difference persists after puberty. Given the fact that prognosis is worse in girls the sex difference we've identified merits more attention. 1Fuller NJ et al. Four-component model for the assessment of body composition in humans: comparison with alternative methods and evaluation of the density and hydration of fat-free mass

Conclusions: These results show that pancreatic status plays an important role in relation with PUFA status in CF patients and particularly in

(the initial fatty acids in the n-6 and n-3 pathways) in neutral lipids, nonesterified fatty acids, and phospholipids

This may, in some way, account for the fact that 18:2n-6 is more readily metabolized, and therefore depleted, in CF animals than in WT mice. Conclusion: DHA supplementation increased the localization

539 DOES AN INTEGRATED CLINICAL AND NUTRITIONAL APPROACH PREVENT

This case control study examines the impact of clinical approach on pre and post-CFRD clinical course. 48 patients with CFRD (mean age 25.9+5.9yrs) matched to 48 CF controls (25.9+6.3yrs) for age, sex and Pseudomonal status had parameters of clinical status and nutritional intervention recorded annually from six years pre diagnosis of CFRD to two years post. Weight and body mass index (BMI) were lower at all time points to diagnosis of CFRD (NS) but were stable as a % of control values

Intravenous antibiotic treatment intensified, peaking at one year post diagnosis 47.7 days/yr (CFRD) v 34.7 (CF) NS but there was no difference in nebulised antibiotic use. An aggressive clinical approach prevents nutritional decline and delays respiratory decline until the year preceding diagnosis of CFRD

Data show mean (SD) or %

1,2 ; Battezzati

With advancing age insulin secretory defects and insulin resistance cause glucose intolerance and diabetes in an increasing proportion of Cystic Fibrosis (CF) patients. Prediction of diabetes development in CF is made difficult by unique features: many patients are normoglycemic or even hypoglycemic after overnight fast, and there are repeated changes of glucose tolerance status from normal to diabetes and vice-versa, for many years and for unclear reasons. AIM of the study was to detect predictive factors of definitive CF related diabetes (CFRD) development in patients undergoing Oral Glucose Tolerance Test (OGTT) evaluations routinarily. METHODS. Starting from 2002, all patients followed at the CF Center in Milan aged >10 years and without established CFRD undergo OGTT yearly. Among those who received their first OGTT between 2002 and 2004, 14 had developed definitive diabetes by

4 µU/ml, and 1.4±0.1 vs 1.4±0.2 ng/ml). Glucose (p=0.010) and insulin AUCs (p=0.030) were the most important predictive variables, respectively directly and inversely related to CFRD development (pseudo R 2 for the model: 0.581, p<0.001). Glycated hemoglobin and baseline glucose concentrations were directly related with outcome at univariate analysis whereas c-peptide concentrations were inversely related. In contrast, no relationship emerged between insulin-sensitivity indexes and outcome. Anthropometric (weight, height z scores, BMI) and pulmonary function indexes were also unrelated. CONCLUSIONS. Insulin secretory defects are an important determinant of subsequent CFRD developement

Response rate was 18/18 in group A and 17/18 in group B. Group A included 5 males and 13 females, mean age was 52 months, SD 22.34; group B included 4 males and 13 females, mean age 52 months, SD 20.91. Followgroup B parents do ("poor-very poor" health 6% vs 47%, p<0.002), but over time health perception in parents of children with CF improves, and the gap vanishes ("poor-very poor" health in last year 6% in group B). Post-diagnosis anxiety, depressive symptoms, sleep disturbances, mood changes, and nervousness are less frequently reported in group A (62% vs 94% p<0.002), and the diagnosis was considered to affect the parent-child relationship more in group B than in A (p<0.01). The two groups did not differ in the assessment scores of Internalizing, Externalizing and Total Behavioural/Emotional Problems. However, group B performed worse than the general control population average (mean 12.65 SD 6.72 vs mean 9 SD 6.6, p<0.04). Thirty-five% of parents in group A and 53% in group B changed their family planning projects following the diagnosis

PHYSICAL HEALTH AND ANXIETY SYMPTOMS: DOES MONITORING MEDIATE THE RELATION?

Results: Physical health, monitoring, and anxiety symptoms were related. Children who perceived themselves to be relatively healthy were less likely to have an information processing style characterized by high levels of monitoring (r = -.39, p < .05), and they reported fewer Trait Anxiety symptoms (r = -.50, p < .01). Monitoring, as hypothesized, was associated with increased Trait Anxiety (r = .52, p = .001). Low BMI, a potential indicator of poor nutritional status and physical health among youth with CF, also was related to higher monitoring (r = -.34, p = .05) and Trait Anxiety (r = -.29, p = .10). Tests of mediation indicate that monitoring partially mediated the relation between poor health and Trait Anxiety for both child report of physical health (Goodman test: z = 1.86, p = .06) and for BMI (Goodman test: z = 1.67, p = .09), but not when parental report or FEV1 were used to assess youth's physical health, or when parental report of youth's internalizing symptoms was used as the outcome. Conclusion: Youth who tend to generally scan for and are attentive to changes in their environment, which may include their internal physical environment, were found to report more anxiety. Moreover, this monitoring style appears to partially mediate the relation between physical health and anxiety among youth with CF. If monitoring is found to be a risk factor leading to an increase in future anxiety symptoms among youth with CF

Data was analysed using a thematic analysis: "framework" 3 Results: Young people say that befriending is fun, offers opportunities for new experiences, a confidant outside the family, and gives them a boost psychologically. Carers see befriending as confidence building for their children, providing time out for themselves, and helping with the big questions. Befrienders see their role as mentoring, broadening young people's horizons and providing a safe place physically and emotionally. Challenges include: Forming and ending relationships, having multiple befrienders, ongoing support and training for befrienders, maintaining boundaries, sibling rivalry, and cost. Conclusions: Befriending is a new innovation in CF, and has the potential to make a difference to young people's lives. Careful planning at the outset, ongoing support for befrienders and regular evaluation are essential factors in ensuring its success

Good practice in befriending services for people with learning disabilities

Qualitative Research Practice Sage London 552ଙ

In addition, there was high use of non-CF vitamins (n=102) and protein shakes (n=88) which were not included as a CAM therapies. When prayer was excluded (which was used by 63% of respondents), 53.1% of the patients are still using some form of CAM. The most commonly used CAM therapies in CF were relaxation therapies (n=83), massage therapy (n=61), chiropractor care (n=59), herb/plant product therapies (n=53), homeopathy (n=19), and yoga (n=19). An analysis of the relaxation therapies revealed that deep breathing exercises (n=80) were the most common and frequently combined with other relaxation techniques: meditation (n=43), progressive relaxation (n=31) and guided imagery (n=24). The most common herb/plant products used were: echinacea (n=33), garlic (n=21), selenium (n=20), ragweed or chamomile (n=17) and ginseng (n=14). The use of possible CF specific CAM therapies were fish oils/omega fatty acids (n=37), glutathione (n=13), docosahexaenoic acid (n=11) and curcumin (n=9). In conclusion, CAM is widely used by the CF population with "prayer for health" being the most common modality, however, many patients are utilizing multiple CAM therapies (data not presented)

Within the framework of a weighted satisfaction model of quality of life, we investigated the importance ratings of adolescent and adult patients regarding disease-specific aspects of living with CF. Method: 108 outpatients (aged 15-47 years, M=27.0, SD=7.9; FEV1 20-125%, M=62.6, SD=26.2) repeatedly filled in the CF-specific module of the Questions on Life Satisfaction (FLZ-CF) to measure satisfaction with nine CF-specific aspects of life in relation to the subjective importance of each life domain. A ranking list of the most important life domains across the study group was determined, and intra-individual changes of the importance scores were analysed. Associations of importance ratings with changes of pulmonary functioning (FEV1%) were examined. Results: The most important aspects (% of "very important" or "extremely important" answers) are sleep (96.2%), integration of therapy into daily routine (94.4%), breathing (89.8%), gastrointestinal functioning (86.9%), and eating (86.9%), less important is understanding by others (44%)

1,5 ; Durieu, I. 4,5 1. Département d'information médicale, Hospices Civils de Lyon

Dr. von Hauner Children`s Hospital

Helios Hospital, E.v. Behring, Berlin, Germany Supported by the Cystic Fibrosis Foundation Leroy Matthews Physician Scientist Award, the National Heart, Lung, and Blood Institute

Cystic Fibrosis Foundation Therapeutics Inc. Symptoms 571 PATIENT-REPORTED RESPIRATORY SYMPTOMS IN CYSTIC FIBROSIS: INITIAL VALIDATION

Children's Hospital and Regional Medical Center

The first author is supported by a Second year clinical fellowship from

Razvi, S. 1 ; Quittell, L.M. 1 ; Sewall, A. 2 ; Marshall, B.C. 2 ; Saiman, L. 1 1. Pediatric Pulmonary Medicine, Columbia University, New York, NY, USA; 2. Cystic Fibrosis Foundation, Bethesda, MD, USA Significance: Significant improvements have been made in diagnostic and therapeutic strategies for CF patients in recent decades. We hypothesized that these changes could potentially impact CF respiratory microbiology. Thus, we examined longitudinal trends in the annual incidence and prevalence of CF respiratory pathogens from 1995 to 2005.Methods: CF Foundation Patient Registry data, estimated to include 90% of the CF population in the U.S., was utilized in this analysis. Patients were included if results from at least one respiratory culture were in the Registry from January 1995 to December 2005. Patients were excluded after organ transplantation. To avoid misclassification of incident cases, a retrospective review of Registry data from 1985 to 1994 was performed to establish the culture status of included patients prior to 1995. Thus, incident cases were subjects with first detection of a given pathogen in a given year. Prevalent cases were defined as subjects with at least one positive respiratory culture for a given pathogen in a given year. All submitted culture results were included.Results: The number of CF patients submitting Registry data increased from 19,735 in 1995 to 23,347 in 2005 . The proportion of subjects meeting inclusion criteria remained relatively constant (85% to 91% per year). The median age of the study cohort increased from 13.1 years in 1995 to 15.1 years in 2005. During the study period, the incidence of Hemophilus influenzae remained stable (10.3% to 10.6%) as did the prevalence (15.3% to 17.3%). The incidence of Pseudomonas aeruginosa ranged from 20.8% in 1996 to a peak of 24.2% in 2004. The prevalence of P. aeruginosa declined from 60.4% in 1995 to 56.1% in 2005. There is a trend for both an increasing incidence and prevalence of Staphylococcus aureus; the incidence increased from 21.7% in 1995 to 33.2% in 2005 and the prevalence increased from 37.0% in 1995 to 52.4% in 2005. The age specific prevalence of S. aureus remained highest in children aged 6-17 years. The incidence of methicillin-resistant S. aureus (MRSA) increased from 0.1% in 1995 to 7.0% in 2005, with a parallel increase in prevalence from 0.1% to 17.2%. The highest prevalence of MRSA was noted in subjects 18 years of age and older. The incidence of Burkholderia cepacia complex decreased from 1.3% in 1995 to 0.9% in 2005, while the prevalence remained relatively stable (range 2.9% to 3.6%). Both the incidence and prevalence of Stenotrophomonas maltophilia increased (incidence: 2.6% to 6.4%, prevalence: 3.5% to 12.4%).This retrospective study aims at assessing the respiratory function and the cystic fibrosis (CF) co-morbid conditions according to the patients' nutritional status.Method: data were collected from the French Observatory (ONM) 2004 (n=4533) (VLM, INED). The last year data for weight (W), height (H), FEV1, FVC were extracted from the database (exclusion criteria: transplan-tation); were also analysed the genotype, pancreatic enzymes consumption, pseudomonas aeruginosa (PA) colonization, cirrhosis, diabetes, pulmonary transplantation and mortality (SAS 9.1). We used international age-adjusted standards for BMI z-score for children (C) and BMI for adults (over 18 years) (A) (Cole TJ BMJ 2000) to define underW (UW), normal W (NW), overW (OW) and obesity (OB) .Results: data are available for 2647 patients in the C cohort and for 1498 patients in the A cohort (36 %). Nutritional status subgroups prevalence (%) is: 6.1, 87.1, 5.6 and 1.2 in C and 37.3, 57.7, 4. 3 and 0.7 in A for respectively UW, NW, OW and OB. Mean age in A is significantly increasing with BMI (p<.0001). Frequency of 508 del/508 del in C and A is lower in OW/OB (p<.02) as well as the use of pancreatic enzymes (p<.001). OW/OB patients have the best FEV1 and FVC values whatever the gender and the age, with significantly less PA colonization in A (p<.01). We could identify a positive correlation between the pulmonary function and BMI. The CF co-morbid conditions demonstrated a lower prevalence of diabetes (12/257) and cirrhosis (5/257) in OW/OB. None of the OW/OB C or A are on a transplant waiting list versus 1.7 % in UW/NW (n=69/4042) and none died versus 1 % in UW/NW (n=42/4042).Conclusion: The observed prevalence of OW and OB in CF is respectively 5.2 and 1 % whereas the French ObEpi 2006 study collected 29 and 12 %. Our results suggest that increased BMI is associated with better FEV1, FVC, lower prevalence of PA, cirrhosis and diabetes. The potential risks of chronically high BMI have not been studied in this population yet, but justify further investigations in this longer life expectancy cohort. Materials and Methods: We retrospectively identified abdominal CT scans of 38 consecutive patients (19 females, 19 males, mean age of 28 years) with cystic fibrosis and a control group of 38 consecutive patients (21 females, 17 males, mean age of 40 years) scanned as potential renal donors. Three readers reviewed all scans and recorded the presence and location of colonic wall redundancy, and the wall thickness of the ascending, transverse, and descending colon. Clinical information on cystic fibrosis patients, including CFTR gene mutations, was queried from our cystic fibrosis patient registry database. Additionally, medical records of all cystic fibrosis patients were reviewed to determine the indication for abdominal CT.

Results: Colonic wall redundancy was seen exclusively in patients with cystic fibrosis, and was noted in 11 of 38 patients each for reviewers 1, 2, and 3 (p<0.05). Colonic wall redundancy was seen in 11 of 28 adult patients with cystic fibrosis (39%), but was not seen in any children age 17 or younger (p<0.05). Excellent agreement was found for the CT identification of colonic wall redundancy among readers (kappa=0.91, p<0.001). Cystic fibrosis patients with colonic wall redundancy had significantly thicker ascending colonic walls (mean 4.0 mm) vs. those without wall redundancy (mean 1.8 mm) or controls (mean 1.2 mm), (p=0.03). Three patients with colonic wall redundancy had follow-up CT, and all showed temporal stability (mean of 50 months). Among adult cystic fibrosis patients, CFTR gene mutations were available for 10 of 11 patients with and 13 of 17 without colonic wall redundancy. While the common ∆F508 mutation was the predominant mutant allele among patients with normal colons at CT (only 3 of 17 patients, or 23%, had an identified mutation other than ∆F508 on either allele), a higher prevalence of less common non-∆F508 mutations was seen patients with colonic wall redundancy (7 of 10 patients, or 70%, p<0.05). The G542X mutation was seen exclusively in patients with colonic wall redundancy (3 of 10 patients, or 30%, p=0.06). There was no significant difference in the proportion of patients with abdominal pain (>0.7), pancreatic insufficiency (p>0.99), diabetes mellitus (p>0.1), history of meconium ileus Background: Airway inflammation in CF is associated with marked remodelling and bronchiectasis. Pro-inflammatory mediators such as advanced glycation end-products (AGEs) and the soluble receptor for AGE (sRAGE) may perpetuate this response in the lung and in other organs such as the kidney 1 . The total body burden of AGEs reflects exogenous sources from the diet, endogenous production by the body, tissue degradation and renal clearance, which may be reduced in renal impairment. The accumulation of endogenous AGE is accelerated in conditions of high oxidative stress and inflammation, and AGEs have been implicated in the pathogenesis of diabetic nephropathy. The burden and significance of AGEs in CF has not been determined, but if elevated, dietary modification of AGEs may represent a novel anti-inflammatory approach for CF. The aim of this study was to determine serum levels of AGE and sRAGE, and identify clinical correlates of AGE and sRAGE levels.Methods: Adults with CF (n=58, 35 males, 86% pancreatic insufficient, mean age 33.9±7.7 years (range 23 to 62 years), mean FEV 1 %predicted 58.7±22.6%predicted (range 20-110)) and healthy adults (n=24, 13 male, mean age 32.7±9.2 years (range 21-55 years)) were studied. CF participants provided 1-3 serum samples each over a six month period, while healthy controls provided a single sample. Serum was analysed for levels of advanced glycation end-product (AGE-CML) and soluble receptor for AGE (sRAGE) (ELISA). For each CF participant, levels in the multiple samples were averaged, and the median for the study sample reported. Clinical data, including BMI, presence of CF-related diabetes mellitus (CFRD) and serum HbA1c level were collected from the medical record. Mann-Whitney tests were used to compare CF and control levels, while Spearman rank correlations were used to identify clinical correlates of AGE and sRAGE.Results: Prevalence of CFRD was 27.6%. Mean BMI was 21.8±2.7kg/m 2 and mean HbA1c level was 5.9±1.1%. Median (IQR) levels for AGE-CML were 1650 (1158, 1986) BACKGROUND: Osteopenia is diagnosed in cystic fibrosis (CF) using Dual-energy X-ray absorptiometry (DXA). Areal BMD from DXA is subject to error when bones are smaller in volume than reference standards (T-scores). Normalization of bone size by use of Zscores in CF is controversial and not widely utilized, thus comparing to larger bone areas. Reports in CFTR-deficient mice (CF mice) reveal osteopenia when measured by DXA. We hypothesized that use of pQCT, which eliminates bias of size, would more accurately analyze BMD.METHODS: Femurs were collected at necropsy from 4 CF mice and 4 C57BL/6J (B6) mice at 6 wks, and 7 CF mice and 7 B6 mice at 14 wks (all female) for pQCT. Time points were chosen to coincide with pre-pubertal and adult ages for comparison to human disease. Total mineral, trabecular and cortical densities were measured. Student's t-test was used to detect significant differences (p<0.05).RESULTS: Femur measurements from both CF mice and B6 mice are listed as mean ± SD in the table (below). Length was greater in B6 mice compared to CF mice at 6 and 14 wks. Total area at the metaphysis and diaphysis were greater in B6 mice at 6 and 14 wks. Data are consistent with larger bones in B6 mice. Total mineral density 1 , trabecular density 2 and cortical density 3 , however, were greater in CF mice compared to B6 ( 1 6 & 14 wks, 2 14 wks, 3 6 wks).DISCUSSION: Our study demonstrates greater BMD in CF mice when volumetric data are analyzed and size differences are accounted for by use of pQCT. These findings persisted at adult age, suggesting a normal deposition of bone. In clinical management, DXA imaging is readily available compared to pQCT, and can be used as a predictor of BMD and fracture risk. Correct reference ranges must be utilized to minimize erroneous values secondary to size. Z-scores allow correction for differences in bone area and size, even in adults. More studies into understanding bone mineral deficit and tools of measurement are needed in CF. Meanwhile, bone measurements by radiographic imaging must be taken in context of overall health, pubertal progression, and size of individuals. Gainesville, FL, USA; 2. Nemours Children's Clinic, Orlando, FL, USA; 3. SUNY Upstate Medical University, Syracuse, NY, USA; 4. Case Western Reserve University, Cleveland, OH, USA; 5. Genentech, San Fancisco, CA, USA Background: Greater growth rates, BMI and fat-free mass are associated with improved lung function in individuals with CF (Pedreira, et al. Pediatr Pulmonol. 2005) . Current treatments for weight gain have focused on nutritional supplements and appetite stimulation, and not underlying issues of catabolism or chronic disease. We hypothesized that anabolic effects of GH would not only improve weight and height, but LBM as well.Methods: Sixty-seven prepubertal children with CF and height less than or equal to the 10th percentile were randomized to daily GH (n=35) or observation (n=32) for a period of 1 year, followed by an off-treatment observation period of 6 months. Children randomized to GH received somatropin injections daily at 0.3 mg/kg/wk. Height and weight were measured every three months. Height was evaluated as height standard deviation score (SDS) to control for differences in age and sex. LBM was measured by DEXA scan at time zero, at 6 and 12 months, and at the end of the 6-month observation period. In addition, for LBM, change from baseline was calculated for subjects for whom the same DEXA equipment was used at both time points. The preliminary results of the first 12 months of the study are presented here.Results: Data for 27 subjects in the observation arm and 29 subjects in the GH arm, who completed 12 months, are available. As shown in the table below (listed as mean ± SD), gain in height, weight and LBM were significantly greater in the GH treated group than in the observation group over the 12 month period.Discussion: GH improves growth in prepubertal children with CF as measured by height SDS. In addition, GH significantly improves weight gain and this gain is, in part a result of the significantly greater increase in LBM in the GH-treated group than in the observation group. The relationship between the improvement in LBM and other outcomes in CF deserves further exploration. The National Cystic Fibrosis(CF) Registry Database notes Cystic Fibrosis Related Diabetes Mellitus(CFRD) as a complication in 15.6% of CF patients at all ages. CFRD is associated with poorer nutrition and increased pulmonary morbidity. The cause of CFRD is not completely known but insulin resistance may be associated with fibrotic damage to pancreatic islet cells due to chronic inflammation. Other studies suggest protein energy malnutrition(PEM) in early life leads to impairment of insulin secretion and Cystic Fibrosis-related diabetes (CFRD) accounts for increased morbidity and mortality in patients with CF and occurs in approximately 30% of patients by the age of 30 years. CFRD NET (Network for Epidemiology and Trials) is a consortium of four large UK CF centres caring for 1052 adult patients with CF. The aim of this group is to undertake research into the diagnosis, investigation and management of CFRD. In this abstract, we estimate the prevalence of diabetes in a screened population of adults with CF.

Methods Annual review data were collected on attending patients with CF aged 16 and above during 2006. 75g OGTTs were performed after fasting for at least 8 hours on patients without known diabetes. Each OGTT was categorised as either normal (2 hour glucose 3.5 -7.8mmol/l), impaired (2 hour glucose ≥ 7.8mmol/l and < 11.1mmol/l) or diabetic (2 hour glucose ≥ 11.1 mmol/l). All patients with a diabetic OGTT were followed up with serial BM monitoring to determine whether or not they had CFRD. Fasting plasma glucose (FPG) was considered to be elevated ≥ 7.0 mmol/l, isolated impaired fasting glucose (IGF) was defined as a value between 6.1 mmol/l and 7.0 mmol/l and hypoglycaemia was defined as a blood glucose <3.5mmol/l. In addition, we obtained information on age, sex and prescription of anti-diabetic medications subsequent to OGTT. In three of four centres, HbA1c was routinely performed on all patients.

Of 1052 patients (median age 26 years, 57% male), 344 (33%) had established diabetes and were therefore excluded from further screening. Of the remaining 708 patients, 392 (55%) underwent formal OGTT testing. In this latter group 34 (9%) had a diabetic OGTT, 58 (15%) an impaired OGTT and 49 (13%) had evidence of reactive hypoglycaemia at 2 hours. Of the 34 patients with a diabetic OGTT, 28 (82%) only had an abnormal 2 hour value, 4 (12%) only had an elevated FPG and 1 (<1%) had both. All patients with diabetic OGTTs underwent blood glucose monitoring and 20 (59%) went on to treatment with hypoglycaemic agents in the calendar year. None of the 4 patients with an isolated elevated FPG had diabetes.HbA1cs were available in 23 of 34 newly diagnosed patients with diabetes (in the centres which performed HbA1c). In this group median HbA1c was 6.1%.

This study confirms the high prevalence of diabetes among screened patients and the growing burden of diabetes management in adult cystic fibrosis clinics. It further highlights the importance of performing screening for diabetes in this population; the majority of patients were identified on the basis of abnormal 2 hour values. Despite this only half of all suitable patients underwent OGTTs in the four centres committed to screening. These patients merit further study, as previous work suggests that the decline in clinical status occurs several years before diabetes becomes apparent. Cystic Fibrosis related Diabetes (CFRD) occurs in up to 40 % of patients with cystic fibrosis (CF), the incidence rising with increasing age. The likelihood of developing long-term complications secondary to diabetes increases with poor glycaemic control and duration of diabetes. As survival has improved for people with CF those who develop CFRD may live with diabetes for several years.

Aim: To establish the frequency of diabetic complications in patients with cystic fibrosis related diabetes.Method : Patients with CFRD attending the Adult CF service at the Royal Brompton Hospital between April 2006-March 2007 were screened for diabetic complications and cardiovascular risk factors.

A total of 78 patients (male/female: 44/34) including 10 post transplant patients were screened. Mean age was 33.1 years (17-52), mean HbA1c 8.0% (5.6-16.2% ) and average duration of diabetes was 7.3 years (<1-38 years).As immunosuppressive therapy can also cause many of the complications associated with diabetes the results are presented separately for nontransplant and post transplant patients (table 1) . 9 patients in total had retinopathy: 7 background retinopathy,1 proliferative retinopathy undergoing laser therapy and 1 maculopathy. The average duration of diabetes in those with background retinopathy was 5.8 years with only 1 patient having diabetes for >10 years. The patient with proliferative retinopathy (non-transplant) had been diabetic for 15 years. A raised creatinine level was identified in all transplanted patients with microalbuminuria but none of the non-transplanted patients. 1 patient (nontransplant) had macroalbuminuria. 2 post transplant patients were on treatment for hypertension and a further 6 patients (2 post transplant and 4 nontransplant) had elevated blood pressure (>140/90) at screening requiring follow-up. None of the patients had evidence of cardiovascular disease or stroke.Conclusion: Macrovascular complications were not seen. Microvascular complications occurred but were less common than the reported incidence in type 1 and type 2 diabetes. This may reflect the relatively short duration of diabetes (mean 7.8 years) of patients in this study. A further study comparing CFRD patients with a non-CF diabetic control group of similar duration of diabetes is warranted. Cystic fibrosis (CF) is a disease that leads to serious disturbances in nutritional status and bone calcification. Comparison of two methods for assessment of bone mineralization: DEXA and hand radiograms in diagnosing osteopenia or osteoporosis. Study was performed in a group of 26 CF patients (10F, 16M), aged 7-30 yrs. Nutritional status was assessed using BMI, Cole's Index and BMC. Radiograms of non-dominant hand were assessed according to normalized optical density comparing to aluminum standard. Bone density was also assessed using DEXA. For statistical analysis, backward stepwise binary logistic regression (Wald's test) was used. Analysis of data revealed that using BMC, Cole's index and hand radiograms markers we can diagnose bone mass disturbances (Z score <1SD) with precision up to 84.62% comparing to DEXA. Sensitivity and specificity of this method was respectively 86.67% and 81.82%. False negative results were obtained in 2 patients and false positive were also in 2 patients. Hand radiograms method could be an alternative for DEXA in screening of bone density disturbances in CF patients. The study was partly supported by grant of Ministry of Science and Higher Education no 2 P05E 041 28.In a very recent publication(1) a high rate of fasting (13%) and reactive hypoglycaemia (15%) was described in a group of n=129 CF patients older than 8 years who received an oral glucose tolerance test. Reactive hypoglycaemia was related to the 120minute glucose concentration in the OGT test. We were surprised by the high percentage of asymptomatic hypoglycaemic situations in CF patients. As a part of a prospective intervention study in CF patients with CFRD, we screened more than 1400 patients 10 years or older with CF for CFRD. In this multi-centres population OGT was performed in the morning after overnight fasting according to WHO standards. Using the same definition for fasting hypoglycaemia (glucose < 60mg/dl; 3.3mmol/l) and reactive hypoglycaemia (glucose < 50mg/dl; 2.8mmol/l at 120minute during OGT test)as in the other study we calculated the percentage of CF patients with fasting or reactive hypoglycaemia in our cohort. Results: OGT were done in 1495 patients with CF. Age(mean±SD)19,6±8,6 years; BMI Z-score -0.73±1,1;height Z-score -0,8±1,1 and weight Z-score -1,0±1,3. OGT was categorised according to ADA criteria. Normal OGT (n=782 all ; fasting hypoglycaemia (FH) n=21(2,7%), reactive hypoglycaemia (RH)n=18 (2,3%),IFG (n=329 all, FH n=0, RH n=5 (1,5%), IGT(n=131,FH n=2(1,5%), RH n=0),FGT (n=75 all, FH n=0,RH n=0) and Diabetes Mellitus (n=187 all, FH n=3(1,7%)). FH was observed in 26 (1,7%) and RH in 23(1,5%) out of 1495 OGT tests. There were no difference related to age, BMI Z-score, height Z-score and weight Z-score comparing those patients with FH or RH to those without FH or RH in all categories of OGT tests.In this large multi-centres cohort of CF patients we were neither able to confirm the high percentage of fasting nor of reactive asymptomatic hypoglycaemia which was reported recently on a small group of CF patients. Nutritional status measured by BMI Z-scores, height and weight Z-scores were unaffected by hypoglycaemia in our cohort as noticed also in the small group. As the OGT tests in the other study were done in a single centre it might by that a centre specific situation influenced the frequency of hypoglycaemia in that small group. We conclude that large numbers of investigations might be needed to come up with firm conclusions related to frequency of specific aspects of glucose disturbance in CF patients.FH does not exclude Diabetes Mellitus or IGT. As survival from CF improves, surveillance to identify and treat complications associated with longevity is an important component of management. Renal disease has been reported in adults with CF. Risk factors may include CF-related diabetes mellitus (CFRD) and use of nephrotoxic medications. Diabetic nephropathy has been observed in the absence of CFRD 1 , highlighting the need for screening for renal impairment. This study aimed to measure renal function in a sample of adults with CF, and to compare estimated glomerular filtration rate (eGFR) and urinary creatinine clearance (UrCrCl) as markers of renal function.Methods: Adults with CF aged 25 years or over (n=29, 66% male, 83% pancreatic insufficient, mean age 36.7±9.1 years) underwent screening for renal impairment. Of these patients, 15 (9 males) had CFRD, and 14 (10 males) had no history or clinical features of CFRD. 24 hour urine collections were analysed for UrCrCl, with renal impairment being defined as <90ml/min. Serum creatinine level was used to calculate eGFR (MDRD formula), which was classified renal function as normal or mildly impaired (>60ml/min); moderately impaired (30-59 ml/min) or severely impaired (<30ml/min). Sensitivity of eGFR for identification of renal function was calculated using urinary creatinine clearance as the reference method. Unpaired t-tests were used to compared CFRD with non-CFRD patients.Results: Mean UrCrCl was 105±35ml/min, and 8 eight patients (27.6%) had renal impairment (<90ml/min). Of these, only 3 had eGFR suggestive of renal impairment. The sensitivity of eGFR for identifying renal impairment in this sample was 38%. A further 2 patients had moderately impaired eGFR, but normal UrCrCL. Five of the 8 patients with impaired UrCrCl had CFRD. Age, gender, CFRD, FEV 1 %predicted and HbA1c level did not correlate with UrCrCl. The table compares CFRD patients with non-CFRD patients. HbA1c level was higher in CFRD patients. There were no other significant differences in clinical or renal function parameters.Conclusion: Renal impairment, is common in the adult CF population and is not confined to patients with CFRD. Screening using eGFR is poorly predictive of impaired UrCrCl in this population. These results suggest that surveillance to monitor renal function is indicated in the adult CF population, including in patients without CFRD, and that determination of UrCrCl should be included in the screening process. Cystic fibrosis (CF) patients suffer from pancreatic insufficiency resulting in malabsorption of fat soluble vitamins including vitamin D. Many fac-tors contribute to low bone mass and fractures in patients with CF; however, chronic vitamin D deficiency plays a major role. The prevalence of vitamin D deficiency has been reported as high as 81% in some specialized care centers. In addition, reports of occult vertebral fractures have been reported as high as 25%. We sought to determine the prevalence of vitamin D deficiency (defined as 25-hydroxyvitamin D (25(OH)D) < 30 ng/ml) and of vertebral fractures at our CF center. We obtained IRB approval to review the records of all patients seen at our CF center during [2004] [2005] . We collected information related to bone health including 25(OH)D, bone mineral density and lateral spine or chest x-rays to examine for the presence of a vertebral fracture. We reviewed the records of 185 subjects who were seen at our center during the study period. The mean age of the subjects was 29 ± 9 years. Subjects had a mean BMI of 21.2 ± 3 and an FEV1% Predicted of 64.4 ± 25%. The percentage of subjects having an annual 25(OH)D level checked was only 57% and 62% in 2004 and 2005, respectively. The mean 25(OH)D was 22.7 ± 10 and 24.9 ± 10 ng/ml in 2004 and 2005. The prevalence of vitamin deficiency was 78% and 73% in 2004 and 2005. About one quarter of the subjects had bone mineral density testing with half of the tests showing osteopenia or osteoporosis. Twenty-seven percent of subjects had vertebral abnormalites detected on lateral chest x-ray. We sought to determine whether any factors were associated with vitamin D deficiency. We found that taking a multivitamin did not significantly protect against vitamin D deficiency. However, not taking a supplement containing vitamin D other than a multivitamin was associated with a 54% risk of vitamin D deficiency (p=0.04). Subjects having 25(OH)D levels determined in the winter or spring was associated with a 30% risk of vitamin D deficiency (p=0.05). In summary, we found annual testing for vitamin D status was inadequate in our CF center and that when 25(OH)D level was determined, over 70% of subjects were vitamin D deficient. Nearly one quarter of our adult patients already had evidence of vertebral compression fracture seen in lateral chest x-ray. We urge greater screening for vitamin D deficiency in the CF population. Effective protocols to prevent and/or treat vitamin D deficiency are urgently needed in the CF population. Improved vitamin D status in CF patients is one factor that may reduce the high prevalence of vetebral fractures. Support for this study was provided by Proctor and Gamble Pharmaceuticals Background: More CF patients are surviving into adulthood, in part due to the increasing use of new therapies and more aggressive management of chronic respiratory and GI disease. As a result, the recommended daily treatment regimens for most CF adults are both complex and time consuming. Objective: To assess the self-reported daily treatment burden of CF therapies in a cohort of adults with CF.

Methods: In the sixth survey wave of the Project on Adult Care in CF (PAC-CF), an ongoing longitudinal panel study of adults with CF from 10 US CF Centers, respondents were asked to report the type of medications, inhaled therapies, and airway clearance therapies they used during the day prior to completing the survey, as well as estimate the time generally required to complete each type of therapy.Results: Of the 204 respondents (response rate 69%), the median number of therapies reported was 7 (IQR 5-9) and the median reported amount of time usually spent on treatments was 90 minutes per day (IQR 50-150 minutes). Forty-eight percent reported using 5 or more different inhaled therapies or using inhaled therapies 5 or more times during that one day. The most commonly reported inhaled therapies were a bronchodilator (61%), Pulmozyme (49%), an inhaled steroid (32%), hypertonic saline (30%), and Objective: Caring for a child with CF is stressful and often leads to adverse effects for both children and caregivers. Support from family and friends can help reduce the adverse effects of chronic stress, however, there is little research documenting types of support provision as well as sources of support. Moreover, social support may be related to important health outcomes. This study described the types of support provided to caregivers of children with CF and examined relationships between support and health outcomes.Method: As part of a larger intervention study to improve adherence to medical regimens, 89 children with CF ages 1 to 11 and their caregivers were enrolled across 3 CF Centers in Florida: University of Florida, Nemours Children's Clinic, and All Children's Hospital. Caregivers completed the Norbeck Social Support Questionnaire (NSSQ; Norbeck, 1980) in addition to other measures at Baseline and Follow Up. This questionnaire provides information about the number and type of people providing support, as well as its type and quality.Results: Caregivers received the majority of support from their families, spouses and friends and characterized these relationships as longer in duration, with more frequent provision of support. Other individuals who provided support included the CF team, religious leaders, counselors or psychologists, and work associates. Overall, caregivers rated family members and spouses as providing the greatest amount (i.e., "quite a bit") of emotional support. Spouses received the highest ratings for tangible support; they were perceived as providing "quite a bit" of support in comparison to families (i.e., moderate) and friends (i.e., a little). In terms of size of the support network, caregivers in this study had similar networks to those of a normative sample of healthy adults. In addition, there was no difference in amount of emotional or tangible support received by these caregivers. Preliminary baseline results indicated that social support was positively related to both adherence, children's growth (i.e., height and weight percentile), and family income.Discussion: Caregivers of young children with CF reported similar support networks as other healthy adults. However, there was variability in the amount of support provided by source, with spouses providing the greatest amount. Social support appeared to be a protective factor, and lack of support was related to negative health consequences for their children. More attention should be focused on the potentially beneficial effects of social support.Funding was provided by NIH grant #69736

Befriending is reported to be valued by those who have been befriended, offering opportunities for social activities and new experiences and can impact positively on self confidence and self esteem 1. Whilst government policy supports befriending, few schemes collect evidence to demonstrate the effectiveness of these services 2. Young people with CF may suffer social isolation due to chronic illness and treatment demands and a befriend-ing service was offered to those considered socially vulnerable in the Edinburgh area.This project aims to evaluate the effectiveness of this befriending service on young people with CF and their carers. Methods Liverpool, United Kingdom; 2. Child Mental Health, University of Liverpool, Liverpool, United Kingdom; 3. Child Health, University of Liverpool, Liverpool, United Kingdom; 4. Psychology, University of Miami, Miami, FL, USA; 5. Psychology, Birkbeck College, University of London, London, United Kingdom; 6. Mathematics and Statistics, University of Lancaster, Lancaster, United Kingdom; 7. Psychiatry, University of Manchester, Manchester, United Kingdom Introduction: Existing treatments for cystic fibrosis (CF) are time-consuming and labour intensive and biomedical advances are likely to lead to further novel interventions. There is concern amongst clinicians that a high care burden associated with these interventions may compromise the wellbeing of caregivers, reduce adherence to the treatment protocol and increase the likelihood of inadvertent errors in the delivery of interventions. At present, there is no measure of 'treatment burden' and therefore no way to assess the impact of new and increasingly more complex interventions conducted by lay caregivers at home. Furthermore, trials are hampered by the lack of measurable patient-reported outcomes beyond a broad quality of life assessment. To address this gap we have developed a measure of treatment burden for CF using qualitative methods (focus groups, action research, in-depth interviews) with participating parents and a working group of CF team professional staff. This measure addresses the time, effort, meaning and ease of management for caregivers of children with CF up to 13 years and post the first year following a diagnosis. Methods: We report here the pilot phase of the validation of this instrument, which involved (i) cognitive interviewing with n=16 caregivers; and, (ii) n=30 caregivers completing the CLCF instrument together with a quality of life measure at 2 time points. This yielded data for stability, reliability and coherence of the hypothesized constructs. Results: The CLCF takes 15 minutes to complete. Face validity and acceptability are established. For the first pilot sample the age of the child ranged from 1.5-11.5 years. Treatments took, on average 72 minutes per day to complete and an average of 16 minutes of this was devoted to administering enzymes particularly for carers of infants. A prominent concern amongst these parents was their child's height and weight and their efforts were directed at optimising growth. For the child however, management of nebulised medications and physiotherapy were more frequently flagged as a concern. These data suggest this is a reliable, coherent and useful, although complex tool. Conclusion: The caregiver challenge for CF cannot be understood simply in terms of time and effort involved in delivery of treatments. It also involves contextualising interventions for their meaning at multiple levels of explanation for the individual concerned. Structural equation modelling would be an appropriate way to proceed with the main validation when such complex relationships occur between latent variables and would serve as a confirmatory factor analysis for the hypothesised relationships. Sponsored by: Royal Liverpool Children's NHS Trust and University of Liverpool.

Funded by: National Institute for Health Research (NIHR) Research for Patient Benefit Programme.Introduction: The French CF practice recommendations, published at the end of 2002 in parallel with the creation of CF reference centres, advise that each patient should be seen at least every 3 months at a cystic fibrosis reference centre.Objective: To investigate the impact of these recommendations on the effectiveness of the follow-up of patients at the four reference centres in the Rhône-Alpes region.Methods: All patients with cystic fibrosis attending one of the four CF centres between 1996 and 2005 were retrospectively included. The total number of visits was recorded for each patient and each year of the study period. To determine the evolution slope for each patient followed for at least 3 years, a negative binomial regression of number of visits versus time was carried out (confirmed with a repeated model). To estimate the impact of the recommendations, the analysis was restricted to patients with at least two visits before and after 2003 and a second model was adjusted with a new intercept in 2003 to estimate the change in slope as of this point.Results: A total of 650 patients were included in the cohort. The average number of visits per patient rose from 3.68 in 1996 to 4.96 in 2005 (p<0.0001). The proportion of patients with at least 4 visits per year increased from 38% to 70%. The negative binomial regression for the 537 patients having had at least 3 years follow-up confirmed this trend with an average slope of +0.07 (SD=0.18). A total of 352 patients were evaluable for the change in slope. No significant change in trend was observed in 2003: only 34% of patients had a higher rate of growth (this change was significant for only 7 subjects). At the last follow-up visit, patients with increasing rates of number of visits were older (21yrs vs. 17yrs, p<0.0001), had lower %FEV1 (68 vs. 75, p=0.02), had a similar average number of visits before 2003 (3.7 vs. 3.9, p=0.33), and a similar weight-for-age z-score (-0.58 vs. -0.46, p=0.50) .Conclusion: The number of visits per patient is regularly increasing. Since the publication of the practice recommendations in 2003, the growth has tended to slow down: clinicians were already convinced by the need for closer follow-up and had begun to increase the rate of visits. Methods: Patients > 13 years who presented for routine healthcare voluntarily completed four surveys: 1) the revised Eating Attitudes Test (EAT), validated for CF patients by Abbott and colleagues, where higher scores reflect worse attitudes towards eating; 2) the Rosenberg Self-Esteem Scale (RSE), where higher scores reflect better self-esteem; 3) the Body Image Scale (BIS), developed for CF patients by Wenninger and colleagues, where higher scores reflect better body image; 4) the Cystic Fibrosis Questionnaire (CFQ), developed by Quittner and colleagues, where higher scores reflect better HRQOL. Also, FEV1%, body mass index (BMI), and pancreatic sufficiency or insufficiency (based on the need for pancreatic enzymes) was recorded. Regression analyses controlling for age, gender and BMI were used to examine the associations between the surveys.

Results: This study included 37 patients with 41 % males and a mean age of 26 years. The EAT was negatively associated with the RSE (p=0.015, adjusted R 2 =0.127), BIS (p=0.001, adjusted R 2 =0.221) and CFQ (p=0.015, adjusted R 2 =0.122). The RSE was positively associated with the CFQ (p=0.001, adjusted R 2 =0.346). The BIS was positively associated with the RSE (p=0.001, adjusted R 2 =0.529) and CFQ (p=0.001, adjusted R 2 = 0.540). Neither BMI nor pancreatic function was associated with the surveys (p=NS). FEV1% was positively associated with the CFQ (p=0.028, adjusted R 2 =0.106) but was not associated with the other surveys.Conclusions: More negative attitudes towards eating predict worse selfesteem and body image, while more positive body image predicts better self-esteem. Also, attitudes towards eating, self-esteem and body image are significant predictors of HRQOL with body image being the most important predictor.Clinical Importance: HRQOL is an important clinical outcome measure in CF. Clinicians need to be sensitive to attitudes towards eating, self-esteem and body image in CF patients, because they are important predictors of HRQOL.

Riekert, K.A. 1 ; Mogayzel, P.J. 2 ; Bilderback, A. 1 ; Hale, W. 1 ; Boyle, M.P. 1 1. Medicine, Johns Hopkins University, Baltimore, MD, USA; 2. Pediatrics, Johns Hopkins University, Baltimore, MD, USA BACKGROUND: Existing research suggests that self-reported adherence to all aspects of the regimen is likely suboptimal and objective measures suggest even poorer adherence. There is little empirical data on how much adherence is necessary to achieve desired health outcomes. Moreover, there is limited data on how adherence changes dur-Objective: CF may be associated with pain attributed to several etiologies. This study evaluates the prevalence of pain symptoms in adult CF patients and the influence on patients` lives.Methods: 243 patients of 6 adult CF Centers in Germany completed a validated, self report questionnaire during a routine clinic visit assessing characteristics of chronic pain (prevalence, duration, location, quality and intensity of pain symptoms). Furthermore, the impact of pain on different aspects of life was explored using a numeric scale from 1 to 10 with 10 being the worst. Every-day-life was divided into 8 categories: duties at home, recreation, social activities, occupation, sexual life, autonomy, vital activities and CF-therapy. The average intensity of pain within the last 4 weeks was correlated to FEV1, BMI and age.Results: 243 patients (119 male) completed the questionnaire. The age was 18-65 years (mean 28.13±8.13). 182 patients (75%) aged 18-65 years (mean 28.51±8.23)reported pain within the last 3 months. BMI was 14.7-38.3 kg/m2 (mean 20.4±3.1), FEV1 12-142% (mean 54.7±24.5). If asked for pain within the last 4 weeks 174 (72%) reported painful episodes lasting from 1-28 days with a mean of 9.8 days (±7.8). Most patients described pain occurring at more than one site with the head being the most localized site, followed by chest and abdomen. Concerning the quality of pain 85.4% characterized their pain episodes as attacks whereas 8.8% reported them as continious, 5.8% had continious pain combined with pain attacks. 179 patients desribed the intensity of pain as 5.77 (±2.35) on a scale from 0 to 10 with 10 being the most severe pain. The category of life negatively influenced most of all by pain episodes was recreation, followed by occupation and duties at home. Female patients were more limited in their acitivites by pain symptoms than male patients with the highest difference being reported in sexual life. There was no correlation of the average intensity of pain within the last 4 weeks to FEV1, BMI or age.Conclusion: The prelevance of pain in CF is often underestimated. Painful episodes can be the cause of worsening the quality of life for adult CF patients. Assessment of pain should be routinely performed as part of care in CF Centers. Objective. Sleep has been examined in a number of pediatric conditions, with impaired sleep resulting in worse neurobehavioral functioning (Beebe, 2005) . Recent research has shown that slow wave sleep is critical for memory (Marshall, Helgadottir, Molle, & Born, 2006) . To the authors' knowledge, no studies have examined slow wave and rapid-eye-movement (REM) sleep in pediatric patients with cystic fibrosis. Methods. Retrospective medical record review revealed two patients who underwent clinically indicated polysomnography (PSG) who reported feeling tired and lacking energy. Patients were females with cystic fibrosis, ages 12 and 14 years. Results. PSG revealed an increase in stage 2 sleep for both patients, resulting in increased risk and observed hypopneic episodes. Patient -A experienced 16.3% in stage 1, 65.6% in stage 2, and 0% in stages 3 and 4 combined. This patient also had no REM sleep. Patient -B experienced 12% in stage 1, 67.8% in stage 2, and 10% in delta wave sleep, with no REM sleep. Respiratory Disturbance Index (RDI) for patients A and B were calculated at 3.6 and 3.5, respectively. Respiratory Effort Related Arousal (RERA) contributed to RDI by 1.9 and 1.1 points in patients A and B, respectively. In addition, endtidal CO2 for patient A and B were maxed at 50 mm Hg and 48 mmHg, respectively. Their Epworth sleepiness scales were scored at 8 and 10 for patients A and B, respectively. Conclusions. Structure of sleep was abnormal in both patients with decreased slow wave (Delta) sleep in patient B and lack of Delta sleep in patient A. Neither patient experienced REM sleep. Compensatory increased stage 2 (NREM) sleep could cause respiratory related abnormalities, such as hypopnea and apnea in patients with cystic fibrosis Background: Understanding and advancing the application of tools to measure patients' symptoms is critical to advancing our evaluation of potential new therapies used to treat CF. We performed initial evaluation of the measurement properties of a CF-specific respiratory symptom daily questionnaire.

Methods: We planned to enroll 60 CF subjects, stratified by age, at three CF programs, the University of Washington Medical Center and Children's Hospital and Regional Medical Center in Seattle and Mary Bridge Children's Hospital in Tacoma in a prospective assessment of a novel CF specific respiratory questionnaire. CF subjects 2 years and older were eligible for enrollment. Patients (parents for children under the age of 12) completed a daily symptom questionnaire during two 7-day periods of clinical stability and one 7-day period when patients were ill for a total of 21 days. The ill state began when patients/parents sought medical attention for respiratory symptoms. The questionnaires were completed using a secure web-based application or via paper for those patients without access to the internet. Two health related quality of life measures (generic and CF specific) were completed by the subjects at the end of each 7-day period to assess the relationship between the novel questionnaire and health related quality of life. Patients also used pedometers during the well and ill states to assess relationship between symptoms and activity level.Results: 52 CF subjects have been enrolled to date. At the time of this interim analysis, 34 patients had completed 384 questionnaires. Of this total, only 2 individual questionnaire entries have been missed in 2 patients (<1% of possible questionnaires). Interim review of cross-sectional data suggests clear differences in symptom reporting between the ill and well periods. Examples include 46% (13/28) noted difficulty breathing and 32% (9/28) noted tightness in the chest in the preceding 24 hours during the initial well period compared to 70% (14/20) noting difficulty breathing and 70% (14/20) noting tightness in the chest during the ill period.Conclusions: Interim evaluation of this novel instrument demonstrates feasibility of deploying this instrument via the internet with an extremely high completion rate. Using a cross-sectional analysis, the instrument can discriminate between the ill and well state. Further data will be presented regarding within patient variability and discrimination of the instrument.Supported by the Cystic Fibrosis Foundation Leroy Matthews Physician Scientist Award, the National Heart, Lung, and Blood Institute (HL72017-03), National Institute of Health (RR-00037-39), Cystic Fibrosis Foundation Therapeutics Inc. Background: Improved communication in today's health care delivery system is a critically important aspect of patient care. Nurses are in an effective position to improve communication due to their close interaction with patients. As a result, they are able to identify potential problems, communicate them to the healthcare team and improve patient care and satisfaction.Hypothesis: Improving communication between the patient/family and the health care team will improve patient/family satisfaction.Method: As a member hospital partnering with the Institute for Healthcare Improvement's (IHI) initiative, Transforming Care at the Bedside (TCAB), The Children's Hospital of Philadelphia (CHOP) introduced the use of Daily Patient Goal Sheets (DPGS) as a vehicle to improve communication between members of the healthcare team, patients and families and the DPGS was refined to meet the needs of an inpatient medical unit to which CF patients are admitted. Using rapid Plan-Do-Study-Act (PDSA) cycles, a multidisciplinary team engaged in small tests of change in different stages. Initially, patient goals and discharge criteria generated in daily rounds were posted in each room on easel paper. Then large "whiteboards" were placed in each patient room for documenting patient goals, discharge criteria, names of the care team members and questions from patients and families. As a final step in the process, we shifted the health care team rounds to the patient's bedside and implemented bedside shift to shift nurse report and safety checks. Patient satisfaction was continuously monitored over this process using the Press Ganey Survey.Results: Over an eight month period, the Press Ganey Survey results for the unit showed an average increase of 10%. We attribute this increase to better information sharing among patient, family and hospital staff regarding the plan of care for the patient's hospitalization. The health care team also had an improved dialog with the patients/families. Conclusion: Patient/family and healthcare team communication and satisfaction were improved using an organized and stepwise PDSA process to develop Daily Patient Goal Sheets on a medical unit. Background: The UAB/CHS pediatric CF center provides care for approximately 300 patients. The Pulmonary Division has 5 full-time RNs and 3 part-time RNs who rotate phone triage for all calls (including CF) of patients seen by the 9 faculty physicians. Historically, patient calls received during office hours are taken by the receptionist and put into the EMR, then returned by the RN on call. The RN contacts one of the staff MD's, gives report of the patient's condition,treatment orders are received, and the patient notified. There are several issues with this phone triage system, which include inconsistencies in RN and MD response as well as differences in documentation.

GLOBAL AIM: To improve the consistency among RNs in phone triage of sick CF patients.METHOD: An algorithm for pulmonary exacerbation phone triage was developed by the CF center director and lead RNs to improve consistency in symptom assessment and treatment plan. A specific list of systemic and pulmonary signs and symptoms was formed including length of illness. If three OBJECTIVE: Patients with CF often require intravenous antibiotics for treatment of pulmonary exacerbations. Patients often receive peripherally inserted central catheter (PICC) lines or totally implantable venous access devices (TIVADs) for venous access. Few studies have examined complications of TIVAD implantation and little published data exists concerning PICC line complications in CF patients. This study sought to assess the complication rates of both TIVAD and PICC lines as well as to identify possible risk factors for developing complications.METHODS: This retrospective study included patients from 3 CF centers in northern New England. Data was obtained from each patient's local medical record, Port CF, and patient interviews. Demographic data was recorded for all CF patients between 1/1/03-6/1/06. For each TIVAD or PICC line, the following information was recorded: type of line placed, history of prior line placement during the study period, patient age, history of use for blood draws, method of line flushing, and status of the line at the end of the study (if still in use). Complications were defined as catheter occlusion, vascular thrombosis/stenosis, infection, or other local inflammatory reactions.RESULTS: Data was collected for 237 pediatric and 155 adult CF patients during the defined study period. Seventy-three TIVADs and 356 PICC lines were placed during the study period in 205 patients. The TIVAD and PICC line complication rates were 33% (24/73) and 6.5% (23/356), respectively. In pediatric patients, 27% (6 of 22) TIVADs and 6.3% (13/205) PICC lines had a complication. Of the 6 TIVAD complications, three were systemic infections and three were catheter occlusions. Of the 13 PICC line complications, there were two venous thromboses, one line occlusion, two systemic infections, and nine minor incidents of localized phlebitis. In adults, complications were recorded in 35% (18/51) of TIVADs and 6.6% (10/151) of PICC lines. Of the 18 TIVAD complications, there were five systemic infections, seven catheter occlusions, and four venous thromboses. Of the 10 PICC line complications, there were three venous thromboses, three line occlusions, and four minor incidents of localized phlebitis. All adult patients who developed a deep venous thrombosis(DVT) associated with TIVAD implantation were homozygous for the deltaF508 mutation. The presence of diabetes or Burkholderia cepacia complex(BCC) lung infection was associated with DVT with odds ratios of 5.91 (95% CI 1.16-30.1) and 5.11 (95%CI 0.93-28.1), respectively. CONCLUSIONS: Complications of PICC lines were uncommon and usually minor. The rate of TIVAD complication observed was more common over the lifetime of the catheter and was similar to previously published reports. We identified potential risk factors for the development of DVT associated with TIVADs and PICC lines, specifically CF related diabetes, BCC infection, and homozygous deltaF508 genotype. The mechanism by which these factors are associated with catheter complications is unclear and warrants further investigation.