key: cord-023017-k6edtg58 authors: nan title: AASLD Abstracts (pp. 282A–382A) date: 2006-02-10 journal: Hepatology DOI: 10.1002/hep.1840380505 sha: doc_id: 23017 cord_uid: k6edtg58 nan of well-characterized human hepatocyte cell lines could facilitate cell therapies such as HTX and bioartificial livers. Toward this goal, we have developed a tightly regulated human hepatocyte line. Methods Human hepatocytes were immortalized with a retroviral vector encoding human telomerase reverse transcriptase (hTERT) and green fluorescent protein (GFP) cDNAs flanked by a pair of LoxPs. After retroviral transduction, a single cell clone was obtained using flow cytometric cell sorting followed by limiting dilution method. The resultant GFP-positive clones were supertransduced with Tamoxifen-inducible Cre recombinase expression cassette, MerCreMer. Recovery of the reverted form of hTERT-immortalized cells was conducted by Tamoxifen treatment and subsequent GFP-negative cell sorting with MoFlo. The expression of hepatocyte markers and albumin secretion was compared before and after Tamoxifen treatment. Transplantation effect of the reverted cells (lx lo9) into the liver were evaluated in a pig model of liver failure induce by an intravenous administration of D-galactosamine (D-gal). Resu1ts:A non-tumorigenic human hepatocyte cell line TTNT-16-T3 was established. Albumin secretion and gene expression of liver markers were increased in the reverted TTNT-16-T3 cells. Transplantation of the reverted cells via the portal vein of pigs treated with D-gal was effective to prolong the survival by providing liver support until spontaneous hepatic regeneration occurred. Conclusion: We here demonstrate reversible immortalization of human hepatocytes using Tamoxifen-induced CrelLox self-excision. The resulting cell line would be highly desirable for research and therapeutic applications. Acknowledgmenk The work was supported in part by Life Science Project of 2lst Century, Japan. Background and aim: Stem Cells play a key-role in tissue homeostasis. Haematopoietic Stem Cells (HSCs) could migrate into the liver and transdifferentiate, becoming a "liver stem cell reserve''. We aimed to analyse: human HSC (h-HSC) ability to engraft into the mouse liver and to contribute to its regeneration after a toxic damage. Materials and methods: NODISCID mice were so divided: A) chimeric not-damaged mice: mice were submitted to irradiation followed by h-HSC intraperitoneally (i.p.) injection; finally they were killed 22, 27 and 31 days after; 8) chimerIC damageD MICE: mice were irradiated, injected with h-HSCs and, 21 days after, damaged using Ally1 Alcohol (AA), i.p.; killing was performed 1 , 5 and 10 days after the damage; C) damaged not-chimeric MICE: mice were damaged (with AA injection) and killed 1, 5 and 10 days after; D) NOT-CHIMERIC NOT-DAMAGED MICE: mice were killed without any treatment. Spleens (S), Bone Marrows (BM) and Livers (L) were tested by flow-cytometry to detect h-HSCs and immunohistochemistry (L only) to localize h-HSCs and their hepatic derivatives. Results: Flow-cytometry: group A) presence of low percentage of human cells; groups C) and D) human cells undetectable; group B) presence of low percentage of human cells in BM and S while in the L they represented up to 20% of whole cellular population (the difference respect to the group A has high statistical significance, being alpha<0.001%). Immunohistochemistry showed the human cell differentiation into parenchymal cells. Conclusions: Our results suggest that HSCs could be an option to improve thie liver regeneration after a toxic injury, supporting the feasibility of BM-derived liver stem cell hypothesis. We have previously demonstrated that CD8-mediated immune damage of allogeneic liver parenchymal cells is difficult to regulate due to the activity of "immunoresistant" CD8+ T cells. In prior studies we demonstrated that short-term interference with CD401 CD40L (but not CD281B7) costimulation transiently suppressed CDB-dependent hepatocyte rejection. Recently, we have determined that targetintg LFA-1 alone preferentially suppresses CD8dependent versus CD4-dependent hepatocyte rejection. In addition, short-term immunotherapy targeting both LFA-1 and CD40lCD40L costimulation produced synergistic effects and resulted in longterm suppression of CD8-dependent rejection such that hepatocyte survival up to 90 days was achieved in the majority of recipient mice. The purpose of this study was to determine whether targeting both CD40lCD40L and LFA-1 mediated signals results in an immunoregulatory state which controls naive CD8+ T cells. Methods: 2 x lo6 FVBlN (hAlAT, hepatocytes, were transplanted into CD4 KO or C57BL16.SCID (all H-2b) mice, and recipients were treated with anti-LFA-1 mAb (0.3mg d0-6) and anti-CD4OL mAb (lmg, do, 2,4,7) . Hepatocyte survival was monitored by detection of serum hAlAT reporter product by ELISA. Hepatocyte recipients with longterm survival (>60 days) were challenged by adoptive transfer of 2 million naive CDB' T cells (isolated from CD4 KO mice) which are sufficient to initiate rejection of functioning hepatocellular allografts in SCID recipients. Results: Combined treatment with anti-LFA-1 mAb and anti-CD40L mAb significantly prolonged hepatocyte allograft survival in CD4 KO mice such that 95% achieved hepatocyte survival >60 days posttransplant (MST>105 days, N= 17) in comparison to either agent alone (anti-CD40L mAb MST=35 days, N=4, P60 days (by short-term treatment with anti-LFA-1 and anti-CD40L mAb) received adoptive transfer of 2 million nayve purified CD8+ T cells. Continued survival of longterm hepatocellular allografts despite adoptive transfer of CD8+ T cells was observed in 60% (3 of 5) recipients. In 2 recipients, adoptive transfer of CD8+ T cells was associated with loss of hepatocellular allograft function by 31 and 56 days following cell transfer. In contrast, control SCID mice with functioning hepatocellular allografts which were reconstituted with 2 million naive CD8+ T cells rapidly rejected hepatocytes with MST of 17 days (N=5). Conclusion: Targeting of both CD40l CD4OL and LFA-1 not only prevents immunoresistant CDB-dependent immune damage of liver parenchymal cells, but also x lo3 vs. 1.321.3 x lo3 and 3.920.4 x lo3 vs. 0.321.0 x 103, respectively) while at 72 hrs it was 2.4t0.8 x lo3 vs. 2.122.0 x 103 (NS) . In the spleen, significantly higher heparanase treated cells were located within the tissue, showing proliferative activity at 48 and 72 hrs post transplantation. Already by 24 hrs after transplantation the proliferating index of SEC increased from 150.5 in controls to 2256 in heparanase treated rats (I' 80% compared with wild type EBs. The disparity in growth rate between wt and (-/-) ES cells was greater in later stages than in the initial 2 days of differentiation. This was accompanied by prolongation of the epiblast-like stage, associated with markedly delayed expression of endodermal mesodermal and ectodermal markers. Next, we examined the capacity of KLFG (-/-) EBs to differentiate into hepatocyte like cells by withdrawing leukemia inhibitory factor after two days, then culturing EBs under serum free conditions for 8 days, followed by growth on gelatinized dishes. Basic fibroblast growth factor and dexamethasone were added into the medium starting day 6 and day 10. Whereas expression in KLFG (+/+) EBs of alpha-fetoprotein and albumin mRNA occurred after 10 days of differentiation, their expression was almost absent and significantly delayed in KLF6 (-/-) EBs. An impaired ability of KLF6 (-/-) cells to differentiate along hematopoietic and neuronal lineages was also observed. In conclusion, these data indicate a broad and proximal role of KLFG in tissue specific development with specific role in hepatocellular development. Cell transplantation has been recognized as a possible future treatment for liver cirrhosis. One of the major questions is which cell will be of greatest useSpleen might be a major source of stem cell, if exist, since splenomegaly is a common in cirrhotic patients and even splenectomy is performed for treatment. Recent advance in the isolation technique of stem cells based on the efflux of fluorescent dyes Hoechst 33342 has turned out to be an efficient method to purify stem cells called Side population (SP) cells. We utilized this method to isolate stem cells from spleen. We also transfused these SP cells into rat liver treated with carbon tetrachloride (CC14) to see whether these SP cells proliferate and differentiate into both hepatocytes or cholangiocytes. Methods. C57/6J male mice and female NODlSCID mice were obtained from Nihon CREA. EGFP transgenic (tg) rat were obtained from Nihon SLC. Splenocytes were prepared by digesting in 0.5% collagenase solution and forcing tissue through sterile mesh. Splenocytes were then resuspended at lo6 cellslml in Hanks Balanced Salt Solution (HBSS) containing 2% FBS, 1mM Hepes (HBSS+), and ug/ml Hoechst 33342 w or w/o verapamil and were incubated at 37 degrees centigrade for 90min. Splenocytes were then stained for 30min. on ice with FITC or PE conjugated various antibodies against Sca-1, c-kit, Thyl.1, CD45 and CD34. Splenocytes were washed in HBSS+ alone 3times and then in HBSS+ with 2ug/ml propidium iodide. Splenocytes were analyzed and sorted on a dual-laser FACStar Plus flow cytometer. ( excitation:36nm, emission:424/BP44 & 660/BP20 filter & 640-nm short-pass dichroic mirror). We also prepared SPcells from donor EGFP tg rats' spleen as described above. These spleen EGFP positive SP cells were directly injected into liver of anesthetized recipient NODlSCID mice. The recipient NODlSCID mice were injected with 20 ug. CC14 intrapentoneally every once a week before and after transplantation. Immunohistochemical staining were performed to identify engrafted EGFP positive cells, liver specific proteins such as cytokeratinl8J9, AFP and albumin. Results. Initial Hst333422 staining of low density splenocytes showed the SP fraction to be readily detectable ( 0.3 -1%). These cells were verapamil sensitive and could be stained with Hst33342 after verapamil treatment. The frequency of each surface markers positive cells were as follows (Table 1 and 2). Two weeks after cell transplantation, donor derived EGFP positive cells were engrafted. The some engrafted cells distributed in a cluster indicating donor cell proliferation. Hepatic differentiation of engrafted cells were shown by double staining of EGFP and rat albumin or cytokeratinld. We observed some of AFP positive cells in EGFP positive cells, but very few of cytokeratinl9 positive cells. Conclusions. The adult rodent spleen had consistently detectable fraction of SP cells. The surface marker profiles were different from those of bone marrow SP cells. The splenic SP cells could be engrafted in liver and differentiate into hepatocyte like cells. Our results suggest that splenic SP cells could be exploited for the repair of damaged liver. BackgroundlAim : It has been reported that bone marrow cells (BMCs) can replace liver in a murine model of tyrosinaemia and correct this metabolic disease through the fusion of donor BMCs to recipient hepatocytes. The aim of this study was to test whether or not this replacement is a general phenomenon in other liver injury models. The following three models were tested; (1) hepatitis B transgenic mouse (TgN(AlblHBV)), (2) albumin-urokinase transgenic mouse (TgN(AlblP1au)) and (3) carbon tetrachloride (CC14) treatment model. As the selective liver injury models, irradiated TgN(Alb1HBV) and TgN(AlblP1au) were transplanted with lx106-lx107 BMCs from green fluorescent protein (GFP) transgenic mice (TgN(ActbEGFP)) or from @-galactosidase transgenic mice (TgN(MtnLacZ)). As a non-selective liver injury model, irradiated C57BL/6 mice were transplanted with 1x106 BMCs from TgN(ActbEGFP) or TgN(MtnLacZ) followed by the administration of CC14 (0.02mllkg animal weight, twice a week for 4 weeks). A CCI4 protocol without preparative irradiation was also tested, in which C57BL16 mice were first administered CCl, 8 times, then followed by transplantation of TgN(ActbEGFP) or TgN(MtnLacZ) BMCs. After the analysis of the donor cell engraftment in the peripheral blood, these recipient mice were sacrificed and checked for donor-derived hepatocytes at 2-47 weeks post-transplantation. Sixty liver sections for each animal (containing approximately 1 . 5~1 0~ hepatocytes), were analyzed for GFP positive cells and the whole livers were inspected for @-galactosidase expression with X-gal cytohistochemistry. However, there were no GFP positive hepatocytes and no gross blue staining of the livers with X-gal in any of the 18 recipient mice. GFP positive cells were only located in sinusoids or associated with larger vessels. They were readily distinguishable from hepatocytes through their morphology and were negative for albumin by immunostaining. In addition, the livers from 4 female animals with gender mismatched BM transplantation were also tested with Y chromosome FISH analysis, which might be a more sensitive method to detect donor derived cells than transgenic epitope tagging. Total of 5 isolated hepatocytes were positive for Y chromosome in 4 . 1~1 0~ hepatocytes analyzed, although it still remains to be elucidated whether these cells arise from spontaneous fusion events or from transdifferentiation. Conclusions: These results demonstrate that there is little or no contribution of BMCs to the replacement of injured livers in these models. Thus, we do not believe that BM derived cells can generally lead to a cure of liver damage. Background &Aims: Mouse embryonic stem (ES) cells are clonal cell lines derived from the inner cell mass of developing blastocysts and have multi-lineage differentiation ability. We previously reported that ES cells can be made to differentiate into hepatocytes possessing high metabolic activities by transfection of hepatocyte nuclear factor-3beta (HNF-3b). In the present study, we investigated the expression of hepatobiliary organic anion transporters and bilirubin uridine diphosphate glucuronosyltransferase (ugtlal) in undifferentiated and differentiating HNF-3btransfected ES (HNF-3 b-ES) cells. Materials & Methods: HNF3b-ES cells were established from a mouse ES cell line. Undifferentiated HNF-3b-ES cells were main-tained in gelatin-coated dishes without feeder cells in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 0.1 mM of Z-mercaptoethanol(2ME), 0.1 mM of non-essential amino acids (NEAA), 1 mM of sodium pyruvate and 1000 Ulml of leukemia inhibitory factor (LIF). Undifferentiated HNF-3b-ES cells were allowed to differentiate in DMEM supplemented with 10% FBS, 0.1 mM 2-ME, 0.1 mM NEAA, 1 mM sodium pyruvate, and 20 nglml of fibroblast growth factor 2 (FGFZ) in the absence of leukemia inhibitory factor (LIF). Differentiating HNF-3 b-ES cells were collected for RT-PCR analysis on day 14, and for Western blotting on days 14 and 28. Results: The expression of organic anion transporting polypeptide 1 (oatpl), multidrug resistance-associated protein 1 (mrpl), mrp2, mrp3, and ugtlal was not seen in the undifferentiated HNF-3 b-ES cells by RT-PCR, whereas all were expressed in differentiating HNF-3 b-ES cells. Protein expression for oatpl, mrpl, mrp2, mrp3, and ugtlal was also observed in the differentiating HNF-3 b-ES cells by Western blotting. An immunofluorescence examination revealed that oatpl was co-located with desmoplakin, a marker for the basolateral (sinusoidal) membrane, and mrp2 was co-localized with CD26, a marker for the apical (canalicular) membrane, though they were both expressed throughout most of the cell membranes. We have used cDNA microarray analysis of clonal murine hepatoblasts to identify genes that are significantly and preferentially expressed in undifferentiated hepatoblasts compared to adult mouse liver. Unique stem cell genes were identified by overexpression in undifferentiated hepatoblasts compared to adult liver. 61 genes were identified by these criteria and were designated candidate stem cell genes (CSCG). 54 CSCG were distributed on all chromosomes of the mouse except chromosomes 12,14 and the Y chromosome. However, chromosomes 2, 3 and 19 showed a preferential distribution of CSCG compared to the total number of genes mapped on these chromosomes. Approximately half of our CSCG are commonly expressed in other murine stem cell populations including embryonic stem cells, neural stem cells, mesenchymal stem cells and hematopoetic stem cells, suggesting that hepatoblasts share a molecular signature with other stem cell populations. Some of the commonly identified stem cell genes include: Sc14a2, c-myc, Nek2, phosphodiesterase 7, nicastrin, SMARCA 3 and SMARCA 5 and PIAS3. These genes function in a variety of pathways including ion transport, anion exchange, cell cycle control, chromatin organization, DNA binding activity and signal transduction. Approximately 25 % of CSCG are expressed in early endoderm, fetal liver and/or the pancreatic bud. A search for homology to known proteins indicated that some of these genes contain homology to membrane transporters, zinc finger proteins and spindle apparatus components. This class of genes is of particular interest because they may represent new hepatic lineage markers. The results of this analysis leads us to conclude that a network of pathways some of which are in common with other stem cell populations function together to maintain the liver stem cell phenotype. Valentina M Factor, Aranzazu Sanchez, Ju-Seog Lee, Tanya N Hoang, Snom' S Thorgeirsson, National Cancer Institute, Bethesda, M D Rat liver epithelial cells (RLE) were isolated from normal adult rat livers and established as a cell line . RLE cells were found to express undifferentiated characteristics resembling adult liver stem cells. We have addressed cellular and molecular mechanisms controlling lineage commitment of hepatic stem cells using RLE-13. To induce RLE differentiation along the hepatocytic lineage, differentiation protocols were developed consisting of a sequential treatment with the demethylating agent 5-aza-cytidine (5AC), fibroblast growth factors 1 and 2 (FGF), oncostatin M (OSM) and hepatocyte growth factor (HGF) in the continuous presence of the synthetic glucocorticoid dexamethasone (Dex). These treatments resulted in a remarkable enlargement in cell size and organelle complexity as shown by FACS and confocal microscopy. Morphological maturation of RLE was paralleled by a decrease in cell proliferation. More significantly, RT and real time PCR analysis revealed an induction of hepatocyte specific markers such as TAT, TTR, glucose-6-phosphatase, and connexin 32. In addition mRNAs encoding liver enriched transcription factors HNF lalp, HNF 3ru, HNF 4, and ClEBP p were notably induced along with hepatocytic differentiation. The PCR results were supported by a cDNA microarray analysis. Comparison of the gene expression profiles following the treatment protocols reflected an activation of the hepatocytic differentiation program as judged by increased expression of a differentiation-associated gene set (phosphofructokinase, glutathione-S-transferase) and decreased expression of a cell cycle regulated gene set. Currently, transplantation studies are performed to identify the potential of the differentiated RLE to repopulate and restore the diseased livers. The results indicate that cell lines derived from adult liver stem cells may provide a renewable resource for transplantation. Adult liver stem cells (ADULIS) can be isolated from rodent bone marrow. When cultured under specific conditions in co-culture with isogeneic hepatocytes, ADULIS fiom 14 days bile duct ligated (BDL) rats are transdifferentiating into a hepatocyte-like lineage and are able to produce urea from ammonia. The aim of the study presented is to describe the hepatocyte specific metabolic capacity of cultured ADULIS from normal or BDL rats in single or co-culture with isogeneic hepatocytes, with or without Interleukin-3 (IL-3). Methods: ADULIS were isolated by a two-step immunoisolation procedure (i.e. Beta-2-microglobulin negativity and Thy-1 positivity) from rat femoral bone marrow (Male Wistar rats, 250g) and cultured on a matrigel layer in 24-well polystyrene dishes at a cell density of 50'000 cells/cm* in small hepatocyte media. Isogeneic hepatocytes were isolated by a two-step collagenase perfusion technique and seeded (100,000 cells/cm2) on an inlay with a collagen-coated PTFE membrane (poresize: 0.4pm) for co-culture experiments. Of note, the culture media was supplemented with 5% isogeneic serum and dexamethason M) was administered after culture day 3. IL-3 was added in the corresponding experimental groups (10nglml). After removal of the hepatocyte-inlay (in the co-culture groups) ADULIS cultures were exposed to 1.5mmol NH4CI for 5 hours in DMEM and urea formation determined thereafter with a colorimetric assay. Cells were harvested in Trizol, total RNA extracted and after reverse transcription cDNA used for real-time PCR determination of 18S(rm~) content to standardize the metabolic signal for cell number. Relative ureagenesis values were compared statistically using Jandel Scientific. The significance level was set at pC0.05. Results: The amount of ADULIS isolated from normal rat femoral bone marrow (n=lO) was 1.4620.75 x lo6 and 1.8321.97 x lo6 in animals (n=6) after seven days of BDL (p=n.s.). Relative urea synthesis in cultures from normal animals was 1.0320.42 in single and 1.3820.41 in co-culture at culture days 3, 6, 9, and 12. With addition of IL-3 to the culture media urea genesis was determined to be 1.6820.6 in single and 2.6521.0'2 in co-culture. In cell cultures from seven days BDL rats relative urea formation was 1.5821.43 in single and 2.6351.32 in co-culture. With the addition of IL-3 to the culture media values were 2.3920.51 in single and 3.1620.81 in co-culture. Addition of IL-3 to the culture media increased the metabolic signal in all cell cultures from normal animals (ANOVA on Ranks, p<0.05) but not in cultures with cells isolated from BDL animals (p=n.s.). Co-culture induced stronger ureagenesis under all culture conditions examined (Paired t-test: p<0.05). Conclusions: To exclude immunological interference from adult immunocompetent bone marrow cells cultures were strictly isogeneic (i.e. ADULIS, hepatocytes, and serum from the same animal). Co-culturing ADULIS with isogeneic hepatocytes increased ureagenesis in all paired culture experiments. As there is no direct cell contact between hepatocytes and ADULIS paracrine soluble, so far undetermined, factors must be involved. Interestingly by the addition of IL-3 transdifferentiation was inducible in cultures of ADULIS from normal animals but no additional effect was detectable in the cell cultures from BDL animals. In further studies it will be necessairy to determine if cholestatic serum also induces accelerated and more pronounced hepatocyte specific metabolic capacity in ADULIS from normal animals. Factors responsible for this transdifferentiation should then be isolated from cholestatic serum and might be used to support the failing liver in vivo potentially by activation of the ADULIS pool in the bone marrow and the liver. Disclosures: Saji, Shinji Tamura, Yuichi Yoshida, Shinichi Kiso, Ayuko Iizuka, Hitoshi Matsumoto, Takako Kawasaki, Yoshihiro Kamada, Yuji Matsuzawa, Graduate School of Medicine, Osaka University, Suita, Osaka, Japan Background and Aims: Recently, the evidence that bone marrow cells (BM cells) have trans-differentiating potential into hepatic lineage cells has been described in animal transplantation experiments and human pathology of bone marrow transplantation recipients. However, the molecular mechanism underlying this phenomenon has remained unclear because of lack of effective culture system. To address this issue, we developed a novel in vitro culture system in which murine BM cells are cultured with growth factors and investigated factors required for the transdifferentiation of EM cells into hepatic linage cells. Methods and Results: (1) BM cells were isolated from the femora of 6-to 10-week-old male C57BL/6] mice. They were cultured in three-dimensional culture system using type I collagen gel and stimulated without or with the growth factors (EGF, HGF, HB-EGF, OSM, BMP-4, bFGF, FGF-4; 2OngIml) for 12 day. To investigate the effect of these growth factors, the gene expression of albumin was examined by quantitative RT-PCR. When BM cells were cultured with HGF, OSM, bFGF and FGF-4, albumin was induced effectively. Among these growth factors, bFGF was found to induce albumin the most effectively in the cultured BM cells. (2) BM cells were cultured in collagen gel with bFGF (2Onglml) for 6, 12, 18 days. The gene expression of hepatocyte-specific markers (albumin, cytokeratin 18, alphal-antitrypsin, glucose 6 phosphates and tyrosine aminotransferase), cholangiocyte-specific marker (cytokeratin 19) and liver-enriched transcription factors (HNFlalpha, HNF3alpha, HNF3beta, HNF4alpha, GATA4, GATA6, ClEBPalpha, and CIEBPbeta) were examined by RT-PCR. Upon the stimulation of bFGF, BM cells were found to express cytokeratin 18, alphal-anitripsin and glucose 6 phosphates, cytokeratin 19 and liver-enriched transcription factors including HNFlalpha, HNFJalpha, HNF3beta, GATA4 and GATA6. (3) BM cells were cultured in collagen gel with bFGF (20ngIml) for 6 days. We assessed expressions of albumin and CK18 proteins of cultured BM cells by immunohistochemistry. About 5% cells of them were positively stained for both albumin and cytokeratin 18. Conclusions: We established a novel in vitro culture system of BM cells and demonstrated that basic fibroblast growth factor could induce the trans-differentiation of BM cells into hepatic lineage cells the most effectively. Furthermore, this conversion was associated with induction of liver-enriched transcription factors including hepatocyte nuclear factors and GATA family proteins. These results indicate these liver-enriched transcription factors are the major components, which induce this trans-differentiation. Disclosures: Ayuko Iizuka -No relationships to disclose Yoshihiro Kamada -No relationships to disclose Takako Kawasaki -No relationships to disclose Shinichi Kiso -No relationships to disclose Hitoshi Matsumoto -No relationships to disclose Yuji Matsuzawa -No relationships to disclose Yukiko Saji -No relationships to disclose Shinji Tamura -No relationships to disclose Yuichi Yoshida -No relationships to disclose Crtteil, France; Nadine Martin, Hbpital Henri Mondor, Criteil, France; Dominique Couchie, Anne M Preaux, Yannick Laperche, Elie 5; Zafrani, INSERM U581, Cre'teiZ, France Liver can regenerate from oval precursor cells when the replication of hepatocytes is inhibited. These cells proliferate in the periportal region, migrate in the lobule and differentiate into hepatocytes. Stromal-derived factor-1 (SDF-1) is a chemokine which plays a major role during embryogenesis. Since SDF-1 and its sole receptor CXCR4 are involved in the differentiation of progenitor cells (e.g. B lymphopoiesis, digestive epithelial cell renewal), the aim of our work was to study the expression of SDF-1 and CXCR4 in a model of hepatic regeneration from precursor cells. Hepatic regeneration from oval cells was induced by treating rats with acetylaminofluorene followed by partial hepatectomy. Rats were sacrificed the day of surgery (day 0) and at days 1,5,9,14 and 21 after partial hepatectomy. Rare oval cells were observed at day 1. Their number was moderate at day 5, marked at day 9 and 14 and declined thereafter. Oval cells predominated in the periportal region and usually formed canalar structures resembling cholangioles. Individual or cholangiole-forming oval cells strongly expressed SDF-I, as demonstrated at the mRNA level by in situ hybridization (ISH) and at the protein level by irnmunohistochemistry. Interlobular bile duct epithelial cells were also positive for SDF-1 protein but negative for its mRNA. CXCR4 mRNA hepatic levels, as assessed by quantitative RT-PCR, paralleled the degree of oval cell proliferation. ISH showed marked CXCR4 mRNA expression in individual as well as in cholangiole-forming oval cells. In order to determine which role the SDF-1ICXCR4 couple could play in hepatic regeneration from oval cells, rats were treated or not with fucoidan (25 mglKg ip, twice daily), a sulfated polysaccharide known to bind to SDF-1 and to block its biological effects. As compared to untreated animals, oval cell proliferation was markedly decreased in five of the seven fucoidan-treated rats. In conclusion, oval cells express SDF-1 and its receptor CXCR4 during hepatic regeneration from precursor cells. Our results strongly suggest that the SDF-1lCXCR4 couple acts in an autocrinelparacrine way to stimulate the proliferation of these precursor cells. Disclosures: Dominique Couchie -No relationships to disclose Background: Bone marrow cells and embryonic stem cells are being used for cell transplantation. Although these cells are known to be multipotent and capable to become hepatocyte, several problems have been pointed out, such as cell fusion and teratoma generation. Thus we paid attention to side population (SP) cells, those with a verapamil-sensitive ability to efflux Hoechst 33342. SP cells have been identified in several tissues. The bone marrow SP phenotype appears to be a common feature of stem cells, but hepatic and splenic SP cells have been less well characterized. Aim: We tried to separate and culture SP cells from non-parenchymal liver cells (NPCs) and splenocytes, and examined whether they would obtain hepatic phenotype. Methods: 8-12-week-old female transgenic rats that ubiquitously express EGFP were used to isolate non-parencymal liver cells by the standard collagenase two-step perfusion method, and splenocytes were isolated by forcing the tissue through sterile mesh, then mononuclear cells were separated by Ficoll density gradient centrifugation. They were resuspended at lo6 cellslml in Hanks Balanced Salt Solution (HBSS) containing 2% fetal calf serum, 1mM Hepes (HBSS+), and 5uglml Hoechst 33342 (Hst) w or wlo Verapamil and were incubated at 37 degrees centigrade for 90min. Cells were washed twice in HBSS+ and then resuspended in HBSS+ with 2uglml propidium iodide. Cells were analyzed and sorted on a dual-laser FACStar Plus flow cytometer. RT-PCR was performed with RNA extracted from freshly isolated SP cells. Isolated SP cells were cultured with male rat primary cultured hepatocytes in collagen gel sandwich. Cell morphology was monitored under a phase-contrast microscope and a fluorescence microscope. The expression of albumin, cytokeratin 18, and cytokeratin 19 was analyzed by immunohistochemistry. Fluorescence in situ hybridization (FISH) for the sry gene was performed to exclude ceIL fusion. Results: The rate of NPC-SP was 0.05% of isolated NPCs. 89% of NPC-SP expressed CD45, while 55% of NPC main population (MP) did. Freshly isolated SP cells were smaller than mature hepatocytes. They looked like monocytes or lymphocytes. These cells did not express liver specific markers, such as albumin, alpha-fetoprotein, cytokeratin 18, or cytokeratin 19. During one-week culture with hepatocytes using collagen gel sandwich method, we could find no change in GFP-positive SP cells, but after further culture, we found several colonies consisting of GFP-positive SP cells, whose shape became polygonal. After 2-week-culture, most of GFP-positive cells were stained positive for albumin. NPC-SP cells died within one week without hepatocyte co-culture. NPC-MP cells formed colonies after 2 weeks culture with hepatocytes, but unlike SP cells, they looked like fibroblasts and were negative for albumin. Freshly isolated SP cells from splenocytes were also small and monocyte-like, and expressed no liver specific markers. Spleen SP cells could culture for more than 2 weeks in the same way as NPC-SP cells, and looked like hepatocytes, but did not proliferate. After having been cultured, most of GFP-positive cells were stained positive for albumin and cytekeartin 18, and negative for cytokeartin 19; on the other hand splenic main population cells after 2 weeks' culture expressed those very little. Conclusion: We succeeded in culturing liver and spleen SP cells. SP cells which did not express mRNA of liver specific markers, such as albumin or cytokeratin 18, became to express those after the culture with hepatocytes. These results suggest that these SP cells have plasticity and obtain hepatic function. Thus, tissue stem cells, especially SP cells, are useful for cell transplantation. Spleen SP cell might be a candidate for source of cell transplantation for treatment of liver cirrhosis. Miura, Takashi Goto, Ken-ichiro Mikami, Kunio Nakane, Kazuo Yoneyama, Hirokazu Nagai, Kunihiko Terada, Toshihiro Sugiyama, Katsuyuki Imai, Haruki Senoo, Sumio Watanabe, Akita University, Akita, Japan Background Epimorphin, a mesenchymal cell surface-associated molecule identified as an epithelial morphogen, is detected on stellate cells in the liver. We have reported that a contact with hepatic stellate cells (HSCs) promotes differentiation of hepatic stem-like cells (HSLCs). Here, we show the involvement of epimorphin in that process.Materials and Methods: HSLCs and HSCs were isolated from healthy adult rats using two-step collagenase perfusion and Percoll gradient centrifugation, and maintained standard medium, Dulbecco's modified Eagle's medium with 10% fetal bovine serum. HSLCs were cultured in stellate cell-conditioned medium, co-cultured with HSCs to maintain cellcell contact or in the presence of epimorphin. Phenotypical and morphologic changes were investigated by RT-PCR and with an electron microscopy, respectively. Results: HSLCs are polygonal in shape and assume a cobblestone appearance when cultured in standard medium. Transmission electron microscopy showed that HSLCs , 20 to 25 micro-m in diameter, have a round to ovalshaped nucleus, a few vacuoles, scant organelles and a high nucleuslcytoplasm ratio compared with normal hepatocytes. The phenotypic properties of HSLCs did not change as well as their size and shape during this experiment. HSLCs characterized by RT-PCR are follows: positive, c-kit, musashi-1 (a neural stem cell marker), alpha-fetoprotein, cytokeratinl9, connexin43; negative, albumin, transferrin, tyrosine aminotransferase, gamma-glutamyl transpeptidase. HSLCs cultured in stellate cell-conditioned medium had no phenotypical and morphological changes. HSLCs co-cultured with HSCs expressed albumin, transferrin, and tyrosine aminotransferase, which were inhibited by an anti-epimorphin antibody. Furthermore, epimorphin induced the markers not only for hepatocytes including albumin, transferrin and tyrosine aminotransferase, but also for cholangiocytes including gammaglutamyl transpeptidase in addition to increased expression of connexin43 and cytokeratinl9, with decreased expression of c-kit and musashi-1. In addition, ClEBP beta was enhanced, which has been reported to mediate through morphogenesis by epimorphin. HSLCs co-cultured with HSCs piled up and subsequent development of bile-canalicui-like structures, which was dramatically inhibited by an anti-epimoirphin antibody. HSLCs, close to epimorphin, stated piling up, changed their shape from flat to cuboidal, became rich in mitochondria and rough endoplasmic reticulum and formed bile-canalicui-like structures. Conclusions: HSLCs have a self-renewing capacity and multilineage differentiation potential. Epimorphin is involved in differentiation of HSLCs though a contact with HSCs. Disclosures: Takashi Goto -No relationships to disclose n order to identify new and differentially expressed genes in fetal rat liver that are specific for epithelial stemlprogenitor cells and genes involved in liver progenitor cell differentiation, we used murine cDNA microarrays containing 8,976 cDNAs, available at the Functional Genomic Facility, AECOM. The expression pattern of fetal liver stemlprogenitor cells was studied from embryonic day 13 through birth, 7 days after birth and in adult liver. The driver RNAs were isolated from cells adhered to the dish after plating the cell suspension in order to remove the blood cells. Reference RNA was isolated from the livers of newborn rats. We found that 511 genes present on the cDNA microarrays were developmentally regulated. These genes fall in two major hierarchical clusters, according to their pattern of expression. The 281 genes that are down-regulated during fetal liver development were distributed in functional groups and further analyzed. In this study, special attention was paid to genes that were induced in fetal liver but were not expressed (or expressed at a very low level) in adult liver. These genes are of special interest because they can serve as specific markers for identification and for isolation of liver stemlprogenitor cells. In addition, these genes represent links to understanding the fetal liver specific molecular pathways that govern cell proliferation, survival, apoptosis and differentiation. To determine which of the 281 over-expressed genes in 13-14 day fetal liver that are down-regulated in adult liver are progenitor cells specific, we searched in the available databases whether the expression of these genes in adult liver was previously reported. Seventy genes were further analyzed: the clones of interest were hybridized to radioactive labeled 32P cDNA synthesized from fetal and adult liver RNAs. For 48 of the clones, we found that there was little or no expression in adult liver. The expression level of 25 selected clones was analyzed further by quantitative PCR, and they were confirmed as highly induced in fetal hepatoblasts compared to adult liver. Half of the 48 clones are ESTs. The known genes fall in different categories, the major four being: genes related to transcription; signal transduction; morphogenesis, histogenesis and organogenesis; cell adhesion, de-adhesion and migration. Some of the known genes over-expressed in fetal liver that are not expressed or expressed at very low level in adult liver are: GrblO (AA260248), Fh12 (AA023645), Tnc (AA270625), Peg3 (AA003064), Hey1 (AA049474), Enah ((AA217593), Pkcd (AA276844), Lox (W96914), Shcbpl (AA265225), Magoh (AA254528), Manba (AA200473), Klf5 (AA432818), Gpc3 (AA274932), Pcolce2 (AA153907), Ppap2c (AA220316), Nfkb2 (AA060802), Adam19 (AA051790), Akapl2 (AA387076), Tagln (BC003795. It should be noted, that most of the 48 clones and all those listed here that we have identified as liver progenitor cells specific, are expressed in stem cells of embryonic, hematopoietic, or mesenchymal origin. Two of the presented genes encode cell surface proteins: a disintegrin and metalloproteinase domain 19 (Adaml9) (meltrin beta), glypican 3. Using in situ hybridization, we are currently verifying whether our putative liver progenitor cell specific genes are expressed in hepatoblasts and in rare progenitor cells that remain in the adult liver. Identifying and cloning new genes that are expressed uniquely in liver stemlprogenitor cells will allow us to design a method for isolation of these cells and to study their role in liver development, growth control and regeneration. Hepatocyte transplantation has been shown to be an effective treatment for liver diseases including fulminant hepatic failure and metabolic defects in liver function. However, the availability of sufficient numbers of hepatocytes with which to conduct clinical studies limits the wide-spread availability of this therapy. Previously, we have shown that human placenta derived stem cells (PDSC) differentiate along neural or hepatic lineages depending on the culture conditions and suggested that these stem cells could be a source of cells for clinical transplantation. Here we report on a protocol to further optimize hepatic differentiation of PDSC. A three stage culture system was applied to induce hepatic differentiation. AE cells were propagated in DMEM based standard media which contains EGF 10 nglml for a week (Stage I) and subsequently with growth factors such as FGF-1, 2, 4, 7, 8, Oncostatin M, HGF, andlor dexamethasone (DEX) and insulin-transferrin-selenium (ITS) for 2 weeks (stage 11). The cells were then changed to media supplemented with different nuclear hormone receptor agonists for 1 week to stimulate hepatic maturation (stage 111). Total RNA samples were examined for hepatocyte specific gene expressions with real-time quantitative PCR (RTQ-PCR). Additional studies examined hepatic differentiation on different culture substrates including collagen, gelatin, fibronectin, ARG-GLY-ASP-SER, and Matrigel 1%, 20%, 100% were also examined. In the three stage differentiation system, 20% (vlv) Matrigel coated plates and DEXlITS containing media, and phenobarbital (PB) were identified as the best conditions to upregulate liver specific gene expression. The principal human drug-metabolizing enzymes (Cytochrome P-450, CYP) were also examined by RTQ-PCR. CYPlA1, 2C8,2D6, and 3A4 were induced with either rifampicin or PB. In parallel, expression of the relevant liver-enriched transcription factors, HNF-4, ClEBP-a and CIEBP-b mRNAs were increased under conditions which induced hepatic differentiation. Under slightly different conditions, differentiation of PDSC into cells which expressed Pax6, PDXl and Nkx2.2, insulin and glucagon was observed. Flow cytometric analysis showed isolated nayve AE cells contains a population of cells which express the embryonic stem (ES) cell markers, SSEA-3, 4, TRA 1-60, and TRA 1-81. Furthermore, the placental stem cells formed embryonic body (EB)-like structure when cells were cultured on Matrigel. The EB-like structures retained the expression of SSEA-3, 4, TRA 1-60, and TRA 1-81. These data indicate that cells can be isolated from term placenta which express markers of ES cells. Under appropriate conditions PDSC expand, and differentiate into cells with characteristics of hepatocytes, neurons, or pancreatic cells. We propose that PDSC could provide a new source of cells for clinical transplantation and regenerative medicine. Unlike with ES cells, there should be no social, ethical or religious objections to the isolation and use of placental-derived stem cells. Using the SSH technology, we have identified previously 643 induced clones in 13 day fetal compared to adult liver; 202 of these clones were independent transcripts, 64 of them represented ESTs and4 appeared to be new sequences (Petkov et al., Genomics 6 8 197-209,2000) . Applying the rapid amplification of 5' and 3' cDNA ends to subtracted genes (GeneRace Kit, Invitrogen), we have obtained the full-length cDNAs for two of the subtracted clones. One of them (clone 2G8) encodes a novel putative serine proteasel subtilase highly expressed in fetal liver and comparatively low in adult. The expression of 2C8 mRNA was analyzed by Northern blots with RNA isolated from different rat tissues: fetal and adult liver, isolated 14 day fetal hepatoblasts, brain, bone marrow, kidney, spleen, intestine, colon, stomach and muscle. On Northern blot with total RNA, a strong signal was detected with fetal liver/ hepatoblasts RNA and a weak signal with adult liver RNA, which showed that the expression of this mRNA is developmentally regulated. To confirm the liver specific expression of 2G8, quantitative RT-PCR was camed out. The results of this analysis confirmed our previous results and revealed also a very low expression of 2G8 mRNA in the colon. The 2G8 cDNA is 3345 nucleotides in length and codes for a novel protein of 691 amino acids. We were able to identify in the data bases the sequences of a rat contig which contains the complete gene for 2G8. The gene coding for 2G8 comprises 21,920 bp of the rat genome and includes 12 exons and a long 3'-UTR segment of 1,000 nucleotides. The initiation of the translation begins with an AUG codon, 188 nucleotides downstream from the 5'-end of mRNA. Using specific primers designed from the contig sequences, we obtained clones corresponding to the upstream promoter region and downstream of the gene in the PCR4 TOP0 vector (Invitrogen). These two upstream and downstream clones were ligated to the 2C8 cDNA upstream and downstream sequences, respectively, so that the whole gene with the 5' and 3' regulatory sequences was obtained in the PCR4-TOP0 vector. Analysis of the structure of 2G8 showed that this protein resembles 8 others mammalian subtilases, named convertases described during the last decade: furin, PC1/3, PC2, PC4, PACE4, PC516 and PC7ILPCIPC8 and SKI-1ISIP. These proteases are synthesized as proproteins and secreted from the cell. 2G8 contains a putative signal sequence for release from the ER located between amino acid 30 and 31 (alanine 4 glutamine). However, it does not share the consensus cleavage sequences characteristic for the other convertases: (K,R)-(X)n-(K,R) for processing, release and secretion of the active enzyme form. At present, we do not know the processing site of the proprotein. In situ hybridization experiments showed that this serine protease is expressed in fetal liver hepatoblasts. Convertases are secretory proproteins implicated in tissue specific processing of hormones, growth factors, metalloproteinase, extracellular matrix proteins, viral proteins etc. They show tissue specificity and some are implicated in development and differentiation. 2G8 is a novel convertase, it is developmentally regulated and most likely functions in processing and activation of growth factorslcytokines or extracellular matrix proteins implicated in morphogenesis. Further studies of its promoter region will reveal the control sequences and transcriptional activators and repressors regulating its expression during development. College, London, UK INTRODUCTIONIAIMS: Bone marrow cells (BMCs) can contribute to regeneration of the chronically damaged liver but in human studies and animal models the magnitude of this axis is highly variable. In a murine model of hepatitis B we examined whether this pathway of regeneration is enhanced by inhibiting endogenous hepatocyte regeneration. METHODS: 2 month old female mice transgenic for Hepatitis B surface antigen (HBs-tg) received lethal irradiation and were then transplanted with C57B1/6J male BMCs by tail vein injection. 6 weeks later half the mice were treated with retrorsine, a pyrrolizidine alkaloid, to irreversibly block proliferation of endogenous hepatocytes. Mice were sacrificed at 3 and 6 months following retrorsine injections. Y chromosome containing hepatocytes were identified using in situ hybridisation and phenotype markers (positivity for cytokeratins 8/18, cytochrome P450 and glycogen, negative for CD45). RESULTS: In the control mice with chronic liver damage there was an increase in Y chromosome positive hepatocytes over time, but the proportion remained <5% of the total number of hepatocytes. However, 19.3% (corrected count) of Y chromosome containing hepatocytes could be found repopulating the livers of the mice that had received retrorsine. CONCLUSIONS: BMCs contribute to the regeneration of the chronically damaged liver, but under conditions where regeneration of endogenous hepatocytes is possible this is minor. However, when chronic liver damage occurs and regeneration of the endogenous hepatocytes is inhibited, the contribution to liver regeneration from the bone marrow is significantly enhanced. Beaujon, Paris, France; Thierry Poynard, Groupe Hospitalier Pitie-Salpetriere, Paris, France Background Portal hypertension depends in part on the development of hepatic fibrosis. Thus, since FibroTest (FT) is a potential biochemical marker of fibrosis, this test was prospectively used to evaluate the presence and severity of portal hypertension in patients with different liver diseases. Methods: 89 consecutive patients (56 males; 32% chronic viral hepatitis C, 9% hepatitis 8, 33% alcoholic liver disease, 7% transplanted, 19% miscellaneous) with transjugular liver biopsy had same day measurements done for hemodynamic parameters (gradient, free hepatic, wedged pressures), histological parameters (Fibrosis scoring system in 4 stages, necroinflammatory activity grades and modification of architecture in 3 classes), and biochemical parameters via blood sample (FT for fibrosis and ActiTest for activity). The measurements of histological and biochemical parameters were done blind to any other characteristics. Sensitivity (Se), specificity (Sp), predictive values (NPV and PPV), Spearman correlation (SC), accuracy (AC), kappa (K) and the area under the ROC curves (AUROC) were assessed. The main endpoint was the prediction of elevated portal pressure (EPP), defined by a gradient of 2 5 mm Hg. The secondary endpoint was the prediction of highly elevated portal pressure (HEPP) of 212 mm Hg (HEPP). Results: The mean FT value was 0.47 (se 0.05) for 15 patients without EPP, 0.73 (0.04) in 23 patients with moderate EPP and 0.87 (0.03) in 49 patients with HEPP (p3 fold better thanNS3 alone, and skewed T cells towards Thl polarization resulting in elevated levels of IFN-.)I (higher than with NS3 alone) but not IL-4 or IL-10. T cells also acquired an activated phenotype (CD69+). To test in viva efficacy of NS3-DC peptide fusion protein, NOD-SCID mice were xenotransplanted with HCV-nake T cells, and vaccinated with three separate weekly injections of autolo-gous DCs (106) pulsed with nothing, NS3 alone, or the NS3-DC pep-tide#3 fusion construct. Vaccination using DCs pulsed with NS3-DC peptide# fusion protein significantly increased NS3-specific T cell proliferationlactivation and enhanced the expression of IFN-y, and TNF-a compared to NS3 alone but no detectable levels of IL-10 or IL-4 were observed. These data indicate that targeting HCV subunits to DCs using specific DC-peptides represents a novel vaccine approach that may facilitate targeting of immunogenic antigens to DCs to elicit specific T cell immune responses against NS3 of HCV. This immunotherapeutic strategy can be implemented against the broad range of immunogenic antigens of various pathogens. It has recently been reported that the administration of Eta in the setting of chronic HCV infection enhances IFN-induced viral clearance (1). The mechanism of these synergistic effects remains to be understood. Our hypothesis was that Eta enhances antiviral effects of IFN by increasing T cell reactivity to antigens. Methods: Peripheral blood mononuclear cells (PBMCs) were isolated in CPT tubes from blood of two healthy volunteers. Cells were washed and then cultured in complete media 24-well plates in the absence of any antigen (negative control) and in the presence of immobilized anti-CD3 antibody (I-CD3, 150 nglml, positive control) as a nonspecific T cell stimulant. I-CD3 stimulated cells were treated immediately with Peg IFN alpha-2a at two different concentrations consistent with physiologic doses in humans (10 nglml or 100 nglml) with or without the addition of Eta at two different concentrations that were also consistent with physiologic doses in humans (1.65 pglml or 3.30 pglml). Supernatants from each of the previous conditions were collected and assessed by ELISA for IFN-7 secretion as a marker of T cell activation. Results: Our findings are displayed in the figure below. There was no spontaneous IFN-7 detected in cells that were not stimulated. IFN-y was detected at a modest level after exposure to I-CD3 alone, a response that became greater after exposure of cells to Peg IFN alone or Eta alone. Production of IFN-7 was substantially enhanced in cells that were exposed to a combination of Peg IFN and Eta compared to those exposed to Peg IFN alone or Eta alone. Conclusion: Our findings suggest that Eta has a synergistic effect to that of Peg IFN in promoting in vitro activation of PBMCs and IFN-7 production in healthy individuals. Increased T cell activation with Eta may also suggest that TNFA suppress T cell function and may contribute to refractoriness to INF therapy in HCV patients. The influence of absolute dose and dose in mg per kg body weight of ribavirin on sustained virologic response (SVR) in interferonbased combination treatment of chronic hepatitis C as well as the influence of dose reduction is still controversial. Here, we address this problem by reanalyzing data of 343 previously untreated patients with chronic hepatitis C from a multicenter trial who were treated with interferon alfa-2a plus ribavirin and amantadine or interferon alfa-2a plus ribavirin (Berg et al, Hepatology 2003 ,37,1359 -1367 and completed therapy. Thereby, we used a multivariate approach to account for correlations of the dose of ribavirin with body weight and body mass index (BMI) and known predictor variables from this data set which are low baseline HCV RNA, high platelet counts, high pretreatment ALT, and low y glutamyl transpeptidase (GGT) as well as HCV genotype non-1. Because per protocol dosage of ribavirin was weight-adjusted (1000 mg for body weight below 75 kg and 1200 mg for body weight 75 kg or more) and body weight was positively correlated with GGT and negatively correlated with platelet counts (Spearman rank correlation, plOO kcopieslml before therapy) and serotype 1 (Genotype l a or lb). Forty-two patients were treated with IFN alpha-2b (6 MU) in combination with ribavirin (600-800 mg) daily for 2 weeks, following 22 weeks with IFN alpha-2b (6 MU) 3 times a week (the combination group). Another 40 patients who were treated with IFN alpha-2b alone between 1998 and 1999 (before introduction of ribavirin in Japan) with the same schedule are used as control (the monotherapy group). Serum samples were taken before, 2 weeks after administration and 12 weeks after cessation of therapy. Serum IL-18 are examined by enzyme linked immuno-sorbent assay (ELISA), using human IL-18 ELISA kit (MBL, Nagoya, Japan). The IL-18 ratio is defined as serum IL-18 level before administration divided by serum IL-18 level 2 weeks after administration. HCV-RNAs are quantitatively examined before and 2 weeks after administration. Sustained viral responses are confirmed 12 weeks after cessation of administration by RT-PCR. Results: There are no differences in the background of patients between the combination group and the monotherapy group. In the combination group, the decline of HCV-RNA level highly correlates with the IL-18 ratio in patients with higher viral titer (HCV-RNA>500 kcopieslml before therapy) despite no correlation is observed in patients with lower viral titer (HCV-RNA50 KJlml, PCR methodology). After 48 weeks of therapy, these patients are expected to achieve a lower rate of sustained viral response (SVR) than patients with undetectable HCV RNA at week 4 (VR4). Models based on viral dynamics and data from pilot studies with conventional interferon (IFN) suggest that a treatment period longer than 48 weeks may lead to a higher SVR rate in these difficult-to-treat patients. This trial compares 72 versus 48 weeks of treatment with PEGASYS(') plus COPEGUS('I in treatmentna'ive patients without VR4. Methods: This phase 111, randomized, parallel group, multicenter study is being conducted in Spain in accordance with the principles of Good Clinical Practice.(l) After signing the informed consent form, all patients received the study treatment (peginterferon alfa-2a [40KD] [PEGASYS(']] 180 pg once weekly and ribavirin [COPEGUS(')] 800 mg daily 1400 mg bid]) for 4 weeks. At this time, viral response (VR) was assessed using the COBAS AMPLICOR HCV(') Test v2.0 (detection limit HCV RNA 50 IUlml). Patients achieving VR4 continued therapy for an additional 20-or 44-week period, according to their HCV genotype and viral load. Patients without VR4 were randomized 1:l to continue the study treatment for either an additional 44 weeks (total active treatment duration 48 weeks or an additional 68 weeks (total active treatment duration 72 weeks ). A 24-week follow-up period is ongoing to determine the SVR rate. Results: The mean age of the patients was 42 years, the mean weight was 73.5 kg, and 37% were female. The prevalence of HCV genotype 1 or 4 was 78%. VR4 occurred in 36% out of 511 patients, (23% of those with HCV genotype 1 or 4,83% of those with other HCV genotypes). HCV genotype distribution, viral load, age, and body weight were similar in the randomized study groups (Group-48 and Group-72). Among patients without VR4, HCV RNA was undetectable in 62"h in Group-48 and in 69% in Group-72 at week 24. The biochemical response rate (BR) at week 24, defined as normal ALT levels, was similar as well: 81% in Group-48 and 78% in Group-72. The dropout rate after 24 weeks was 9% in Group-48 and 6% in Group-72. At the end of therapy (EOT), HCV RNA was undetectable in 93% of patients with VR4. The EOT VR rate in Group-48 was 74% versus 81Y0 in Group-72. The rate of BR in patients with VR at EOT was also higher in Group-72 (85%) than in Group-48 (77%). Dropout rates and reasons for dropping out by the EOT visit were different between groups: 17% in Group-48 and 36% in Group-72. Therapy was well-tolerated and no unexpected adverse events were observed. The prolongation of therapy from 48 to 72 weeks did not cause an increase in the frequency of neutropenia or thrombocytopenia. Conclusions: PEGASYS(') 3 80 p g weekly plus COPEGUS(') 400 mg bid administered for 72 weeks offers higher rates of VR and BR at EOT than 48 weeks of treatment in patients with detectable HCV RNA four weeks after treatment initiation. The 800 mglday dose of ribavirin used in the trial has been proven to be optimal in patients infected with HCV genotype 2 or 3, but not in those with genotype 1 who require 100011200 mglday. This evidence was not established when the trial was initiated. "The TeraViC-4 Study Group also included V. Ripollks Background ISIS 14803 is an antisense phosphorothioate oligodeoxynucleotide inhibitor of hepatitis C virus (HCV). In a previous Phase I study, plasma HCV RNA reductions 2 1.0 log were observed in 3 chronic HCV patients treated with ISIS 14803 for 4 weeks. Aims: The goals for this Phase I1 study were to evaluate the safety, antiviral efficacy, and pharmacokinetics of ISIS 14803 given for 12 weeks. Methods: The study was conducted in noncirrhotic chronic HCV patients. Following 2 weeks of thrice weekly intravenous dosing at 2.5 mglkg ideal body weight (IBW), patients were treated for an additional 10 weeks at higher doses given either once (Group A) or twice weekly (Group B). The two doses studied were 4 (initial 3 patientslGroup) and 6 mglkg IBW. Results: Of 43 patients enrolled into the study, 41 had genotype 1 HCV, 39 had plasma HCV RNA levels 2 2 x lo6 copieslml, and all but two had previously failed interferon-based therapies. Two of the 16 Group A and 5 of the 18 Group B patients treated at 6 mglkg IBW had plasma HCV RNA reductions 2 1.0 log. Three in Group B had reductions of 3.2-3.8 log. HCV levels remained 2.5 log reduced at 110 days after the last ISIS 14803 dose in one of these patients. Seventeen patients had transient elevations in ALT including all those with HCV RNA reductions. Peak ALT levels ranged 5-30 times the upper limit of the normal range (x ULN). For those with HCV RNA reductions, the ALT flares were temporally associated with the reductions. The ALT flares were not associated with any clinical signs, symptoms, or sequelae. There were no changes in prothrombin time or albumin level. Minor alkaline phosphatase (5 2.2 x ULN) and bilirubin (5 1.5 x ULN) elevations occurred in 2 patients. Dosing with ISIS 14803 was continued during 7 ALT flares, all of which resolved to baseline levels despite continued dosing. Three flares, ranging from 9-29 x ULN, occurred 3-8 weeks after the last dose of ISIS 14803, when presumably little or no drug remained in the liver. Similar ALT elevations have not been observed in the clinical studies of other phosphorothioate oligodeoxynucleotides. Aside from the ALT flares, common adverse events in this study were mild to moderate headache, fever, chills, fatigue, nausea, myalgia and other flu-like symptoms. The fever and chills usually occurred several hours after the end of the 6 mglkg infusions and resolved within 24 hours. Two serious adverse events that were considered possibly related to ISIS 14803 occurred in the study: one patient had cryoglobulinemic glomerulonephritis and one was hospitalized for precautionary monitoring of an ALTlASTlbilirubin flare. The former event was possibly due to an exacerbation of a pre-existing condition. The ISIS 14803 plasma pharmacokinetics observed in this study were consistent with those seen in the previous Phase I study. Conclusions: ISIS 14803 appears to have significant antiviral activity in some chronic HCV patients. The plasma HCV RNA reductions (up to 3.8 log) achieved in this study were in difficult-to-treat patients (i.e., genotype 1, high virus level, andlor nonresponders to interferon-based therapy). Although, ALT flares were observed in some patients, the results suggest the flares may not be due to a direct hepatotoxic effect of the drug. Other than the ALT flares, ISIS 14803 treatment was generally well tolerated. This study established the dose of 6 mglkg IBW twice weekly for further clinical evaluation. A study of ISIS 14803 in combination with peginterferon and ribavirin for patients not achieving an early virologic response during standard peginterferon and ribavirin therapy is in progress. with the long serum half-life of albumin. The objectives of this phase 1/2, open-label, dose escalation study were to evaluate the pharmacokinetics, safety, tolerability, immunogenicity, and pharmacodynamics of Albuferon in subjects with hepatitis C virus (HCV) who had previously failed IFNa containing regimens. Methods: Subjects were initially enrolled in 4 sequential dose groups (10-12 subjects per each group at 7 pg, 20 pg, 40 pg, and 80 pg). Within each dose group a minimum of 5 and a maximum of 6 subjects per cohort were enrolled sequentially to receive 1 or 2 subcutaneous (SC) doses of Albuferon administered 14 days apart. Dose escalation beyond 80 pg included single injections of 120 pg, 180 pg, 240 p g 320 pg and 400 pg. Plasma Albuferon concentrations and antibody to Albuferon were measured by ELISA. HCV RNA was measured using the Amplicor HCV monitor kit (Roche). 2' 5' oligoadenylate synthetase (OAS1) mRNA levels in whole blood was measured by a research-based Taqman PCR assay. Results: Of the 63 subjects currently enrolled, 96% were infected with HCV genotype 1 with a mean baseline viral burden of 2 million copies/mL. 78% of subjects enrolled in these cohorts had previously failed pegylated IFNa containing regimens. Albuferon pharmacokinetics showed linear increases in AUCo.8 and C, , , with mean terminal half-lives of 139-158 hours with doses of 80 pg and higher. T, , , occurred between days 4 and 5. There is an approximately 40% increase in AUC after the second injection in the double injection cohort at 80 pg. Albuferon was well tolerated and there were no discontinuations. No subjects developed detectable anti-Albuferon antibodies. Adverse events were transient and most were mild to moderate. The most common adverse events were injection site erythema (39%), headache (30%), fatigue (26%), mylagia (21%) and arthralgia (19%). The mean reductions in the nadir neutrophil counts in the higher single dose cohorts ranged from 40-60%. There was induction of OASl mRNA expression in all cohorts, with approximately 9-fold increase in median values at day 7. In the single dose cohorts (=80 pg), 52% of subjects showed a maximal 0.55 log or greater reduction in HCV viral load during the first two weeks. Also, 20% or greater reductions in ALT levels were observed in 36% of subjects in these single dose cohorts. Conclusions: In the ongoing phase 112 study, Albuferon demonstrated a favorable safety and immunogenicity profile. The pharmacokinetic profile supports dosing every 2-4 wks given its reduced clearance and extended half-life of up to 158 hours. All cohorts showed evidence of biological activity, as demonstrated by OASl induction, with anti-viral activity evident in the higher single dose cohorts. Disclosures: V Balan -Human Genome Sciences, Inc.: Investigator T Bambury -Human Genome Sciences, Inc.: Investigator G Everson -Human Genome Sciences, Inc.: Investigator W Freimuth -No relationships to disclose H Mesghali -No relationships to disclose J Murray -No relationships to disclose D Nelson -Human Genome Sciences, Inc.: Investigator L Novello -No relationships to disclose B Osborn -No relationships to disclose J Recta -No relationships to disclose G Subramanian -No relationships to disclose M Sulkowski -Human Genome Sciences, Inc.: Investigator J Zhong -No relationships to disclose INTRODUCTION Pirfenidone (PFD) is an orally bioavailable pyridone derivative that affects a variety of cytokines, including inhibition of TGFbeta, TNF-alpha, PDGF, and EGF. PFD has been shown in clinical studies to improve physiologic parameters in patients with pulmonary fibrosis. We have previously shown that PFD decreased hepatic fibrosis in two different rat models of cirrhosis of Hepatology 37: [797] [798] [799] [800] [801] [802] [803] [804] [805] 2002) . Our objective in this pilot clinical study was to evaluate the safety and preliminary activity of PFD in patients with cirrhosis of varying etiologies. METHODS All patients had histologic andlor clinical evidence of cirrhosis. PFD was given orally at a dose of 400 mg TID for 12 months. Physical examination and labs including ALT, AST, bilirubin, albumin and prothrombin time, platelet count were assessed at baseline and on monthly basis. HCV RNA levels were measured in patients with chronic hepatitis C (Cobas Amplicor HCV Monitor v2.0). Liver biopsies were obtained at baseline and after 12 months of treatment and were read independently by two hepatopathologists who were blinded to the biopsy sequence. Modified Histological Activity (HAI) Index of Knodell and Ishak fibrosis stage were used to assess changes in necroinflammatory scores and fibrosis stage, respectively. Change in steatosis was also assessed. RESULTS A total of 26 patients with cirrhosis due to hepatitis C (15), ethanol (S), amyloidosis (l), autoimmune disease (1) and Budd-Chiari syndrome (1) were included. The mean age was 57 years (range, 29-75) with 13 males. Liver biopsies at end of therapy showed a 2-point or greater reduction in the HA1 necroinflammatory score in 41% of the patients. Steatosis decreased in 33% of the patients, was unchanged in 42% and worsened in 25%. While there was no significant reduction in the Ishak fibrosis stage, improvement in interstitial fibrosis was noted. Evidence of cell regeneration was seen in some patients. HCV RNA levels were measured in 15 patients with chronic hepatitis C. At 6 months, 9 patients had a decrease in viral load, 2 patients remained unchanged and 4 patients displayed an increase in viral load compared to baseline. No patient had a sustained virologic response. 4 out of 15 (27%) HCV patients had normalization of ALT, 7 out of 15 (47%) had decreased ALT values, 1 did not change (7%) and 3 patients showed a modest increase in ALT (20%). PFD was well tolerated with the predominant drugrelated adverse events being nausea, photosensitivity rash, and itching occurring in 15% of the patients and which improved after 2 to 3 months of therapy. CONCLUSIONS: In this pilot study, treatment of cirrhotic patients with pirfenidone for one year was well tolerated. A significant reduction in necroinflammation ( 2 2-point reduction in HA1 grade) and steatosis was observed in a substantial proportion of patients. A subset of patients with chronic HCV infection showed on-treatment reduction in HCV RNA levels. Longer treatment duration or a less cirrhotic patient population may be needed to demonstrate effects on fibrosis. These data support conducting clinical studies to evaluate a potential role of I'FD in the treatment of steatosis, or in combination therapy for chronic hepatitis C. This work was supported by Grants of Marnac, Inc and InterMune, Inc. Disclosures: Arnulfo Alvarez -No relationships to disclose Gil Arechiga -No relationships to disclose Juan Armendariz-Borunda -No relationships to disclose BACKGROUND: Viral kinetic modeling of HCV response to interferon-based therapy provides important insights into factors associated with treatment outcomes. HCV/HIV coinfected patients appear to have lower overall response rates than that observed in monoinfected subjects, but the reason for this is not clear. We speculated that kinetic responses in key parameters would be decreased in coinfected subjects. METHODS: HCVlHIV coinfected patients and HCV monoinfected patients prospectively matched for treatment, genotype, age (k 5 yrs), gender, race and histology were evaluated. Coinfected patients were randomized within the context of a large US. multicenter trial (ACTG 5071) to receive pegylated interferon alfa-2a + ribavirin vs. interferon alfa-2a + ribavirin. Quantitative HCV RNA (Roche COBAS AMPLICOR) kinetic testing was performed at 0, 6, 12, 24, 48, and 72 hours and at days 7, 14. Non-linear regression and linear models were evaluated in an effort to best predict response and identify prognostic factors. RESULTS: Twenty-seven subjects underwent viral kinetic sampling and evaluation. These included twelve HCVlHIV case subjects (10 men, 2 women) and 15 matched HCV controls (some patients double-matched). The mean age of coinfected subjects was 46.1 years (range, 38-60). Among HIV+ subjects the mean CD4+ count was 325 cells/mm3 (range, 175-855). 11/12 had baseline HIV RNA 1400 copieslml. Mean HCV viral load was 6.6 log IUlml among coinfected vs. 6.2 log IUlml in controls. 75% of coinfected subjects had HCV Genotype 1. The remainder were Genotype 2.25% of coinfected patients had no detectable virus 24 weeks after completion of 48 weeks of therapy (SVR). Overall SVR in control subjects was 46.6%. Efficiency ( E ) of Phase 1 response (A) slope at 72 hours, lambda2 (slope of Phase 2 decline) were calculated. Efficiency of clearance ( E ) at 72 hours was highly associated with actual viral clearance across all groups (p=O.OOl) but A2 was not. Regression analysis failed to demonstrate a relationship between E and baseline CD4+ count, HCV viral load or genotype. In contrast, #955;2 was significantly associated with genotype. Viral kinetic parameters were not predictive of SVR. CONCLUSION: Viral kinetic modeling demonstrated that the efficiency of clearance at 72 hours was significantly associated with viral clearance during treatment. Coinfection status did not affect key kinetic parameters. PEG-IFN was superior to standard interferon in terms of viral clearance efficiency, particularly in the coinfected group. Diminished SVR in coinfected patients may be related to immune factors that are operative after reduction of viral loads to undetectable levels. The prevalence of chronic hepatitis C (HCV) infection is greater in African-Americans (AA) than in Caucasian-Americans (CA). However, small studies and post-hoc analyses of larger clinical trials have not found racial differences in disease severity at study entry, though indicate that interferon and ribavirin combination therapy is not as efficacious in AA as in CA. It is important to identify disease and patient characteristics that differ between AA and CA in a large study designed to ascertain whether the apparent racial difference in response to therapy can be explained. Aim: The Study of Viral Resistance to Antiviral Therapy in Chronic Hepatitis C (Virahep-C) is investigating reasons for the lack of sustained viral response (SVR) to pegylated interferon and ribavirin therapy in previously untreated patients chronically infected with HCV genotype 1. A key objective is to determine whether previously reported response rate differences between AA and CA exist in a large multi-center cohort. Methods: Virahep-C will recruit 200 AA and 200 CA from 8 clinical centers in the United States. Hypothesizing that AA and CA patients differ by factors that influence SVR which could explain racial differences in SVR, we compare demographic, clinical, histological, virological and health-related quality of life variables (HRQOL) before treatment begins. Liver biopsies performed within 18 months of study entry were assessed without knowledge of patient race by a central pathologist. Results: Based on the first 155 participants recruited, age and sex distributions are similar in the two racial groups with patients averaging 48 years of age and 2/3 being male More than 80% of the cohort has at least a high school education, but AA are more likely to have less than a high school education. Body mass indices tend to be higher among AA than among CA (p=.06), though waistto-hip ratios are similar (mean 0.9 in each group). There are no racial differences in mode of HCV transmission, estimated duration of HCV infection, or baseline serum HCV RNA levels. AA are more likely to have sub-genotype l b (p=.004).The Ishak fibrosis score does not differ by race, but severe lobular inflammation is greater in AA patients (p=.03). Diabetes and hypertension are more common in AA. The SF-36 physical component score in CA is similar to the general population and higher than among AA (p=.002). The SF-36 mental component score is similar in the two racial groups and comparable to the general population. Summa-wlConclusion: With few exceptions, pretreatment virological, clinical and liver biopsy findings were generally similar in the two racial groups. Differences in response are unlikely to be explained by these factors. lshak Flbrosls Score 58 (%) Severa lobular influnmation (%) Hypertension (%) Diabetes ( Bar-llan University, Ramat-Gan, Israel Background: Combination therapy with peginterferon alfa and ribavirin for 48 weeks achieves sustained virologic response rates (SVR) of 54-61%0 in patients with chronic hepatitis C. However, the SVR rates are highly variable according to baseline (HCV genotype) and on-treatment parameters (initial viral decline). Thus, it must be anticipated that current standard therapy recommendations lead to under-treatment in some and over-treatment in other individual patients. Objectives: Comparison between a dynamically individualized treatment schedule according to the early virologic response versus the standard of care combination therapy with peginterferon alfa-2a (40KD) (Pegasys @) (PEG-IFN) (180 pg qw) plus ribavirin (Copegus @) (RBV) (1000-1200 mg qd) for 48 weeks. The primary aim of the study was to increase SVR, while optimizing the available drugs, treatment duration, quality of life and the socio-economical burden of therapy. The secondary aim of the study was to enable a comprehensive analysis of virallhostlimmune correlates of response to treatment (ongoing). Methods: All patients (n=273) were initially treated with PEG-IFNlRBV for 6 weeks and initial viral kinetics were defined according to centralized serum HCV RNA quantifications (Cobas Amplicor@ HCV Monitor v2, Roche Molecular Systems) on baseline and days 0,1,4,7,8,15,22 and 29 . After classification into viral response categories at 6 weeks, patients were randomized (n=270) to individualized therapy (arms Al, A2, B1, B2, C, D) or continuation of standard therapy (STD arm). Treatment tailoring included: in rapid viral responders (RVR)discontinuation of RBV (Al) or shortening of treatment duration to 24 weeks (A2); in slow partial responders (SPR)addition of histamine (Bl) or prolongation of treatment to 72 weeks (B2); in flat partial responders (FPR)addition of histamine (C); in null responders (NUR)retreatment with high-dose of PEG-IFN (360 pg qw) plus RBV (D). SVR was defined as undetectable serum HCV RNA (< 50 IUlml) at the end of 24 weeks of untreated follow-up. Results: Demographic and baseline virologic parameters were similar in the standard and the individualized treatment groups. According to the initial viral decline patients were categorized as RVR (51% for genotype 1 and 94% for genotype 2-3), SPR (35% and 5% accordingly), FPR (5% and 0% accordingly), NUR (10% and 1% accordingly) or unclassified (1%). The overall SVR rates for genotype 1 patients were 49% for the individualized and 56% for the standard treatment arm (NS), and for genotype 2-3 patients 90% and 87% respectively (NS). SVR rates (by ITT analysis) according to HCV genotype and within each initial viral response category and arm are given in the Table. Conclusion: The overall sustained virologic response rates of 53% in genotype 1 and 88% in genotype 2-3 patients with chronic hepatitis C treated with peginterferon alfa-2a (40KD) in combination with ribavirin support previously presented data. Individualized treatment according to initial viral kinetics appears to be clinically feasible, but did not improve the sustained virologic response rate with the drugs and dosages usable at the time of this study. Nevertheless, the possibility that in rapid viral responders discontinuation of ribavirin does not decrease the sustained virologic response rate warrants future prospective trials. Supported by the European Community (QLK2-2000-00836), Hoffmann La-Roche and Maxim Pharmaceuticals. 10% 10% (Hepatology 2000; 32(Suppl) :844A). The rectally administered IFN is transferred not into blood but into regional lymphatic system and reaches to the thoracic duct via intraperitoneal lymphatics (Pharmaceut. Res. 1986; 3,116) . Therefore, the IFN suppository appears to have different antiviral mechanism from the conventional IFN injection. In this study, we performed a combination of the IFN suppository and IFN injection. The antiviral effect and immune-related markers were examined comparing with the controls (IFN injection therapy). (PATIENTS AND METHODS) Fourteen patients with chronic hepatitis C were given an IFN suppository and IFN alpha (6-10M units) injection daily for 4 weeks. After that, only IFN injection was continued three times a week for 20 weeks. The control group contains 9 patients with chronic hepatitis C, who received IFN alpha (6-10M units) injection daily for 4 weeks and were followed by the IFN injection three times a week for 20 weeks. The viral loads were 1.52-2400KIUlml (median 280KIUlml) in the IFN suppository group and 18.7-616KIUlml (median 97KIUlml) in the IFN injection group. The genotype population (lb/2a,Zb) was 5/9 and 3l6, respectively. There is no significant difference in the clinical background in the two groups. Serum HCV RNA was tested every 4 weeks. Peripheral blood NK cell activity and CD4lCD8 ratio were investigated before and 4 weeks after the beginning of the therapy. The IFN suppository contains low dose IFN alpha (BALL-1). (RESULTS) Serum HCV RNA turned negative in 73.3% of the IFN suppository group and in 66.7% of the IFN injection group (N.S.) after 4 weeks (end point of IFN suppository administration). The HCV RNA seronegative rates of the two groups were 80.0% and 77.6% (N.S.) after 24 weeks, respectively. However, while the NK cell activity was decreased in 78% of the IFN injection-treated patients after 4 weeks, only 36% of the IFN suppository patients had decreased NK cell activity (p=0.049, chi-square test). Furthermore, whereas serum ALT levels were significantly decreased after 4 weeks in the IFN injection group (83.5+47.9 IUll vs. 53.3+30.1 IUll, p=0.026, paired t test), no change was seen in the IFN suppository group (80.1IU/1+63.9 vs. 80.5+53.7, N.S.). (CON-CLUSION) The conventional IFN therapy decreased NK cell activity and serum ALT levels. However, the IFN suppository combination therapy maintained augmented NK cell activity and elevated ALT levels. These findings suggest that the IFN suppository continuously activates the hosts' immunity during IFN treatment. We used the IFN suppository only for 4 weeks; however, longer administration may lead to more frequent elimination of HCV The aim of the study was to evaluate the efficacy and tolerability of induction dose pegylated-interferon and ribavirin vs pegylatedinterferon and ribavirin as treatment strategies in relapsers to standard interferon + ribavirin in chronic hepatitis C patients. METHODS: 110 patients, virological relapsers after a first treatment with standard interferon and ribavirin for at least 6 months, were randomized to receive either pegylayed-interferon alpha-2b 2 Fglkg Qw plus ribavirin 800-1000mg Qd during 8 weeks then Peg-Interferon alpha-2b 1 pglkg Qw plus ribavirin 800-1000mg Qd during 40 weeks (n= 52 pts) or Peg-Interferon alpha-2b 1.5 pglkg Qw plus ribavirin 800 -1000mg Qd during 48 weeks (n= 58 pts). Efficacy assessments consisted of serum HCV RNA level by PCR and serum ALT at the end of treatment and after 24 weeks of follow up (week 72) and liver fibrosis with METAVIR score before treatment and at week 72. Safety evaluations included adverse events and laboratory tests. RESULTS: patients were not different for baseline characteristics in induction and non induction groups including sex (male 58 % and 71% ), mean age ( . Liver fibrosis METAVIR score decreased in 41% of patients in non-induced group and in 19% of patients in induced group (p=0,16) whatever the response (34% in sustained responders and 25% in non responders or relapsers). CONCLUSION 1) Combination therapy with pegylated-interferon is efficient in half of relapsers to standard combination therapy. 2) The response is similar to the response observed in naive patients. 3) 8 weeks induction dose does not increase the virological response. 4) We observed a regression of liver fibrosis in 30% of patients whatever treatment regimen and virological response. BackgroundAnemia has been shown to be an important factor in impairing HRQL in cancer patients receiving chemotherapy, with changes in Hb directly correlated with changes in HRQL (Gabrilove et al., J Clin Oncol2001). Since 29-36% of patients with HCV develop anemia as a side effect of combination IFNlRBV therapy (Rebetron@ Package Insert), it is important to assess the relationship between Hb and HRQL in this population. The objectives of this analysis were to (1) descriptively correlate changes in Hb to changes in HRQL in an anemic HCV-infected population; and (2) analyze the independent relationship of Hb to HRQL. Methods: HRQL scores and Hb values were obtained from clinical trial data of 185 anemic HCV-infected patients (Hb 10.8 t-0.9 gldL) who had been receiving IFNlRBV therapy for 12-14 weeks (Afdhal et al., DDW 2003) . During the 8-week double-blind phase, patients were randomized to receive either epoetin alfa 40,000 U once weekly or placebo. HRQL was measured using the Short Form-36 Health Survey (SF-36), an accepted and validated tool that measures 8 domains of HRQL (Table l) , and the Linear Analog Scale Assessment (LASA), which measures constructs of overall quality of life, energy, and activity. Patients were categorized into the following groups based on their change in Hb levels between randomization and week 9: (1) decrease in Hb; (2) Hb increase from 0 to <2 gldL; and (3) Hb increase 2 2 gldL. Changes in HRQL (by domain of each instrument) corresponding to the aforementioned Hb categories were summarized (Table 1) . Regression analyses were conducted to evaluate the independent impact of Hb change on HRQL improvement. The factors included in the regression model (in addition to Hb change) were: age, gender, baseline Hb, baseline HRQL domain, fibrosis status, RBV dose change, duration of HCV therapy, and HCV RNA. Results: Changes in Hb values were directly related to changes in HRQL for all 8 domains of the SF-36 and the LASA (Table 1) . HRQL changes were seen in an Hb-dependent manner; patients whose Hb increased 2 2 g/dL had higher improvements than patients whose Hb increased between 0 to <2 gldL. Similarly, patients whose Hb decreased from randomization to week 9 had mean decreases in HRQL for most of the domains (Table 1) and minimal increases for others. Regression analysis substantiated that the Hb change was a significant independent predictor (P=.006 to R.001) of HRQL change in all subscales of the LASA and the SF-36, with the exception of bodily pain and general health. Conclusions: Similar to cancer patients receiving chemotherapy, improvement in Hb is a strong independent predictor of improvement in the HRQL of HCV-infected patients. Patients on combination IFNlRBV therapy should be monitored and considered for anemia treatment to improve HRQL and potentially enhance their adherence to HCV therapy. virological response. The current trial was designed to see if better results could be achieved by retreating with higher doses of peginterferon alfa-2b (PEG2b) + weight-based ribavirin. AIM to compare the efficacy, safety and tolerability of three different doses of PEG2b + weight-based ribavirin among interferonlribavirin non-responders. METHODS: Patients were randomized to 1 yr of treatment with PEG2b 0.5, 1.5 or 3.0 mcglkglwk, plus ribavirin 12-15 mglkglday. Treatment assignment was stratified for sex, race, HCV genotype and histologic fibrosis. Treatment was stopped at 24 wk if PCR(+). Doses were reduced by 33% for toxicity; growth factors were not allowed. RESULTS: Patients: Enrollment took place between February 2001 and November 2002, with 963 patients recruited from 100 centers; data forms have been received on 794 thus far. Enrollment was stopped in the low-dose group after FDA approval of higher doses of PEG2b. The study population is 31% female, 16% African-American, 93% genotype l, and 63% F2/3/4. Efficacy: On-treatment virological response rates were dose-related at 24 wk but less so at 48 wk (see table) . This was partly due to a higher rate of discontinuation after a satisfactory response at 24 wk on the higher dose. On-treatment response rates were lower among African-Americans and patients with more advanced fibrosis. Sustained response data will be available for most patients by October 2003. 'Tolerability: The rate of dose reduction was 33% on PEG2b 1.5 mcglkglwk, vs. 45% on 3.0. Rates of discontinuation were the same for PEG2b 1.5 and 3.0 mcglkglwk: 25% vs. 24% overall, and 13% vs. 14% for adverse events. The frequencies of subjective adverse events were similar between the two groups. Some degree of iieutropenia was observed in 27% on 1.5 mcglkglwk, vs. 32% on 3.0; however, only 14 patients in the study discontinued treatment for neutropenia overall. CONCLUSIONS: 1) Thirty to 40 percent of interferon/ OF HIGH-DOSE PEGINTERFERON ALFA-PB + RIBAVIRIN M S L D ABSTRACTS 313A ribavirin non-responders achieved initial clearance of viremia on peginterferon alfa-2b + ribavirin. 2) Doubling the PEG2b dose to 3.0 mcglkglwk resulted in a higher viral clearance rate at 24 wk but a similar rate at 48 wk. 3) The higher dose of PEG2b 3.0 mcglkglwk was well tolerated, with a higher rate of dose reduction but an identical rate of discontinuation. Beliefs about therapy and psychosocial factors may influence the course of treatment and therefore could be important for developing comprehensive medical care plans that improve treatment adherence and optimize response to therapy. Objectives: (1) To describe baseline indices of social support, depression, and self-efficacy (i.e., perceived confidence in the ability to perform health behaviors necessary for successful treatment) of patients with HCV genotype 1 participating in a treatment trial of combination peginterferon and ribavirin therapy. (2) To assess the degree to which baseline psychosocial measures are interrelated. Methods: The sample consisted of the first 155 patients enrolled in the Virahep-C study, a multicenter, collaborative clinical trial, designed to examine reasons for non-response to HCV therapy in African American and Caucasian patients. Using a touch-screen computer, patients completed the 21-item Medical Outcomes Study Social Support survey (MOS-SSS); the 20-item Centers for Epidemiological Studies Depression scale (CESD); 24 questions about self-efficacy, and the SF-36 quality of life instrument. Questionnaires were completed during an 8-week period prior to initiation of therapy. Mean scores were calculated for each of the 4 subscales of the MOS-SSS (l=supported none of the time, 5=all the time): a) tangible support (material aid, behavioral assistance); b) affectionate (expressions of lovelaffection); c) emotional-informational (expressions of understandinglencouragement); and d) social interaction (others to have an enjoyable time with). Similarly, mean scores were calculated for 5 subscales of the selfefficacy instrument (O=no confidence, 10= high): a) obtaining help; b) communicating with physicians; and managing c) symptoms, d) depression, e) medications. Correlation analysis was used to compare each subscale with the CESD and SF-36. Results: Of the participants in the analysis, 43% were African American and 33% were female. On average, perceived social support and self-efficacy beliefs were high at the initiation of treatment. Mean scores for the tangible, affectionate, emotional, and social interaction subscales were 4.2 (SD=0.9), 4.4 (SD=0.8), 4.4 (SD=0.7) and 4.2 (SD=0.8), respectively. For self-efficacy, the mean scores were 8.0 (SD=2.0) for obtaining help; 9.2 (SD=1.5) for communicating with physicians; 7.3 (SD=2.2) for managing symptoms; 8.3 (SD=1.6) for managing depression; and 9.4 (SD=1.3) for managing medications. The distribution of social support and self-efficacy scores were similar for both racial groups and genders. Depressive symptoms were common in the sample at baseline with 55% having a score of at least 16 on the CESD and 7% having a score of at least 28. In general, the social support and self-efficacy subscales were unrelated to the CESD and to the subscales of the SF-36. Conclusion: African American and Caucasian study participants rate their social support and self-efficacy beliefs high in the weeks prior to initiating combination antiviral therapy for HCV, although depressive symptoms are common. Social support and self-efficacy instruments may measure different psychosocial constructs as compared with other instruments used in HCV research like the SF-36 and CESD. As a result, social support and selfefficacy may represent new and important predictors of adherence and sustained virologic response (SVR) in HCV treatment. Baseline data from Virahep-C will be used to evaluate factors associated with adherence and SVR to determine if racial differences exist among these measures. If so, this information can be used to develop new initiatives for educating patients and families prior to initiating HCV therapy. Background and aims: Interferon(1FN) a-2blribavirin therapy is an effective anti-viral therapy for the patients with hepatitis C virus (HCV) genotype 1. However, the response to this therapy is not satisfactory because only about 40-50% of the patients become sustained responder(SR)s. Further, the patients receiving IFN a-2blribavirin therapy have been frequently withdrawn because of severe adverse effects. The efficacy of IFN therapy is mediated by both genomic type and viral load and to immunological response of the hosts. Reportedly, some amino acids were identified to modulate host's immunological response. In order to achieve more higher SR ratio after IFN a -2blribavirin therapy and to decrease the number of withdrawn patients, it seems important to improve the nutritional condition and to upregurate potential immunological response by modifying the balance of amino acid. In the present study, we investigated the dynamics of a panel of 41 amino acids during IFNa-2blribavirin therapy, then analyzed the effects of nutritional condition on the clearance of HCV. Materials and methods: Seventy patients with chronic HCV infection were included after informed consent. Six-million unit of IFNa-2b in combination with ribavirin (600-800mglday) were given during 6 months. During the first 2 weeks of the therapy, IFN a -2b was given daily. Blood samples were obtained after overnight fasting at Zweeks, 8weeks, and 24weeks of the therapy and served for the analysis of amino acids. At the same time, peripheral blood mononuclear cells (PBMC) were collected and the number of IFNy-and IL4-producing cells were measured by FACS analysis. ThllTh2 ratio was calculated as following; Thll Results: HCV RNA was undetectable in 22%, 52%, and 78% of the patients at 2, 8, and 24 week of the therapy, respectively. ThllTh2 ratio decreased gradually during the therapy. ThllTh2 ratio at the end of therapy (24 week of the therapy) was significantly smaller than that at the beginning of the therapy (17.8210.9 vs 13.558.0, p=O.O11). However, there was no significant correlation of Thll Th2 level with the rate of HCV eradication. Total amino acids (TAAs) and branched chain amino acids (BCAAs) decreased gradually during the therapy. Even at 8week of the therapy, there was a significant decrease in TAA and BCAA (p=0.017 and p 75kg 1200mgl day in two divided doses). Randomisation was stratified according to viral genotype (1 and non-1). The primary end point was the virological outcome of treatment with sustained response defined as an undetectable HCV RNA PCR 6 months post cessation of therapy. In addition there were a number of secondary aims: 1. Viral kinetics. Serum was collected at baseline, days 3,7,10 14 and week 12 for quantitative rt PCR to determine HCV viral load. The ability of early phase kinetics and the presence or absence of a 2 log drop at week 12 to predict eventual outcome is assessed in the context of mild disease. 2. Health economics. The collection of resource use data in HCV infected patients who have mild, moderate and cirrhotic disease was performed as an integral part of this study. This data has been incorporated into a Markov chain based model to assess likely future costs of the disease and their possible avoidance by treating mild cases. Response rates from this trial will form the basis of the model. 3. Quality of life assessments. SF36 and Euroquol data were collected at baseline, weeks 1 2 2 4 and 48 during treatment and at post treatment weeks 12 and 24 in order to compare quality of life between treated patients and controls before during and after therapy. Results. 98 patients received treatment and were matched with 98 controls (no treatment). Sustained viral response rates will be available at the end of July when the follow up period of the trial is complete and will be included in the presented results. End of treatment results are shown in table 1. 30% of patients infected with genotype 1 and 61% of patients infected with genotype non-1 achieved end of treatment responses. Sustained viral response rates to date are shown in Table 2 .20% of patients infected with genotype 1 and 53% of patients infected with genotype non-1 have achieved sustained viral responses so far. 25% of patients in the treatment group were unable to complete at least 6 months of therapy due to side effects of medications. There were 4 hospitalisations in the treatment group, 3 of which were related to adverse events. Patients with mild HCV have lower response rates to those with more severe histological change on liver biopsy. Side effects are common and frequently result in treatment discontinuation, lowering response rates. The decision to treat patients with mild histological change must be weighed against the side effects. Deferring treatment until later in the disease process does not prejudice response to therapy and may increase likelihood of success. The HCV NS5B polymerase is essential for HCV viral replication and infectivity as demonstrated in a chimpanzee model. The enzyme adopts a unique molecular structure that resembles a "thumb-palm-finger" which has some features different from other known DNA and RNA polymerases, highlighting the attractiveness for this enzyme as a target of antiviral therapy. From an in-house high throughput screening campaign and the subsequent lead optimisation, we have identified a novel chemical series of substituted thiophene-2-carboxylic acids that have good potency against HCV polymerase in vitro. X-Ray crystallography studies revealed that these compounds bind to an allosteric site that is -35 angstrom away from the active site. Further studies on selected members of the series confirmed the ability of these compounds to inhibit the sub-genomic replication of HCV in Huh-7 cells. Moreover, a representative compound was also found to have good in vitro safety index and favourable in vitro and in vivo metabolic and pharmacokinetic properties. In an open and prospective trial we investigated, if a pre-treatment with citalopram as an SSRI can reduce the frequency of IFN-associated depression in hepatitis-C-infected psychiatric risk patients during methadone substitution. 36 patients with a chronic hepatitis C were treated with pegylated interferonalpha 2b and ribavirin according to body weight. 11 patients without any psychiatric history (group A) were compared to 25 patients during methadone substitution. The methadone substituted patients were separated into two groups: the first group (group B, n = l l ) were only treated with IFN-alpha and ribavirin and the second group (group C, n=14) received a two week pre-treatment with citalopram (20mglday) before combination treatment with IFN-alpa and ribavirin was started. Antidepressant treatment was continued over the study period of four months. The Hamilton Depression Rating Scale (HAMD, 17-item) was used and major depression defined as >20 points. Patients were followed over the first 4 treatment months. Group differences were calculated with ANOVA for parametric and chi2-test for non-parametric scales. Results: HAMD scores at baseline were significantly higher in the psychiatric groups as compared to the controls (p=O.OlO Dallas, T X BACKGROUND: Peg plus RBV is the mainstay of chronic hepatitis C treatment. However, an upper limit in terms of safety and efficacy for dosing with Peg has not been evaluated. METHODS: We studied a group of nake patients using either conventional weight-based dosing of Peg; (1.5 pglkg) or double the dose (3.0 pglkg) plus RBV 13+2mg/kg/day for 48 weeks. Genotype 1 patients were randomized 1:l and to receive medication accordingly in unblinded fashion, with intent to enroll a total of 1,100 naYve patients N, an additional 300 previous NR or R to any type of interferon i RBV in a non randomized arm, using only the 3.0 pglkg Peg dose. Values for HCV RNA at week 12 were compared to those obtained at baseline. RESULTS: To date, 573 patients 306 N, 193 NR, 74 R have enrolled in the study. 175 patients 106 N, 48 NR, 21 R have reached week 12. HCV RNA results at week 12 are shown in the table below. Treatment was well tolerated in most patients. Dose reductions were required in 20% of N 1.5pglkg Peg and 29% of N 3.0pglkg Peg patients. Side effects were equivalent between the two groups as shown in the table below. Serious adverse events (SAE's) were less than 5% and equally observed in both arms of the N patients. None were directly attributed to study drug. There were 25 patients in the wk 12 analysis that discontinued therapy. The reasons these patients were discontinued after reaching wk 12 were side effects (36%), +HCV RNA (32%). It is interesting to note that 40% of the patients that were discontinued had 2 2 log drop or negative HCV-RNA at week 12 and a positive HCV RNA was not a reason for their discontinuation. CONCLUSION: 3.Op.glkg Peg dosing does not appear to improve 12 wk viral response, but we await 24 wk results and sustained viral response rates. Increased side effects as a result of doubling the interferon dose were not observed. More data is needed to verify the long-term efficacy and safety of this dosing regimen for patients with chronic hepatitis C . This study was supported by a grant from Integrated Therapeutics Group a subsidiary of Schering Plough. Background and Aim: One of the major adverse effects of the combination therapy for chronic hepatitis C is ribavirin-induced hemolyhc anemia. However, little is known about the mechanism of this anemia. Oxidative stress has been suggested as potentially important pathological mechanism in hepatitis C. This burden may cause peroxidation of erythrocyte membrane phospholipid in conspiracy with the accumulation of ribavirin triphosphate in the erythrocyte, which potentially attenuates the mobility of erythrocyte membrane. The aim of this study was to examine the change of fatty acid composition in erythrocyte membrane and the effects of vitamin E and vitamin C on fatty acid composition in combination therapy. Methods: Thirty-nine patients with chronic hepatitis C were enrolled in this prospective study. They were randomized to receive daily oral vitamin (500 mg of vitamin E plus 750 mg of vitamin C) (vitamin group) or none (control group), in addition to injections of 6 million units of interferon-a-2b daily for two weeks, followed by thrice-weekly for 24 additional weeks, plus daily oral ribavirin (600 or 800 mg) for 26 weeks. Blood samples were obtained at 0, 2, 4, 8, and 26 weeks after initiation of therapy. Phospholipid was separated by one-dimensional thin-layer chromatography after extraction of total lipid from the erythrocyte ghosts. Following transmethylation, fatty acid methyl esters were quantified by gas chromatograph. a-tocophenol concentrations in erythrocyte were determined by high performance liquid chromatography. Plasma thiobarbituric acid reactive substances (TBARS) were measured by spectrofluorometer. Results Seven patients were obliged to suspend or cease receiving therapy because of adverse effects including anemia by 8 weeks after initiation of therapy. Twenty-one patients have completed the assigned therapy up to now. Among the 26 kinds of fatty acid analyzed, the mean content of 205n-3 polyunsaturated fatty acid (PUFA) significantly decreased at 8 (0.96 5 0.12 mol% vs. In a cross-sectional study we investigated MxA expression in patients with chronic HCV infection(n = 60 ) and in patients with chronic HCV infection receiving IFN-alpha therapy (n = 23) as well as in healthy controls (n = 22). In a prospective study with 16 chronically infected patients (HCV genotype 1) known to be IFN non-responders, we followed MxA gene expression during combination therapy with pegylated IFN-alpha2b and ribavirin. Results: Patients with chronic HCV infection showed higher MxA gene expression levels than healthy controls, indicating that hepatitis C virus induces IFN production. However there was no correlation between MxA gene expression and clinical parameters (viral load, ALT, liver histology). Patients with chronic HCV infection receiving IFN-alpha had significantly higher MxA levels than patients without treatment or healthy controls. Further we addressed the question whether MxA expression in PBMCs during therapy in previous non-responders could be a predictive marker of therapy outcome. MxA expression was clearly induced in all but one patient compared to pretreatment values (median: 10.2 fold). Importantly, the level and kinetics of MxA expression did not correlate with the response to antiviral therapy. Surprisingly, one patient showed an unusually low pre-treatment level of MxA expression and a total lack of MxA induction during IFN therapy, which was associated with complete non-response to therapy. In a neutralization assay, we could detect neutralizing antibodies to IFN-alpha2b in the serum of this patient whereas no neutralizing antibodies were found in all others patients. Hepatitis C virus (HCV) infection is a common cause of transfusion acquired hepatitis and is today a major health problem throughout the world. The present study reports the prevalence of Hepatitis C virus (HCV) infection in general population and in various high-risk groups from the city of Hyderabad in South India. A total of 308 out of 3589 (8.6%) people (both general and risk groups) tested positive for HCV RNA by RT-PCR, while anti-HCV antibody positivity, as determined by third generation EIA, was found to be 5.8% (209/3589). This suggests that a number of cases go unreported, as screening of blood and blood products is done primarily by ELISA. Among 124 chronic renal failure (CRF) patients with history of either renal transplant or hemodialysis, 46% were infected with HCV alone, 6.5% HBV alone while 37.1% were found to be co-infected with bothe HCV and HBV. Our findings implicate these viruses as the major cause of post transplant hepatitis in Indian patients with CRF and indicate the necessity for immediate implementation of stringent screening procedures for these viral infections. Additionally we report here that tattooing and slashing (a cultural practice among one sect of muslims) increase the mode of transmission of HCV infection. In addition, we also report a high incidence of HCV infection ( BACKGROUND In Japan, long-term treatment with glycyrrhizin, given as SNMC, is used to reduce ALT and the risk of hepatocel-GLYCYRRHIZIN GIVEN AS STRONGER NEO-lular carcinoma in patients with chronic hepatitis C. Phase 1/11 studies confirmed the ALT lowering effect of glycyrrhizin therapy in western patients with chronic hepatitis C. The treatment duration in these studies was limited to 4 weeks. ALT response was lost during follow-up. AIM 1) To evaluate which dose frequency is required beyond week 4 to maintain the initial ALT response. 2) To evaluate the effect of SNMC treatment on liver histology. METHODS: HCV-RNA positive patients with ALT >2x ULN, fibrosis stage 2 3 or necro-inflammation score 2 6 (Ishak's score) who were not eligible for interferon therapy (prior non response, absolute contraindications) could enter this multi-center, randomized, open phase I1 clinical trial. All patients were treated 4 weeks with six infusions weekly of 100 ml SNMC (Minophagen Pharmaceutical Co. Ltd., Japan) containing 200 mg glycyrrhizin. Patients with an ALT response at week 4 (defined by a decrease of more than 50 percent of the baseline value or ALT 51.5 ULN) were randomized to continue treatment in three dose frequency groups for a total of 26 weeks. SNMC was given 6 times weekly (group l), Objectives: It is important to maintain reduced serum alanine aminotransferase (ALT) levels in cases with chronic hepatitis C (CH-C) that do not respond to interferon (IFN) and in those with no indication of IFN therapy. We reported previously that dietary restriction of iron intake reduces serum ALT levels in such patients. We evaluated CH-C patients treated with iron-restricted diet for two or more consecutive years, mainly focusing on the balance of energy intake, physical examination, and changes in hematological indices of nutrition. Methods: Twenty-two patients with CH-C (males, 18; females, 4; mean age, 56 year-old) that consulted our outpatient department were enrolled in this study. The inclusion criteria were as follows: 1) elevation of ALT levels above the upper normal limit for 3 months or more; 2) positive tests for HCV-antibody and HCV-RNA; 3) absence of other causes of CH (alcoholic liver disease, drug-induced liver injury, hemochromatosis) and negativity for hepatitis B surface antigen and for serum anti-nuclear and anti-mitochondria1 autoantibodies. Twenty cases had received IFN therapy for more than 12 months before the beginning of the study; none of them responded to IFN therapy. Dietary prescriptions included iron intake 7 mglday or less, energy intake 30 kcallkglday, protein intake 1.1-1.2 glkglday, and a fat energy fraction of 20%. Nutritional balance was evaluated based on meal records, and instructions was given when necessary. Results: The average energy intake before dietary prescription was 2184 kcal(36.7 kcallkg)lday, and it was significantly reduced to 1655 kcal(28.5 kca1lkg)lday (p < Om), and then maintained stable at 30 kcallkglday. The average protein intake before dietary prescription was 85.7 g (1.45 glkg)lday and it was reduced to 1.1-1.2 glkglday after the prescription. The average fat intake of 66.5 g (1.1 glkg)lday and the average fat energy fraction of 27% before the dietary prescription were significantly decreased to 30.8 g (0.52 glkg)lday; p < 0.01 and 16% (p < 0.001), respectively, after dietary instructions. The fat energy fraction was maintained at a level of 20% or less. Carbohydrate intake did not change remarkably during the observation period, although the carbohydrate energy fraction significantly (p < 0.001) increased. The average iron intake decreased significantly (p < 0.001) from 9.6 (before) to 6.1,5.2, 5.1,5.2, and 5.1 mglday 6, 12, 18, and 24 months after die* prescription, respectively. Body mass index (Bh4I) before diet prescription was 23.9 on average; BMI had no significant change throughout the course. The body fat percentage was 24.6% on average before the diet instructions, and it significantly decreased after the diet. The average values of aspartate aminotransferase and ALT before diet prescription were 65 IUll and 66 lull, respectively, and they were significantly reduced to 48 IUll and 49 IUll, respectively, after 24 months (p < 0.01). Serum iron levels significantly decreased after 18 (p < 0.01) and 24 (p < 0.05) months, while unsaturated iron binding capacity tended to increase. The average serum ferritin levels were 376, 210, 189, 189, 141 nglml before and 6,12, 38, and 24 months after diet, respectively; there was a significant reduction (p < 0.01) in the values measured before and after the diet instructions. The average levels of hemoglobin, albumin and cholinesterase did not change significantly during the follow-up period. Conclusions: Restriction of iron intake is safe and well tolerated for a long period. The results of our present study suggest that decreased dietary intake of iron may constitute an important adjuvant therapy in patients with CH-C. CHRU, Nice, France; C Henquell, CHRU, Clermont Ferrand, France; C Dizrcha, S Ughetto, P Dechelotte, Hotel-Dieu, Clemont Ferrand, France; H Lafeuille, CHRU, Clermont Fewand, France; G Bommelaer, Hotel-Dieu, Clemont Fewand, France Peginterferon (PegIFN) Background: HCV Co-infection is common among patients with HIV disease. It has been documented that HIV co-infection accelerates the course of liver disease in patients with HCV, and that liver failure is higher in co-infected patients. At this moment, there is no effective treatment for HCV non-responders to interferon therapy. Mono-therapy and interferon-ribavirin combination is only effective in 2-3% and 5-10% interferon resistant patients respectively. Treatment strategies to slow the progression to cirrhosis and to prevent liver failure in this population are needed. Objective: This is a report of a study that examines the efficacy of Peg IFN alfa 2-a (Pegasys) vs PegasyslRBV 800mg in co-infected HCVlHIV patients that have not responded to a previous 6-12 months course IFN-alfa. The study also examines the histological benefit of treatment in this population. Patients and Methods: 76 patients were randomized 1:l to receive either Pegasys 180mcg weekly x 24 weeks (group 1) or Pegasys 180mcg weekly plus 800mg RBV for 24 weeks (group 2). Patients that have HCV non-detectable or 210g decrease from baseline at week 24 (groupl), were added RBV 800mg. All similar responders (group 2) continue treatment for a total of 48 weeks. Baseline demographics were similar between both groups. More than 70% of patients were nonresponders to IFN monotherapy.Most patients in both groups (80%) were genotype l. The mean HIV Baseline log was 2.92 (SD.64) group 1, vs 2.85 (SD.57) group 2 and most patients had baseline CD4 levels>400 cells, and were in stable antiretroviral therapy. The majority of patients (81%) are non cirrhotic, with mean grading group 1 6.63 (SD 3.24), group 2 7 (SD 3.64), and mean staging, group 1 3.51 (SD 2.2), and group 2 3.48 (SD 1.70 Conclusions: This study is completed and pending some results, that will be available for presentation. Sustained virological response(SVR) (Intention to treat), will be of the order of 5-20%. (11 group 2 results are pending). In this study, 10 patients were discontinued because adverse events and 10 were lost to follow up, before week 24th.SVR in patients that completed at least 24 weeks of therapy will be of the order of 7-26%.A significant number of end of treatment responders are relapsing at week 72th. Histology analysis show improvement in both groups of the mean grading and staging after treatment. In responders and relapsers the FPR becomes static or regressive.These results show that Pegasys lRBV therapy is effective in this population.The study also suggests that longer duration of therapy ,and higher doses of RBV should be studied in coinfected nonresponders. * -Pending results, will be available for presentation. Disclosures: Josi. Rodriguez-Orengo -No relationships to disclose Maribel Rodriguez-Torres -Roche Laboratories: Investigator; Other: Grant to perform study. Adverse events by week 12 of therapy were grade 2 to 3 in severity and were therapy related. See Table 2 . Sixteen patients who continued therapy to week 24, three dropped out due to adverse events (1 anemia, 1 hyperbilirubinemia and 1 related death due to self-overdose). This patient had a history of hypothyroidism and mild untreated depression. At 48 weeks of therapy, 1 patient discontinued treatment due to suicidal thoughts. There was no difference between the ethnic groups regarding the drop out rate due to adverse events. HIVfHCV coinfected population had a poor tolerance to Pegylated Interferon and Ribaiirin. This was markedly increased in the first 12 weeks of therapy. The use of high dose Pegylated Interferon and Ribavirin led to a significant drop out rate in comparison to the low dose pegylated interferon. Background During liver injury, poorly characterized factors activate quiescent hepatic stellate cells (HSC) to become proliferative, matrix-synthesizing myofibroblasts. Activated HSC are the major source of liver collagen and thus, play a key role in the fibrotic response to liver injury.The SNS appears to promote fibrosis in injured livers because hepatic fibrosis is increased in the spontaneously hypertensive rat, which has an overactive SNS. Conversely, prazosin, an adrenoceptor antagonist, inhibits fibrosis in toxin-damaged rat livers. HSC express several neuronal proteins, including glial fibrillary acidic protein (GFAP). They also contain synaptic vesicles and receive autonomic fibers. Therefore, our HYPOTHESIS is that HSC are hepatic neuroglial cells that produce and respond to neurotransmitters in order to become activated during liver injury. Methods: HSC were isolated from normal mice and Dbh-lmice that cannot produce norepinephrine (NE) due to targeted disruption of the dopamine P-hydroxylase (Dbh) gene. Lysates of culture-activated HSC were analysed by RT-PCR, immunoblot and HPLC to determine if they express adrenoceptors, catecholamine biosynthetic enzymes andlor produce NE. The effect of adrenoceptor antagonists and NE on HSC growth in vitro was also assessed. Then in vivo activation of HSC by hepatotoxic diets was evaluated in control and Dbh-lmice by comparing numbers of a-smooth muscle actin (ASMA)+ cells with immunohistochemistry and the induction of TGFP-1 and collagen a-1 gene expression by ribonuclease protection analysis of whole liver RNA. Results: HSC express a-and P-adrenoceptors, tyrosine hydroxylase and Dbh. HSC from control, but not Dbh-I-, mice release NE. Endogenous NE is an autocrine growth factor for HSC because a-and P-adrenoceptor antagonists inhibit proliferation of HSC cultured from control mice. Moreover, HSC from Dbh-lmice, which cannot make NE, grow poorly in culture and are rescued by addition of NE. Exogenous NE also promotes HSC proliferation. Inhibitor studies demonstrate that the latter effect is mediated via G-protein coupled adrenoceptors that activate mitogen activated kinases and phosphatidylinositol 3 kinase pathways. HSC activation in response to diet-induced liver injury is inhibited in Dbh-Imice, as evidenced by reduced hepatic accumulation of ASMA (+) HSC and inhibited hepatic induction of Background & Aims: Hepatic cirrhosis is six times more prevalent in obese individuals than in the general population, and obesity is one of the risk factors for liver fibrosis in which plasma adiponectin levels are decreased. Adiponectin is an adipocytokine, which we previously identified by screening adipose-specific genes in the human cDNA project. Hepatic stellate cells (HSCs) play central roles in liver fibrosis. When they are activated, they undergo transformation to myofibroblast-like cells, then proliferate, migrate, produce transforming growth factor-pl (TGF-Pl), and various extracellular matrix proteins, and express a-smooth muscle actin (a-SMA). We previously reported that adiponectin suppresses not only the proliferation and migration of HSCs, but also the TGF-pl-induced fibrogenic gene expression in HSCs. Adiponectin could have biological significances in liver fibrosis. In this study, in order to clarify the effect of adiponectin on liver fibrosis in vivo, we tested the role of adiponectin on liver fibrosis using adiponectin-knockout (KO) mice and adenovirus mediated adiponectin expression system. Methods: (1) To investigate the anti-fibrogenic effects of physiological concentrations of adiponectin, male wild type (WT) mice and KO mice were used. Mice were each injected with a dose of carbon-tetrachloride (CCl4) (300 pllkglbw) intraperitoneally twice a week for 12 weeks to induce liver fibrosis. (2) To investigate the anti-fibrogenic effects of excessive concentrations of adiponectin, male WT mice were used. Mice were injected CC14 (1000 pl/kg/bw) intraperitoneally twice a week for 12 weeks. Mice were divided into 6 groups. Control, received an injection of corn oil only; Gr.1, mice treated with CC14 for 12 weeks; Gr.2, mice treated with CC4 for 12 weeks after infusion of adenovirus producing adiponectin (AdADN); Gr.3, mice treated with CC4 for 12 weeks after infusion of adenovirus producing P-galactosidase (AdLacZ); Gr.4, mice treated with CC14 for 12 weeks with AdADN infusion at 6 week; Gr.5, mice treated with CClb for 12 weeks with AdLacZ infusion at 6 week. Results: (1) KO mice showed extensive liver fibrosis with an enhanced expression of TGF-P1 and connective tissue growth factor (CTGF) compared to wild type (WT) mice (P<0.05). The hydroxyproline content and the numbers of a-SMA positive cells in mice liver significantly increased in KO mice. (2) The fibrosis areas were significantly decreased in Gr.2 and Gr.4 compared to those in Gr.3 and Gr.5, respectively. The hydroxyproline content in mice liver significantly decreased in Gr.2 and Gr.4 compared to that in Gr.3 and Gr.5, respectively. Moreover, in Gr.4, the hydroxyproline content was significantly decreased compared to that in 6-weeks of C C 4 treatment even though CCl, was given for an additional 6-weeks (total 12 weeks). Conclusions: The findings indicate that adiponectin attenuates liver fibrosis and could be a novel approach in its prevention. Disclosures: (tTG) is observed in mature scars and might promote wound repair by protecting the neomatrix from degradation by matrix metalloproteinases (MMPs).However, such ECM crosslinking has the potential to hinder the matrix remodelling required for resolution of liver fibrosis. We have explored this hypothesis by examining expression of tTG and its crosslink product epsilon-(gamma-glutamyl) lysine in livers of: a) rats administered CC14 for 6 weeks (6wk C C 4 cohort)after which liver fibrosis resolves spontaneously after 28 days; b) rats administered CC14 for 12 weeks (12wk CC14 cohort) which develop micronodular cirrhosis; c) 12 wk CCl4 treated rats allowed to recover for 365 days after last CC14 dose (12 wk+365d cohort) which show partial resolution of fibrosis but with persistence of a macronodular cirrhosis. Although tTG was identified by immunohistochemistry within and around fibrotic bands in 6 wk CCld cohort, no crosslinks were detectable. However, both tTG and crosslinks were detected in and around mature fibrotic bands in the 12wk CC14 cohort and in persisting fibrotic bands in the 12wk+365d cohort. Western blotting of liver homogenates from these two cohorts using antibody against the crosslink revealed a major product of 160 kD which was degradable to lower molecular weight products following incubation of homogenates with bacterial collagenase but was refractory to MMP-2. To evaluate if ECM crosslinking in livers correlated with its resistance to MMPs, 10 micron cryostat sections of livers from the three cohorts were incubated with purified active forms of MMP-1 or MMP-2 for 8 hr at 37 C and any residual ECM was stained with Sirius red. ECM of 6 wk CC14 cohort was effectively degraded by these MMPs but that of 12wk C C 4 and 12 wk+365d cohorts was only minimally degraded. HSC freshly isolated from normal rat or human liver showed no tTG expression whilst HSC activated for 7-15 days on plastic substrate expressed tTG protein (by Western blotting) and mRNA (by RT-PCR). There was tTG in cell lysates and in conditioned media of activated HSC. Pure rat tail type I collagen incubated with conditioned media of activated HSC incorporated crosslink antigen, confirming that HSC secreted functional enzyme. However, no crosslink was found in type I collagen incubated with media from HSC which had tTG gene silenced by cell transfection with small inhibitory RNA. Cultured HSC which had tTG gene silenced had enhanced apoptosis (by acridine orange staining) following serum deprivation in culture. This suggests tTG or tTG activity promotes survival of these cells. We conclude that in persistent, incompletely resolving liver fibrosis there is evidence of tTG mediated crosslinking of the liver ECM and resistance to exogenous MMPs. Both features are lacking in ECM of fully resolving fibrosis. ECM crosslinking by tTG might limit resolution of cirrhosis. Activated HSC are a potential source of tTG following liver injury and, as this protein supports their survival, it might contribute to their persistence in cirrhotic liver. We recently defined multifunctional biochemical properties of the Long and Frizzled variants and hypothesized that they may be inherent in the 192-aa-Long-specific and the 235-aa-FZ-specific modules, respectively. We thus transfected pCDNAS.l-V5-His vectors containing variant-specific sequencer3 in mhAT3FlS315 hepatoma cells. This cell line stably expresses a truncated HNF3 protein turning off endogenous expression of liver-specific genes, such as albumin or C18. Consistently, mhAT3FlS315 hepatoma cells transfected with an empty vector showed no endogenous C18 expression. By immunohistochemistry, using variant-specific and tag-specific antibodies and after analysis by confocal microscopy, Long was localized in large supranuclear vacuolae suggesting protein storage/maturation in Golgi structures or in beaded perinuclear vacuolae suggesting RER cysternae. In contrast, FZ was strongly detected in close proximity to the plasma membrane of single cells. Clusters of FZ-transfected cells showed a strong and dense signal at sites of cell-cell contact, outlining single cells. These findings were confirmed by transfection of prep-7 vectors containing full-length Long or FZ forms of human C18 and confocal microscopy analysis after immunohistochemistry using an antibody directed against an epitope common to all C18 forms. Immunoblot analysis of conditioned media showed that the fulllength Long form was secreted into the medium but not FZ. The latter was detected as a heavily glycosylated protein (> 300 kD) in the cell layers, predominantly in the triton-X-100 insoluble pellet after sonication and 10% SDS solubilization, suggesting binding to extracellular matrix proteins. Finally, glycosydase analysis of FZ and Long N-terminal modules showed an important increase in mobility after PNGase F plus sialidase A digestion. These data show that, in addition to the previously described plasma (Long) and tissue (Short) forms of Cl8, the FZ form constitutes the pericellular matrix form, suggesting that the specific modules of C18 regulate tissue targetting through protein-protein interactions. Backgr0und:During liver fibrosis, hepatic stellate cells (HSC) acquire an activated phenotype, proliferate and produce an excess of collagens. Mycophenolic acid (MPA) is a known immunosuppressive drug. which inhibits the proliferation of B-and T-lymphocytes by Inosine Mono Phosphate Dehydrogenase (IMPDH) inhibition, causing an intracellular guanosine depletion. In addition, MPA has shown to inhibit growth of mesangium cells in the kidney and of skin-and tenon fibroblasts. Therefore, we hypothesize that the proliferation of HSC may also be influenced by MPA. In this study we explored whether MPA is able to inhibit the proliferation of primary isolated rat HSC in vitro. Furthermore, we studied the in vitro effects, mechanism of cellular uptake, and in vivo pharmacokinetics of MPA coupled to mannose-6-phosphate modified human serum albumin (M6P-HSA). M6P-HSA is a newly developed, HSC-specific drug carrier, homing towards the upregulated M6PlIGF-I1 receptor on activated HSC. In this way we hope to achieve cell-specific delivery of MPA to the HSC avoiding undesired effects of MPA on the immune system. Meth-odslResults: In primary cultures of HSC a dose dependent reduction in the number of BrdU-positive nuclei was observed after 24h incubation with 750,3000 and 6000 nM MPA, as assessed immunohistochemically. Coupling of MPA to M6P-HSA via a biodegradable ester linker resulted in a conjugate with a maximum drug: carrier ratio of 1.2:l. Analysis of the synthesized constructs was performed by HPLC, FPLC and MS. This conjugate was able to inhibit 3T3-fibroblast growth, as detemined by BrdU-incorporation (ELISA) in a dose dependent manner, reducing proliferation to 38.0 2 14.5%, 30.0 t 18.0% and 18.0 2 16.8% of control at 120, 240 and 480 uglml of conjugate, respectively. When cells were co-incubated with an excess of M6P-HSA, a competitor for receptor mediated uptake of the conjugate, the anti-proliferative effect was reduced by 69 YO. The organ distribution of the conjugate, 1251 labeled, was evaluated by IV injection of a tracer dose in rats 3 weeks after bile duct ligation (BDL-3). 47.3 t-4.1 % of the injected dose accumulated in the liver after 10 min of injection. In contrast, spleen and thymic gland accumulated 1.47 f 2.65% of the injected dose. Intra-hepatic distribution was assessed in BDL-3 rats by immuno-histochemical double staining for HSA and specific markers for Kupffer cells (ED-2), HSC (DesminelGFAP) or Endothelial cells (HIS52). M6P-HSA-MPA showed a non-parenchymal distribution and co-localization with HSC and Kupffer cells. Conclusions: MPA is able to inhibit the proliferation of primary isolated rat HSC. This suggests that stellate cells are dependent on intracellular IMPDH activity to proliferate. Coupling of MPA to MCiP-HSA, a stellate cell specific drug carrier, resulted in a pharmacologically active construct, able to inhibit fibroblast proliferation in vitro after specific uptake via the M6P/IGF-II receptor. The major part of the injected conjugate accumulates in the HSC and Kupffer cells of the liver, and avoiding uptake in the major resident organs for Band T-lymphocytes. Future studies will assess the advantage of this first HSC-selective compound in animals with liver fibrosis. Background: Platelet-derived growth factor (PDGF) is the most potent stimulator of migration and proliferation of mesenchymal cells. The expression of PDGF-p-receptor is increased during liver fibrosis. Our aim was to investigate the effect of p3-integrin subunit blockade using the specific non-peptidic inhibitor EMD409915 on migration and proliferation of hepatic stellate cells (HSC) and human foreskin fibroblast (HFF). Materials and methods: Cell migration of human HSC and HFF was assessed using a scratch assay. Scratches of 600-700 pm width were made in confluent cell monolayers of cells following 24 h starvation in serum-free medium. After wounding. cells were stimulated with PDGF-BB (10 nglml) with or without EMD409915 (Merck, Darmstadt, Germany) at increasing concentrations (10-10M -10-6M). Cell proliferation was measured as DNA synthesis by BrdU-incorporation. Mitogen-activated protein kinase (ERK112, p38, SAPKlJNK and Akt) phosphorylation was evaluated by phospho-MAPK-specific Western blotting of cell lysates after 10 min of stimulation. Results: Stimulation of human HSC cells with PDGF-BB at 10 nglml in the absence of other growth factors resulted in pronounced stimulation of cell migration. PDGF-stimulated migration of human HSC was inhibited dose-dependently between 10-9 and 10-6 M by EMD409915, with complete abrogation of migration at 10-6M. Surprisingly, human skin fibroblast migration was not affected by b3-blockage. There was no effect of p3 integrin inhibition on cell proliferation as measured by BrdU-incorporation, neither in human HSC-nor in HFF cells. Pre-incubation of HSC cells with the p3 integrin inhibitor at 10-6M did not affect the activation of p38 MAPK, p44l42 MAPK or SAPKlJNK. Conclusions: 1. PDGF-BBinduced migration is strongly b3integrin dependent in human HSC, but not in human skin fibroblasts. 2. In contrast to cell migration, HSC and HFF proliferation is p3-integrin-independent. 3. p38, p44l42, SAPKlJNK and AKT mitogen-activated protein kinase signalling pathways are not modified by p3-integrin inhibition. 4. p3-integrin antagonist may be an adjuvant approach to treat hepatic fibrosis. Epithelial to mesenchymal transition (EMT) is defined as a process, in which epithelial cells loose their epithelial characteristics and acquire typical characteristics of fibroblasts. Epithelial cells are tightly attached to their neighboring cells via cell adhesion molecules and they adhere with their basal side to their underlying basement membrane matrices, whereas the apical side faces a lumen. In contrast to immobile epithelium, mesenchymal fibroblasts are specifically designed to invade extracellular matrix (ECM). This is reflected by their prominent mesenchymal cytoskeleton and by their lack of a typical polarity. In the adult organism, EMT occurs in epithelia in response to injury, potentially as a means to replenish fibroblasts, which are required for repair of injury. In organ fibrosis however, enhanced conversion of epithelium into fibroblasts is considered to contribute to progression of disease, as parenchymal epithelial cells acquire phenotypic and functional properties of activated fibroblasts. Recent studies provided increasing evidence that parenchymal epithelium can potentially contribute to activated fibroblasts by EMT, as in mouse models of kidney fibrosis 36% of activated fibroblasts were found to be of tubular epithelium origin. In liver fibrosis, stellate cells are considered the principal source of activated fibroblasts, whereas the role of hepatocytes is considered minor, even though they constitute for more than 80% of the liver mass. Here we provide evidence that the pro-fibrotic growth factors TGF-beta1 and EGF induce expression of fibroblast-markers fibroblast specific protein-1 (FSPl), alpha-smooth muscle actin (alpha-SMA) and type I collagen in primary mouse hepatocytes in vitro. Furthermore, TGF-beta1 and EGF induce acquisition of fibroblast characteristics, migratory capacity and release of MMP-2, suggesting EMT of hepatocytes in vitro. Additionally, progression of liver fibrosis in a mouse model of tetracarbon chloride-induced liver disease is associated with appearance of FSPl positive hepatocytes, indicating EMT, which suggests a role of EMT in liver fibrosis. In conclusion, our results for the first time provide evidence for an active role of hepatocytes in liver fibrosis by undergoing EMT. Human Studies -Serum MCP-1 levels were significantly elevated in both CFLD (929t67 pglml; p30 years) Genotype frequency p=0.0017, Allele frequency P=0.0007, odds ratio=1.8. (Table 1) . Possession of the Val allele is associated with slow rate of disease progression in HCV. The functionally less active forms of CETP may result in decreased liver steatosis as a result of altered composition or increased levels of HDL, and therefore less fibrosis when a second insult such as infection with HCV is sustained. BackgroudslAims: Hepatitis C virus (HCV) is a major cause of chronic liver disease worldwide. However, little is known about exact mechanism of fibrogenesis by specific HCV gene or protein because the study of HCV-induced liver fibrosis has been mainly studied in human or animal model. Moreover, the dynamically molecular study of HCV role in relation to liver fibrogenesis or immune response has been hampered due to lack of efficient in vitro HCV culture system. In the present study, we investigated whether HCV core protein directly influence on the liver fibrogenesis through stimulation of hepatic stellate cell(HSC) in in vitro or not. Methods Human and rat HSCs were isolated using collagenase perfusion system and density gradient method, and other human HSC line (L190) was purchased from ATCC. We established co-culture system of primary HSC and HepG2-core stable cell line which HCV-core gene was transfected into HepG2 cell. Co-culture system was divided into two way, mixed co-culture and separated co-culture system. Immunocytochemical staining was performed to identify the cytokines such as transfoming growth factor pl (TGF-P1) and a-smooth muscle actin(a-SMA) overexpressed from HSC in the liver fibrogenesis. a-SMA, TGFpl, transforming growth factorp receptor I1 (TGFORII), collagen type I were also quantitated by Western blot analysis. The expression of MMP-2 and collagen type I in the culture media was measured and analyzed by each zymogram and ELISA. Results The expression of TGF-p1 and a-SMA was significantly higher in mixed co-culture of HepG2-core plus HSC than in HepG2 plus HSC as negative control. Also the markers related fibrosis such as a-SMA, TGF-pl, collagen type I and TGFRII, MMP-2 and collagen type I were highly expressed in separated co-culture of HepG2core and HSC Conclusions In conclusion, HCV core protein may play a direct role in the fibrogenesis via upregulation of a-SMA, TGF-pl, collagen type I and TGFRII, MMP-2 and collagen type I. Further study is needed to clarify exact signal transduction of fibrogenesis by HCV core protein in the co-culture system. Background: To current knowledge, transforming growth factor beta (TGF-beta) signaling is mandatory to produce liver fibrosis. Various molecular interventions designed to affect the TGF-beta system were successfully used to inhibit fibrogenesis. Activated hepatic stellate cells have been considered as a major producer of extracellular matrix proteins in liver injury and fibrosis. In the present study, we wondered whether follistatin, activin inhibitory protein, reduce apoptosis and prevent liver fibrosis and whether activin plays a key role in liver fibrogenesis. Methods: Wistar male rats weighing around 380g were injected intraperitoneally with dimethylnitrosamine (DMN) (lOpg/g body weight) three times a week for three weeks. Either follistatin or saline were also injected intravenously three times a week. On the 22nd day, blood was collected and biochemical parameters were measured using the standard methods. The liver was either fixed with 4% bufferedparaformaldehyde for histological examination, or frozen immediately in liquid nitrogen for the RNA analysis. Tissue sections were either stained with Hematoxyline-eosin, Masson-trichrome, or subjected to immunohistostaining using antibodies against collagen type IV, alpha-smooth muscle actin (SMA), fibronectin, TGF-beta. The mRNA expression of activin, TGF-beta, TIMP were measured by RT-PCR. Apoptosis was analyzed by TUNEL methods. Wistar male rats were injected with DMN (10kglg body weight) once and liver tissues were obtained and fixed with 4% buffered-paraformaldehyde 0, 12, 24, 36, 48, 60, 72 hours and 7 days after injection for immunohistochemical staining. Hepatocytes and hepatic stellate cells were isolated by collagenase perfusion method and by collagenase-pronase perfusion method and analyzed activin and TGF-beta mRNA expression by RT-PCR. Results: 50% of control rats died, whereas none of follistatintreated rats died within 22 days. The serum level of hyaluronic acid in follistatin-treated rats were significantly reduced. AST and ALP levels were also decreased significantly. Apoptosis was reduced by follistatin dose-dependently. The expression of TGFbeta, TIMP, Collagen lV and alpha-SMA expression were also decreased in follistatin treated rats. We then examined activin expression in liver after DMN treatment. Activin expression was observed at the maximum level in hepatocytes 12 hours after DMN treatment. TGF-beta expression was not detectable 12 hours after DMN administration, and it was sfxikingly increased in stellate cells 48 hours after DMN administration but it was not detectable in hepatocytes. Activin appeared in hepatic stellate cells as well as in hepatocytes 72 hours after Dh4N administration. Inducible nitric oxide synthase (iNOS) has been reported to play pivotal roles in the development of various types of liver injury and iNOS expression may be important in the process of fibrogenesis of nonalcoholic steatohepatitis in obesity. NO-induced active substance was shown to activate matrix metalloproteinase. The iNOS mRNA and protein activity have been found to be induce in rats on a high-fat diet. Objectives We investigated whether 1) induction of NOS occurs in liver of mice fed high-fat diet, and 2) NOS increases matrix metalloproteinase activity and reduces collagen content, thus attenuates liver fibrosis. We compared fatty and fibrotic changes of hepatic tissues in iNOS-knockout (iNOS-'-) and wild-type (iNOS+/+) mice which were fed with high-fat diet for 12 weeks starting from 8 weeks old. Marked induction of iNOS mRNA occurred in iNOS+/' mice, but not in iNOS-/-mice. Immunohistochemically, nitrotyrosine staining which is a footprint of NO-induced active substance, showed positive in iNOS+'+ mice, and negative in iNOS-'-mice. In histopathologically showed fatty metamorphosis, but did not have any distinction between groups iNOS+'+ and iNOS-/-. The extracellular collagen content with Azan staining in iNOS+/+ mice was markedly decreased compared with that in iNOS-'-mice. In gelatin zymography of matrix metalloproteinases we found that proMMP-2 and proMMP-9 were increased both in iNOS+'+ and iNOS-'-mice, but their active form was found only in iNOS+'+ mice. Similar results were obtained from in situ zymography of hepatic tissue. The stronger gelatinolytic activity was found diffusely in hepatic tissues of iNOS+/+ mice than iNOS-/-mice. (129xC57BL16) were subjected to either sham operation or bile-duct ligation. Animals were sacrified two weeks after sur-gery and blood and liver samples obtained. Bile duct ligationinduced elevation of serum liver enzymes was similar between WT and ATla (-/-) mice. However, the liverlbody weight ratio was greater in WT than in ATla (-/-) mice. Bile duct ligated WT mice showed severe septa1 fibrosis, as assessed by Trichromic Masson and Sirius Red staining. In contrast, ATla (-/-) mice showed minor fibrotic lesions, which were mainly located in peribiliary areas. Collagen accumulation, as assessed by morphometric analysis of Sirius red staining and hepatic hydroxyproline content, was markedly lower in ATla (-/-) mice compared to WT mice. Positive Sirius red stained area in bile duct ligated WT and ATla (-/-) mice were 8.150.9 and 4.651.0%, respectively (p<0.05). The increase in hepatic concentration of bioactive TGFb and proinflammatory cytokines (TNFa and ILlb) was attenuated in ATla (-/-) mice compared to WT mice, as assessed by ELISA. Moreover, immunohistochemistry analysis revealed decreased lipid peroxidation products as well as decreased phosphorylation of c-Jun and p42-44 MAPk in ATla (-/-) mice compared to AT1 (+/+) mice. Chronic liver failure stimulates the onset of interstitial liver fibrosis, which eventually results in "capillarization" of the sinusoid, impeding clearance of toxic substances by hepatocyte cells. The goal of this work was to search for the therapeutical effect of adjuvant gene therapy using Adenoviral vectors containing cD-NAs for human urokinase Plasminogen Activator and human Matrix Metalloproteinase-8 (Ad-huPA plus Ad MMP-8) on cirrhosis and its relationship with manganese brain accumulation, striatum dopamine and its metabolites in the rat. Mn+2 is a wellknown neurotoxic metal which has been found accumulated in brain and blood of cirrhotic patients with hepatic encephalopathy. Mnt2 elimination takes place via the hepatobiliary route. METHODS 200 gr Wistar rats underwent bile-duct ligation (BDL) for 4 weeks and concomitantly treated with 1 mglml of MnCl2 in drinking water (BDL/Md2). After this point, five animals were sacrificed and serum, liver and brain tissue (striatum) were obtained. Of the remaining BDL/Mn+2-cirrhotic animals (n=10), 5 were injected with Ad-huPA plus Ad-MMP-8 (3x10" + 1.5~10~' vp/kg respectively) and 5 rats injected with (4.5~10" vp/kg of Ad-P-Gal). This treatment lasted 10 days. Then, biological samples were recollected as before. An additional experimental model represented by cirrhotic rats injured chronically for 7 weeks with CClb were also monitored for their response to this therapy. RESULTS Seven wk-CC4-cirrhotic animals treated with Ad-huPA plus Ad-MMP-8 (total 4.5x101'vp/kg) were monitored at 2,4,6,8, 10 and 12 days after combined gene therapy treatment (n=18). These animals showed improvement in liver fibrosis of up to 55% as compared with their Ad-P-Gal (n=18) treated cirrhotic counterparts. These results correlated with hydroxyproline detemiinations. Furthermore, survival was determinated in 28 additional cirrhotic animals. 14 cirrhotic rats treated with Ad-huPA plus Ad-MMP-8 had a significantly higher probability of survival at 60 days after beginning of treatment as compared with 14 Ad-@-Gal cirrhotic rats. BDL/Mn+2 injured rats displayed tremors, rigidity, and gait abnormalities. Ten days after treatment with combined therapeutic gene therapy, these symptoms decreased. Also, liver fibrosis was evidently less (25%) as compared with Ad-P-Gal treatment rats. Brain tissue (striatum) was recollected 10 days after BDL/Mnf2 animals were injected with either Ad-huPA plus Ad-MMP-8 or Ad-p-Gal alone. Dopamine (3.04 mglgr) was decreased in 30% in Ad-P-Gal treated animals, as compared with 4.79 mglgr of Dopamine found in Ad-huPA plus Ad-MMP-%treated animals. Dyhydroxyphenylacetic acid (DOPAC a main dopamine metabolite) was as high as in Ad-huPA plus Ad-MMP-%treated animals (13.56 mglgr striatum) indicating a higher dopamine turnover in nontherapeutically treated cirrhotic animals. Moreover, animals treated with Ad-huPA+Ad-MMP-8 showed a 50% decrease in ascitis and gastric varices as compared with Ad-P-Gal-treated animals. (Gastroenterology 1221924,2002) . While PPAR-a was reported to be involved in induction of antioxidant enzymes expression and activities (Life Sci 63:135,1998). Furthermore, in respect of hepatic fibrosis, oxidative stress induces hepatic fibrosis and many reports have demonstrated that several antioxidants can inhibit hepatic fibrosis. Therefore, in the present study, we examined whether PPAR-a ligands, i.e. Wy-14,643 and fenofibrates, can suppress hepatic fibrosis by attenuating oxidative stress in experimental rat model and possess a possibility of therapeutic candidate for hepatic fibrosis. Methods: Fibrosis was induced in male Wistar rats by intraperitoneal administration of thioacetamide (TAA) (200 mglKg twice weekly for 6 weeks). In the treated groups, PPAR-a ligands, Wy-14,643 (Wy) and fenofibrates, were fed a diet containing 0.1% (wlw), and PPAR-y ligands, pioglitazone (PIO), were fed containing 0.01% (wlw) all through the experiment. Control cirrhotic rats received saline injections for 6 weeks. After killing all rats, histological examinations (HE, Azan staining, immunohistochemistry of a-SMA), serum value of ALT and hyaluronic acid, mRNA expression of PPARs, acyl-CoA oxidase (ACO), a1 (I) procollagen, and antioxidant enzymes, such as superoxide dismutase (SOD) and catalase. We also determined lipid peroxidation (LPO), glutathione levels, and activities of SOD and catalase in the perfused liver. Results: Semi-quantitative analyses of fibrotic area revealed that Wy co-administration with TAA reduced to only 19% of the area of TAA-treated rats. There was no significant difference in Azan staining of the liver between TAAtreated rats and TAA-treated rats. Wy administration could not lower serum ALT value in chronic and acute TAA injury. These observations suggest that Wy fails to modulate the hepatotoxicity of TAA. An increased expression of PPAR-CY in TAAtreated rats were observed, but the expression of PPAR-a was abolished in TAA-and TAA-treated rats. Wy intensively enhanced mRNA levels of ACO which was regulated by PPAR-a, increase nearly 50-fold than controls, but the expression of ACO diminished in TAA-and TAA-PPAR-A ACTIVATORS, WY-14,643 AND FENOFIBRATES, treated rats as well as PPAR-a. The mRNA levels of a1 (I) procollagen and TGF-/3l were strongly increased in TAA-and TAA-administrated rats than in controls, while they were dramatically suppressed in TAA-treated rats. The catalase mRNA was reduced to 15% of the controls in TAA-and TAA-treated rats, however, TAAtreatment prevented this decrease to 70% of the control levels. Catalase activity increased 2-fold in TAAtreated rats than in controls, while it decreased 60% of the controls in TAA-and TAA-treated rats. In TAA-and TAA-treated rats an increase of LPO was observed, but not in TA-Atreated rats. We c o n h e d that fenofibrates treatment could reduce hepatic fibrosis to approximately 16% of TAA treatment for 6 weeks by semi-quantitative analysis of fibrotic area. Conclusion: Our data indicate that WAR-a activators, not PPAR-y, can markedly inhibit hepatic fibrosis in the experimental TAA-induced rat cirrhotic model. We suggest that PPAR-a and catalase are important in the development of hepatic fibrosis. Therefore, we conclude that suppression of hepatic fibrosis by PPAR-a activators is probably due to their antioxidant effects via increased activitiy of catalase which reduces hydrogen peroxide, and that fibrates, such as fenofibrates and bezafibrates, might be new candidates for the treatment of hepatic fibrosis. Fibrosis is a common end-point in clinical trials of chronic hepatitis C. Liver biopsy remains the gold standard for fibrosis evaluation. However, variability in fibrosis distribution within the liver (sampling variability) is a potential limit to liver biopsy. In order to assess the influence of sampling variability on the accuracy of liver fibrosis assessment with biopsy, we measured liver fibrosis on virtual biopsies of increasing length reconstituted from digitalized images of large liver sections using two different methods of fibrosis assessment. Method Large sections (3cmZ) were performed from 17 surgical liver samples with chronic hepatitis C and various degree of fibrosis. Measurement of fibrosis on the whole section was considered as the reference value. From the digitalized image of the whole section, virtual biopsies of increasing length (2.5 to 200mm) were reconstituted. Fibrosis was assessed independently on each individual virtual biopsy using both image analysis and META-VIR score. Results were compared to the reference value. Results: A total of 10.659 virtual biopsies were studied. Using image analysis, a strong dispersion of area of fibrosis was observed for virtual biopsies smaller than 4Omm. Only 22% of 1,5 mm length virtual biopsies had a measured of area of fibrosis equal to reference value 2 10%. This percentage increased progressively with increasing size (30%, 50%, 69% for biopsies of 25mm, 4Omm, lOOmm length, respectively). Using METAVIR scoring system, 65% of biopsies of 1,5 cm length were correctly categorized according to the score assessed on the whole large section. It increases to 75% for a 2,5 cm size without any substantial benefit for longer biopsies. A same trend was observed whatever the stage of fibrosis. Conclusion: sampling Variability of fibrosis is a significant limit for fibrosis assessment with liver biopsy. This study suggests that a length of at least 25mm is necessary to valuably evaluate fibrosis with semiquantitative score. Sampling variability become a major limitation for more accurate method such as automated image analysis. (Msi-1) , a RNA-binding protein, is highly observed in developing central nervous system and thought to be a mammalian neural stemlprogenitor cell marker. We have reported that cells positive for Musashi-1 (Msi-1) protein appeared during spontaneous resolution of rat liver cirrhosis and that some Msi-1-postive cells were also positive for matrix metalloproteinase (MMP)-13, possibly implicating active participation of stem cells in the resolution of collagen. Though Msi-1 antigen is expected to be expressed during very early stage of differentiation in the cellular lineage of liver, the type of cells appeared in the damaged liver and expressed Msi-1 remain to be elucidated. The present study is designed to characterize the Msi-1-positive cells in cirrhotic liver from the aspect of differentiation, significance and expected function of stemlprogenitor cells in the liver, especially fibrolytic function. Methods: Rat liver fibrosislcirrhosis was established by intraperioneal injection of CC14 twice a week. Liver samples were obtained 2, 5 and 7 days after the last injection of 12-week CC14 intoxication. Liver tissue samples were immunohistochemically stained for Msi-1, c-kit, nestin, a-smooth muscle actin (a-SMA), CK-19 and MMP-13. Gene expression of Msi-1 was also observed by reverse-transcription polymerase chain reaction (RT-PCR). Results: Neural stemlprogenitor cells, as defined by their expression of Msi-1 andlor nestin, were not observed either in normal rat liver or after 4 weeks of C C 4 administration. RT-PCR analysis revealed very slight signals of Msi-1 mRNA at 2 days after discontinuance of 8-week CCld treatment. On the contrary, remarkable increase of Msi-1 mRNA was observed at 2 days after the 12 weeks CCl4 treatment, then the signals decreased gradually. Immunohistochemical analysis showed that a considerable number of cells positive for Msi-1 appeared in the early recovery stage from advanced liver fibrosis, especially at day 5 after the last injection of 12 weeks CCl4 treatment. Expression of Msi-1 in liver tissue was proceeded by and partly overlapped with that of c-kit, a marker of hematopoietic stem cells. Some Msi-1-positive cells observed around perivenular regions were as very small as oval cells, while the size of Msi-1-positive cells present along the resolving fibrous bands varied and some of them expressed MMP-13. Some ductal cells were positive for Msi-1 but not for CK-19 examined in serial sections. Msi-1 positive cells did not express hepatocytes markers such as albumin or AFP. Cells positive for a-SMA and Msi-1 were observed in some portions, but most of Msi-1 expressing cells did not show positive staining for a-SMA or desmin. At day 7, a few Msi-1-positive cells were observed around the remaining fibrous bands. S-ADENOSYL-L-METHIONINE REPRESSES THE fibrosis, constitute a good model for studying the mechanisms responsible for antifibrotic therapy. To test whether administration of S-adenosyl-L-methionine, a precursor of glutathione and an agent shown to prevent liver toxicity in a variety of settings, could repress the activity of the mouse pro-alpha 2(I) collagen gene in hepatic stellate cells, homozygous transgenic mice harboring the -17 kb to +54 bp of the proximal promoter of the mouse alpha 2(I) collagen gene cloned upstream of the Escherichia coli beta-galactosidase reporter gene (LacZ) were used. Chronic liver injury was induced by injecting intraperitoneally 5 mllkg of body weight of CC4 (25% vlv in mineral oil) three times per week for 4 weeks. S-adenosyl-L-methionine was administered intraperitoneally at a dose of 10 mglkg of body weight every day for 4 weeks. Control groups were given either mineral oil or S-adenosyl-Lmethionine alone. Hepatocellular damage and protection by Sadenosyl-l-methionine was confirmed by measuring serum levels of transaminases; S-adenosyl-L-methionine lowered ALT levels in the CCL4-treated mice from 143 to 28 UlL and AST from 171 to 66 UlL (control values in the absence of CC14 were 26 and 58 UlL for ALT and AST, respectively). Hematoxylin and eosin staining in the CC14-treated mice revealed the presence of Mallory bodies, lymphocyte infiltration, centrilobular steatosis, and perivenular and pericellular fibrosis, and S-adenosyl-L-methionine minimized the pathology score. Masson's trichrome staining showed less collagen deposition in mice treated with CC14 plus S-adenosyl-Lmethionine than in the CC4-treated mice. Histochemical analysis using X-gal staining allowed for the precise identification of the cell type in which the beta-galactosidase gene was active as driven by the pro alpha 2(I) collagen promoter. Results indicated activation of the pro alpha 2(I) collagen promoter in mice treated with CCl, and repression of such activation in mice co-treated with S-adenosyl-L-methionine. Immunofluorescence analysis of mice injected with CC14 revealed colocaliiation of alpha-smooth muscle and beta-galactosidase positive cells, suggesting that the activation of the promoter occurred only in hepatic stellate cells. These results suggest that one mechanism by which administration of S-adenosyl-L-methionine, a precursor of glutathione synthesis, could ameliorate liver fibrosis is by decreasing the responsiveness of the promoter of the alpha 2(I) collagen gene to a profibrogenic stimuli such as CCb. These transgenic mice should prove to be useful in further studies on how S-adenosyl-L-methionine exerts this repression of the alpha 2(I) collagen gene and perhaps to studies with other liver toxins such as alcohol. Background Endothelin-1 (ET-I), the most powerful constrictor of the liver vasculature and stimulator of the synthesis, by Kupffer cells, of a potent systemic vasodilator, platelet-activating factor, is also implicated in the fibrosis of lung, kidney and liver. Thus increased hepatic concentrations of ET-1 and its receptors in human and experimental cirrhosis suggest its major role in the pathology of chronic liver disease and its complications (fibrosis, portal hypertension and systemic hypotension). We investigated whether ET-1 receptor antagonism, after the development of fibrosis and cirrhosis, arrestslreverses the progression of chronic liver disease. Methods: Chronic liver injury was induced in rats by CC14 treatment ( of the development of cirrhosis in Group 1 (histopathology score of 3.7 ? 0.41 vs 4.8 2 0.16, p< 0.01) and reversal of cirrhosis in Group 2 (histopathology score of 3.6 2 0.41 vs 5 2 0, p 65%). Patients were analyzed for a) gender, age, estimated length of infection and BMI at the time of biopsy, b) genotype, current ethanol use, HCV RNA within 1 year of biopsy, c) ALT within 30 days of biopsy, and d) histologic activity and fibrosis found on biopsy, using univariable and multivariable analysis. Results: Steatosis was found in 79/185(43%) of liver biopsies; 62 (33%) had grade 1; 13 (7%) grade 2, and 4 (2%) grade 3. Genotype distribution was as follows: 114 genotype 1 (62%), 44 genotype 2 (24%), and 27 genotype 3 (14%). Steatosis was found in 39% (441114) of genotype 1,48% (21/44) of genotype 2, and 52% (14/27) of genotype 3 patients (p = 0.34). Grade 2 or 3 was found in 6% (7/114) of genotype 1, 11% (5/44) of genotype 2, and 19% (5/27) of genotype 3 patients (p = 0.09). There was no significant association found between steatosis and gender (p = 0.93); age at biopsy date (p = 0.09); ALT (p = 0.58); histologic activity (p = 0.27), viral load (p = 0.27) and estimated length of infection (p = 0.22). There was a significant association of steatosis with BMI (< 0.01), fibrosis score (p < 0.01) and current ethanol use (p = 0.03). BMI 2 30 occurred in 52% (42/81) of genotype 1, 39% (11/28) of genotype 2, and 27% (6/22) of genotype 3 patients (p = 0.10). After controlling for BMI, there was no significant relationship found between steatosis and genotype. For BMI < 30, 23% (9/39) of genotype 1, 47% (8/17) of genotype 2, and 50%(8/16) of genotype 3 patients had steatosis (p = 0.08). For BMI 2 30,60% (25142) of genotype 1,55% (6111) of genotype 2, and 67% (4/6) of genotype 3 patients had steatosis (p = 1.00). Multivariable analysis was performed on 5 variables (p 0.25 in univariable analysis) and genotype. These included age (categories < 30,30-40,40-50, SO+), BMI (< 30,30+), current ethanol use (yes, no), Ishak fibrosis score (0-1, 2 21, and estimated length of infection (5 10, 11-15, 16-25, 25+ years) . A significant association with steatosis was found in 3 variables: Ishak fibrosis score 2 2 (OR 4.2, 95% CI 1.5-11.6, p = 0.005), BMI -> 30 (OR 3.7, 95% CI 1.6-8.8, p = 0.003), and current ethanol use (OR 2.5, 95% CI 1.0-5.8, p = 0.04) after adjustment for HCV infection length, which itself was not statistically significant (p = 0.82). Conclusion: Contrary to a number of previously reported studies, this population-based study did not find an association between steatosis and genotype, including genotype 3 in chronic HCV infection after multivariable analysis. We did find that Ishak fi-brosis score, BMI and current ethanol use were associated with steatosis after adjustment for HCV infection length, but age and histologic activity were not associated. infection results in necroinflammatory liver disease characterized by the insidious progression of hepatic fibrosis and loss of functioning hepatocyte mass. A cell-mediated immune response with prominent lymphocytic infiltration of liver is thought to play a major role in pathogenesis, although little is known about the molecular mechanism underlying the liver injury associated with this viral infection. On the other hand, non-immune mechanisms have also been proposed as another mechanistic candidate for such sophisticated processes. The recent intensive interest is that the expression of HCV proteins seemingly alters lipid metabolism and transport in the liver in association with the reactive oxygen species (ROS) mediated carcinogenesis, although the molecular basis of underlying mechanisms as well as responsible elements of HCV gene for this is yet to be established. Thus, the aim of this study was to clarify whether HCV core protein expression directly alters lipid metabolism-relating enzymes to (cause steatosis, especially focusing on nuclear receptors and ATP binding cassette transporter proteins expression. Methods: 1. The complementary DNA clone of full length HCV core (aa 1-191) was derived from serum of HCV I b patient by reverse transcription and nested polymerase chain reaction, and HCV-core expression plasmid (pCAG-HCVcore) was prepared by a standard procedure. Control plasmid was also prepared with P-galactosidase (pCAG-LacZ). 2. HepG2 cells were transfected with pCAG-HCVcore or pCAG-LacZ, thereafter harvested for enzymatic triglycerides (TG) assay, HCV core antigen rneasurement by ELISA at 24 h or 48 h. Also, lipid metabolism-related enzymes mRNA expression was estimated by semi-quantified RT-PCR; mitochondrial and peroxisomal fatty acids ,&oxidation, o-oxidation, ABC transporters (Mdr3, Mrp2, Bsep), microsomal TG transfer protein (MTP), LDL receptor, and nuclear receptors. Resultss: 1. HCV core expression was evidenced at 24 and 48 h at the similar level. 2. Hepatic TG was increased in HCV core expressing cells, along with down-regulation of carnitine palmitoyl transferase 1 A(CPTlA), a precious protein in liver mitochondrial fatty acid@-oxidation (50% at 24 h, 85% at 48) and up-regulation of acyl-CoA oxidase 1 (ACOl), a precious protein in peroxisomal fatty acidso-oxidation (130% at 48 h). Cyp4A11, a precious protein in microsomal o-oxidation, and MTP, a precious protein in VLDL assembly, were not detected or unchanged. Superfamily 2 (SF2) helicases and helicase-like proteins share 6 conserved motifs. Alignments reveal that several additional conserved motifs are present in the SF2 helicase encoded by the hepatitis C virus (HCV). The roles of two such motifs are examined here using structure-based site-directed mutagenesis. The first motif (YRGxDV) forms a loop that connects SF2 helicase motifs 4 and 5, at the tip of which is Arg393. When Arg393 is changed to Ala, the resulting protein retains a nucleic acid stimulated ATPase but cannot unwind RNA, unwinds DNA poorly, and does not translocate on ssDNA. DNA and RNA stimulate ATP hydrolysis catalyzed by R393A like the wildtype but the mutant protein binds DNA more weakly than wildtype both in the presence and absence of a non-hydrolysable nucleotide analogue. Thus, this "Arg-clamp" motif anchors the protein on nucleic acid enabling processive unwinding. The second motif (DFSLDPTF) forms a beta-loop between SF2 motifs 5 and 6 that connects domains 2 and 3. When F444 in this "beta-arm" is changed to Ala, the resulting protein is devoid of all activities. When F438 is changed to Ala, the protein retains nucleic acid stimulated AT-Pase, but unwinds DNA and RNA poorly. In this case, uncoupling of ATP hydrolysis and unwinding is due to the fact that the F438A mutant does not release DNA upon ATP binding like the wildtype. The F438A mutant also has a lower melting temperature than the wildtype indicating the hydrophobic pocket formed by the beta-arm and residues in domain 3 stabilizes the protein. Data support an inchworm model for helicase action and identify two new potential sites for rational HCV drug design. This work was supported by the AASLD Liver Scholar Award from the American Liver Foundation. Tat is an early gene product of HIV-1 that is essential for replication and viral gene expression. This transactivating protein is secreted by HIV-infected cells and taken up by neighboring cells. Tat modulates expression of specific cellular genes and may be a key player in the interactions between HIV and other infections. One of the most common coinfections observed among HIV patients is hepatitis C virus (HCV). If HIV Tat has an effect on HCV replication, this could help explain the rapid development of liver disease and cancer among HIV patients. AIM This study was designed to test the hypothesis that Tat alters HCV replication. METHODS: Soluble Tat was added to the media of HUH-7 cells containing a HCV replicon. Replicon replication was quantitated using a ribonuclease protection assay. Using an LTR-luciferase plasmid to transfect HeLa cells, we developed a series of controls outlining the relative oxidative state of the Tat protein. RESULTS Exposure to Tat protein in the reduced state substantially increased HCV replicon replication. Oxidized Tat was found to have no significant effect on the replication of the replicon. The increase was dependant on length of exposure to Tat and the concentration of Tat. PRESENT WORK: Further assays seek to define the nature of the effect by Tat on HCV replication. CONCLUSION: Tat may upregulate HCV replication directly or indirectly, and thereby play a role in modulation of HCV infection and pathogenesis. Elucidating the complex interactions between HIV and HCV will be critical in evaluating and developing workable treatments for the increasing number of coinfected HIV/HCV persons with liver disease. ACKNOWLEDGEMENTS: This research was supported by NIH grants AI34764, AI54626, AI54238, CA54576 and CA89121. We have found previously that expression of hepatitis C virus proteins in cell lines or in the liver of transgenic mice inhibits interferon alpha induced signaling through the Jak-STAT pathway (Heim et al., J Virol, 1999, 73:8469 and Blindenbacher et al., Gastroenterology, 2003, 1241465) . The aim of the present study was to investigate if the same inhibition takes place in livers cells of patients with chronic hepatitis C. From February 2001 to April 2002, all patients with chronic hepatitis C referred to the outpatient liver clinic of the University Hospital Basel were asked for their permission to use part of the liver biopsy for this study. For non-HCV controls, patients that underwent ultrasoundguided liver biopsies of focal lesions (mostly metastasis of carcinomas) were asked for their permission to obtain a biopsy from the normal liver tissue outside the focal lesion. The protocol was approved by the ethical commission of Basel. Written informed consent was obtained from all patients that agreed to participate in the study. After removal of a 20 to 25mm long biopsy specimen for routine histopathological workup for grading and staging of the liver disease, the remaining 5 to 20 mm long biopsy cylinders were immediately incubated in PBS or PBS with human interferon alpha (lo00 IUlml) for 10 to 60 minutes at 37°C. The samples were then used for the preparation of cytoplasmic and nuclear extracts and in some cases for immunofluorescence studies. In a first part, the ex vivo stimulation of biopsy tissue was validated. Biopsy samples of patient with different degrees of fibrosis were incubated for 0, 10, 20,30 and 60 minutes, and the nuclear translocation (a surrogate marker for activation) of STATl was visualized by immunofluorescence. We found a transient activation of STAT1, with a peak after 20 minutes of stimulation with interferon alpha. The signal was shut down in all biopsies after 60 minutes. Interferon alpha diffused readily in the biopsy cylinders regardless of the degree of fibrosis. In the second part, 44 consecutive biopsies of patients with chronic hepatitis C and 12 consecutive control biopsies of patients who underwent ultrasound-guided biopsy of focal lesions in otherwise healthy livers (mainly liver metastasis of carcinomas) were used for semiquantitative assessment of interferon alpha induced STATl activation using gel shift assays. We found a signhcant inhibition of STATl DNA binding in patients with chronic hepatitis C compared to controls. As in our previous work with HCV protein expression systems, STATl phosphorylation was not impaired in human liver biopsies from patients with hepatitis C. We conclude that interferon induced intracellular signaling can be analyzed semi-quantitatively in human liver biopsies ex vivo by gel shift assays. Using this newly validated method, we found that signaling is impaired in liver biopsies from patients with chronic hepatitis C. The block in STAT signaling is at the level of DNA binding in the nucleus, whereas STAT activation at the interferon receptor functions normally. Methods We studied the intrahepatic and peripheral HCV-specific CDS+ T cell response in a cohort of 15 HLA-A2 positive chronically HCV infected patients by tetramer staining, intracel-M a r IFNgamma staining and CFSE labeling. In addition, viral amino acid sequences corresponding to the four CTL epitopes used in the study were deduced by nucleotide sequence analysis to assess the potential role of viral escape variants. Results (1) Substantial higher numbers of HCV specific CD8+ T cell responses were detectable in the liver compared to the peripheral blood, suggesting compartmentalization at the site of infection. In addition, intrahepatic CDS+ T cells were multispecific whereas peripheral HCV specific CDS+ 'I' cell responses were primarily monospecific. These results suggest, e.g., the persistence of HCV-specific T cells at the site of disease or the direct priming of T cells in the liver. (2) Epitope-specific CD8+ T cells that were detectable in peripheral blood as well as liver were characterized by a good peripheral proliferative capacity after peptide specific stimulation suggesting that proliferation is a prerequisite of peripheral virus-specific CD8+ T cells to accumulate in the liver. This is further supported by the observation in one patient that the lack of proliferation of virus-specific CD8+ T cells in the peripheral blood was associated with the absence of the same CDS+ T cell response in the liver. (3) Despite the strong accumulation of HCV specific CD8+ T cells in the liver, the majority of those cells was unable to secrete cytokines after peptide specific stimulation indicating that dysfunction of intrahepatic HCV specific CD8+ ' I cells might contribute to viral persistence. Importantly, IFN gamma producing tetramer positive CD8+ T cells were only detectable in patients with low viral titers or in patients with high viral titers but with sequence variations in the corresponding viral epitope, suggesting viral escape. The relative contribution of T cell dysfunction versus viral escape to viral persistence is not known. It is important to note, however, that both mechanisms can operate within the same patient. Insulin resistance (IR) is a frequent feature in chronic hepatitis C while risk factors of steatosis are body mass index in patients infected with genotype 1 and viral load in those infected with genotype 3. In patients with chronic hepatitis C, we adressed the following issues: 1) is IR the cause or consequence of steatosis and fibrosis ? 2) what are the risk factors of IR ; 3) does IR play a role (and to what extent) in the occurrence of steatosis ? 4) is IR involved in the progression of fibrosis ? Therefore, this study was designed to assess relationships between IR, steatosis and fibrosis according to HCV genotypes in non diabetic patients. Methods: 152 non-diabetic patients with biopsy proven non-cirrhotic chronic hepatitis C had fasting serum glycemia and insulinemia measurements. IR was evaluated by using HOMA model assessement. 4 groups of patients were defined according to HCV genotypes (1 or 3) and degree of steatosis (absent or mild vs moderate to severe). Results: The 4 groups were similar in terms of age, sex ratio, BMI and disease duration. Prevalence of IR (HOMA higher than1.64) was significantly higher in genotype 1 patients with steatosis than that of other patients (77% vs 27,23 and 21% respectively, p=O.oOOl). IR was significantly associated with extensive fibrosis in genotype 1 patients (p=0.008) but not in genotype 3 patients. Among genotype 1 patients, independent parameters associated with IR were age (p=0.002) and steatosis (p=O.O3) but not fibrosis. Independent risk factors for steatosis in genotype 1 and genotype 3 patients were DR (p=O.O4) and viral load (p=0.02) respectively. Steatosis was associated with significantly higher fibrosis score whatever the genotype (p=O.O1). Among genotype 1 patients, the median progression rate of fibrosis was significantly higher in those having steatosis and IR together than in other patients (0.1 vs 0.05, p =0.02). Conclusions : In non-diabetic patients with non-cirrhotic chronic hepatitis C 1) steatosis and fibrosis are associated with IR in genotype 1 patients but not in genotype 3 patients, 2) among genotype 1 patients, IR mainly depends on age but not fibrosis 3) IR is a major risk factor of steatosis in genotype 1 patients, 4) the combination of steatosis and insulin resistance is a risk factor of fibrosis progression. These results suggest that IR is not the consequence but rather the cause of steatosis and fibrosis progression. . However, in the first month of life, in 4 of these 13 newborns, HCV RNA was only detected in PBMCs and not plasma. In 2 neonates, the SSCP band patterns of PBMC-derived viral sequences were different from maternal sequences in serum or PBMC; however, they were identical to HCV RNA negative strand amplified from mothers' PBMC. The latter strongly suggests that the infection was transmitted through maternal infected PBMCs. Another infant harbored different HCV strains in serum and PBMC; both strains were present in the mother's serum. Only 7 of 13 HCV-RNA positive children developed HCV antibody at one year. HIV infection was found in 2 out of 13 HCV RNA-positive and in 6 out the 35 HCV RNA-negative infants (NS). CONCLUSIONS: We have found that 1) HCV strains infecting neonates may be derived from strains residing in maternal PB-MCs. 2) HIV+ pregnant women who were HCV RNA positive in PBMCs were more likely to transmit HCV to their infants. These data suggest that one mechanism for perinatal transmission of HCV is maternal-fetal transfusion of HCV-infected PBMCs late in pregnancy or during labor and delivery. Additionally, HCV infected neonates may have a limited ability to develop HCV antibodies in the first year of life and thus there may be an underestimation of the prevalence of HCV infection by anti-HCV determination among children born to HCV RNA positive mothers. (HCC) in this clinical setting is very high. However, the molecular mechanisms of how HCV related proteins may contribute to HCC is not well defined. HCV core protein, a viral structural protein, has been implicated in tumor development. The goal of this study was to characterize the transcriptional regulation induced by core protein expression with respect to generations of a malignant phenotype. METH0DS:A cDNA fragment encoding HCV core protein of l b genotype was subcloned into the pTRE2hyg vector and co-transfected with pTet-On vector (Clontech) into a human Huh 7 hepatoma derived cell line using polyamine (Mirus) as a carrier. Stably transformed clones were selected by growth in culture medium containing G418 and hygromycin. Clones positive for HCV core protein expression were screened by immunoblotting from cells grown in the presence and absence of tetracycline. One clone with core expression tightly regulated by tetracycline was chosen for further analysis. Cell proliferation was evaluated by a modified MTT assay, at 6,12,24, 36 and 48 hours after induction of core protein expression. Twenty-four hrs after HCV core expression, the cellular transcriptional changes were analyzed by Microarray analysis using human Genome U133 Array Sets (Affymetrix). The microarray data were validated by real-time PCR on selected genes. RESULTS HCV core protein expression was tightly regulated by the presence of tetracycline and the protein levels were dependent on the amount of the tetracycline present in the culture medium. Core protein expressing cells showed significant increase in growth compared to non-expressing cells at 6,12, 24,36 and 48 hours after addition of tetracycline indicating a proliferation stimulus provided by core protein. Microarray analysis using human gene chip U133 revealed that 1072 of 39,000 genes were significantly changed (> 3 fold), with 626 up-and 446 down-regulated. This screen was quite informative since 52 genes involved in regulation of cell proliferation (38 genes) and apoptosis (14 genes) were changed respectively. CONCLUSIONS: Enhanced growth of liver-derived cells by HCV core protein is due to cellular transcriptional changes induced upon core expression; two major molecular pathway(s) are involved 1) those involved in celI growth control and 2) those involved in cell survival. Such gene regulation by HCV core protein is important to the molecular pathogenesis of HCC Daudi-CD4 cells were infected with pNL4-3, a T-tropic HIV molecular clone. 24 hours later, uninfected and HIV-infected cells were incubated with either 10% high titer HCV-positive patient serum, 10% HCV-negative patient serum, or 10% FBS (no human serum). Infections were continued for 3,4,5, and 6 days; cells were harvested, stained with Annexin V and propidium iodide, fixed, and apoptosis was measured by flow cytometry. Annexin V is a marker of early apoptosis, and propidium iodide indicates cell death either by apoptosis or necrosis. RESULTS: Six high titer HCV-positive sera and six HCV-negative sera were tested in five separate experiments, and in each case, the percentage of viable cells was higher in the HIVIHCV-coinfected cells compared to those infected with HIV only. To further elucidate dynamics of the course of coinfection, we did a time course consisting of harvests every eight hours beginning at Day 3 (the earliest point at which apoptosis had been observed in prior experiments) and continuing until Day 6 plus 16 hours. The results of the five harvests from Day 5 plus 8 hours to Day 6 plus 16 hours are shown in the figure below. The percentages of viable cells and those undergoing apoptosis are shown as measured by flow cytometry. During this 32-hour interval the percentage of viable (negative for apoptosis) cells fluctuated very little in untreated cells, or in cells incubated with HCV-positive serum, or even in HN-infected cells incubated with HCV-positive serum. In striking contrast, HIV-infected cells that were incubated with HCV-negative patient serum dropped to only 16% viability by the end of the 32 hours, compared to 64% viability in the HIV-infected cells incubated with HCV-positive patient serum. This effect was not observed when apoptosis was induced by either of two chemical inducers of apoptosis in Daudi cells: peroxisome proliferator-activated receptor gamma ligand and calpain inhibitor 11. Next, we sought to determine if the effect of apoptosis inhibition was dependent on HCV particles. When H Ninfected cells were kept separate in culture from HCV-infected cells by a membrane that prevented virus passage, apoptosis in the HN-infected cells was still inhibited. CONCLUSIONS: Collectively, these results suggest that in our experimental conditions: 1) HCV-positive sera, but not HCV-negative sera provide a protective effect to HIV-induced apoptosis in Daudi-CD4 cells, and 2) a cellular factor induced by the presence of HCV, rather than HCV particles themselves, is responsible for the inhibition. Methods: Serum and PBMC samples from an individual with acute HCV-infection (Genotype l a ) were collected every two to three months over a time period of 20 months beginning 2 months after infection. PCR amplification of vRNA using a set of overlapping primers for the entire HCV genome was performed and PCR-products population sequenced using an ABI automated sequencer. CD8+ T cell responses were defined using an IFNgamma ELISpot assay using an overlapping synthetic peptide set spanning the whole HCV-genome (Genotype la). Based on the HLA-type the complete viral genome was also screened for putative CDS+ epitopes using a web based epitope prediction program (SYFPEITHI). Results: ELISpot screening with the whole HCV-genome peptide set detected only two ex-vivo IFN-gamma responses (NS3 1073-1081 and NS5B 2594-2602), although neither epitope was associated with the development of mutations. However, over the 20 month period of observation we detected a total of 25 mutations throughout the genome (2 in El, 4 in E2,3 in NS2,5 in NS3, 2 in 353A NS4 and 9 in NS5), in addition to multiple changes in the hypervariable regions. Interestingly, only five of these mutations resided within known HLA-restricted optimal epitopes. However, 17/25 mutations (68%) were located within predicted epitopes based on known HLA-binding motifs of the subject. In order to begin to define the rate at which these mutations develop an intermediate time point of 12 months was also sequenced. Interestingly, 19/25 (76%) of the mutations had already occurred by this time. It is likely that at least some of these mutations are the direct result of immune pressure exerted through CD8+ T lymphocytes. We are currently generating peptide-specific CDS+ T cellines against these regions exhibiting viral evolution, providing an opportunity to determine which of these regions are associated with previously undescribed CD8+ responses. Conclusions: Hypothesizing that evolving mutations in HCV might be the result of immune pressure by CTL, longitudinal full length sequencing of HCV may represent a powerful tool to define additional HCV-specific CD8+ T cell responses, especially those exerting substantial selective pressure. Using this approach we were able to identify a number of candidate regions where CD8+ T cell responses may be present and driving viral evolution. Determining the extent to which viral escape from CD8+ responses occurs during HCV infection will elucidate its overall impact in preventing proper control of HCV. indicate that the main open reading frame of HCV contains RNA structures in the corelARFP (alternate reading frame protein) and NSSB (polymerase) genes. We hypothesized that one or more of these structures are cis-acting replication elements (CRE)s. Methods. We used custom software, thermodynamic RNA folding programs, and classical comparative phylogenetic analysis to build secondary structural models. We then used directed mutagenesis in the subgenomic replicon system to seek stem-loop elements in NS5B that are required for viability. The mutations we introduced maintained the sequence of the polymerase protein, i.e., they were silent, synonymous codon substitutions. Structural probing was carried out on replicon RNA. RNase T1 and lead (II), and nuclease V1, were used to identify loops and helices, respectively. Results. Mutations in NS5BSL3.2 blocked replication, indicating that this novel structure is an essential cis-acting replication element (CRE The aims of this study were: 1) to correlate mean ALT level with HCV RNA by PCR positivity or negativity, 2) to compare clinical, demographic and risk factor data as well as viral load and genotype between persons with PNALT, persistently elevated alanine transaminase (PEALT) and fluxuating alanine transaminase (FLUXALT) who had at least 3 years of data available. Methods: For each person in the cohort from whom 2 2 PCR results were available (n=278), we calculated the mean level of all ALTs measured over a 3-year period after enrollment. For a subset of the cohort, we selected persons who met the following criteria: 2 positive PCR tests 2 1 year apart with the first positive PCR occurring prior to date of the first ALT and a minimum of 6 ALT levels measured over the subsequent 3 years with a minimum interval of 1 month between ALT measurements (n=129). We defined a person as having PNALT or PEALT when 2 5 of 6 ALT levels were normal or elevated, respectively, during the 3-year follow-up period. Persons who did not fit into either of the above 2 categories were defined as FLUXALT. We reviewed clinical data via chart review, demographic and risk factor data via interviews, as well as data on viral load (performed by branched DNA assay version 2.0) and genotype (restriction polymorphism analysis Background: HCV infection rarely presents acutely, but when it does it offers a potential early "window" for effective therapy. It has been suggested that such therapy is efficacious due to enhanced activity of T cell responses, which are most active during acute disease. Methods: 8 subjects with acute HCV infection were comprehensively mapped for CD8+ T cell responses with an interferongamma Elispot assay using 301 overlapping peptides, spanning the entire HCV polyprotein. Responses were confirmed by establishing peptide-specific T-cell lines and intracellular cytokine staining. The screening for HCV-specific CD8+ T cell responses was repeated during and after therapy. Responses detected in the screening assay were longitudinally studied before, during and after therapy using single peptide Elispot ,as well as tetramer assays. We also studied CD4 proliferative responses over time in some subjects using recombinant HCV proteins. Results: CD8+ T cell responses were detected in 6/8 subjects before therapy was initiated and were typically multispecific, with up to 5 epitopes targeted. After the start of therapy, frequencies of HCV-specific CD8+ T-cells declined following suppression of HCV viremia. Therapy also did not increase the breadth of the HCV-specific response as no additional specificities were detected at later timepoints, when Elispot screening was repeated. In 5/8 subjects viral relapse occurred after therapy was stopped, with 2 subjects being retreated successfully. Relapse was not predicted by the presence or absence of HCV-specific CD8+ T-cell responses. During viral relapse, a vigorous expansion of pre-existing HCV-specific CD8+ T-cells was observed for most specificities, with single responses reaching almost 3% of CD8+ T cells as measured by tetramer staining. These responses were unsuccessful in containing HCV. In the two subjects who were retreated after viral relapse, we observed the identical pattern of declining CD8+ T cell responses as in the first treatment course. In contrast to CD8+ T cell responses, CD4 proliferative responses usually became more vigorous after virus was suppressed on therapy. However, such responses also did not protect from viral relapse. Conclusions: These data suggest that although multispecific functional CD8+ T cell responses may be present during acute disease, this does not predict a successful outcome of antiviral therapy. Rather than boosting CD8+ T cell responses, therapy and viral suppression appear to lead to their attenuation, even though CD4+ T cell responses may be restored. Figure) in the fibrous septum of portal tract in cirrhotic liver. Significantly less Fltl immunopositivity occurred in ND liver. Hepatocytes were negative for Fltl. Conclusion: The greater DE of genes involved in immune activation, fibrosis, cellular proliferation and cell signalling indicated that these processes are more active in the cirrhotic liver that has developed HCC than the cirrhotic liver without HCC development. PLGFlFltl signalling has an important role in neo-vascular development within organs. The decreased expression of PLGF and its receptor Flt 1 in cirrhosis with HCC implies that vascular proliferative signals although normally active in cirrhosis itself may be decreased in cirrhotic livers with HCC. Prospective analysis of cirrhosis prior to HCC development is needed to indicate whether these findings represent a premalignant phenotype. BACKGROUND: Hepatitis C virus (HCV) infection often leads to chronic liver diseases including liver cirrhosis and hepatocellular carcinoma (HCC). At least 10 HCV proteins have been identified, which serve as viral structural and non-structural proteins required for viral replication and virion formation. Several viral proteins have been implicated in HCV persistence and the development of HCC. We have previously reported that the NS2 protein inhibits gene expression f+om various cellular and viral promoters in liver and non-liver derived cells. Thus, expression of endogenous cellular proteins was significantly reduced in NS2 expressing cells. In this regard, the NS2 protein was recently found to be an inhibitor of the pro-apoptotic CIDE-B protein and therefore may contribute to hepatic oncogenesis. To understand the molecular basis of repressed gene expression, we constructed successive deletion mutants of the NS2 protein and tested their effect on luciferase expression driven by CMV promoter. METH-ODS: The cDNAs encoding the full-length (1-217), N-terminal part (1-91) and internal parts of NS2 were generated by PCR and cloned into the pCR3.1 vector for expression under control of the CMV promoter. A liver-derived Huh-7 cell line was co-transfected with a plasmid encoding the luciferase reporter gene and plasmid encoding various deletion mutants of the NS2 gene. Inhibition of gene expression was measured by luciferase activity assay at day two after transfection. Cell viability was also evaluated as well. RESULTS: Deletion constructs 61-121, 61-111, 71-121 and 71-111 exhibited a significant inhibitory effect on luciferase activity comparable to the fulllength NS2 protein, whereas the deletion constructs 81-111 and 81-121 lost the inhibitory effect. Therefore, the minimal amino acid residues required for the inhibition of gene expression by this viral non-structural protein was mapped to residues 71-111. Interestingly, this region partially overlaps with the binding site of the NS2 to a newly identified CIDE-B pro-apoptotic protein. IL12 plays an essential role in the antagonism of T helper 2 differentiation and in the induction of antiviral host defence. We postulated that genetically determined attenuation of IL12 production could provide a plausible immunological mechanism able to determine outcome of HCV infection and have investigated a recently described polymorphism in the IL12B gene. Methods: We have extracted genomic DNA from whole blood taken from 196 HCV antibody positive patients. 124 patients had chronic HCV with detectable HCV RNA and 72 had spontaneously resolved infection, testing HCV RNA negative on several occasions. Genotyping for a single nucleotide polymorphism (SNP) at position (A16974C) on the ILl2B gene was performed using polymerase chain reaction and restriction digest. Maximal in-vitro IL-12 production by cultured mononuclear cells in response to SAC (Staph Aureus Cowan) stimulation was determined in 24 cases by ELISA. Results: Of the HCV RNA positive cases 81 (65%) were homozygous for the 'a' allele, 39 (31%) heterozygous 'ac'; and 4 (4%) were homozygous 'cc' whereas of the HCV RNA negative cases 36 (50%) were 'aa', 33 (46%) were 'ac'; and 3 (4%) were 'cc'. HCV RNA negative cases were significantly more likely to be heterozygous for lac' than HCV RNA positive cases (p=0.04). Of the 24 patients studied for IL-12 production, 13 were genotype 'aa' and 11 'ac'. 13 were HCV RNA positive (8 genotype lac' and 3 'aa') and 11 HCV RNA negative (3 genotype 'aa' and 3 'ac'). Maximal IL-12 production with SAC stimulation was lower in the 11 genotype 'aa' cases (mean 104 units) than in the 13 genotype 'ac' cases (mean 228 units) (p= 0.08). HCV RNA positive cases produced significantly less IL-12 than did HCV RNA negative cases (mean 84 units compared to 266 units, p= 0.016). Comparison of HCV RNA positive or negative cases only, revealed a trend for higher maximal IL-12 production with genotype 'ac' regardless of HCV outcome. Conclusion: Cases heterozygous for the variant 'c' allele at position 16974 of the IL-12 p40 gene are significantly more likely to be HCV RNA negative than those homozygous for the 'a' allele. Recent data found carriage of the variant 'c' allele to be associated with greater IL-12 production capacity and our data from a small number of cases supports this. Genetically influenced enhancement of IL-12 production appears to be a factor influencing the outcome of HCV infection. We have previously demonstrated that hepatitis C virus (HCV) core protein expression in Huh-7 hepatoma cells increased reactive oxygen species (ROS) derived from mitochondria without inducing apoptosis. The aim of this study was to investigate whether HCV core protein inhibits the ROSassociated apoptosis induced by deoxycholic acid (DCA). Methods:: We measured ROS level and 8-hydroxy-2'-deoxyguanosine (8-OHdG) content, and evaluated apoptosis in the presence or absence of DCA (500 pM), using a human hepatoma-derived cell line with tightly regulated HCV core protein expression under the control of a Tet-OffrM promoter. Cells were incubated with 5pM of chloromethyl2',7'-dichlorodihydrofluorescein diacetate for 30 min for measurement of ROS. Cellular 8-OHdG content was quantified with enzyme-linked immunosorbent assay. Fragmented nuclei were assessed with 4',6-diamidino-2-phenylindiole staining and DNA fragmentation was evaluated by genomic DNA laddering. The degree of apoptosis was quantified with enzyme-linked immunosorbent assay. We also examined whether the general caspase inhibitor (zVAD-FMK) inhibited the ROS-associated apoptosis induced by DCA. In some experiments, cells were incubated with 500wM of ursodeoxycholic acid (UDCA) in addition to DCA. The experiments were repeated 3 or 4 times. Results:: Strong core expression was detected 72 hours after withdrawal of tetracycline from the culture medium. The expression of core protein increased the basal ROS level (1.6 2 0.17-fold, P<0.05) and 8-OHdG content (1.13 i 0.005-f0ld, P<0.05) of Huh-71191-20 cells without inducing apoptosis. DCA stimulation produced a 4.0-fold increase in ROS content in the presence of core expression and a 2.5-fold increase in the absence of core expression. Similarly, the 8-OHdG content was significantly increased by DCA regardless of HCV core expression (1.13-fold, P=0.02 for core expression, 1.15-fold, P=O.O1 for core non-expression). Nevertheless, HCV core protein significantly suppressed the ROS-associated apoptosis induced by DCA (W0.05). Also apoptosis induced by DCA was almost completely inhibited by zVAD-FMK. UDCA significantly decreased the ROS (P<0.05) and the 8-OHdG content (P=O.OOS) in the core-expressing cells and attenuated DCA-induced apoptosis. On-treatment virological response (OTR) was defined as complete loss or greater than 100-fold drop in HCV viremia within 3-6 months of therapy by quantitative or qualitative RT-PCR (Roche Cobas Amplicor v2.0), based on recommended early virological testing for continued therapy. Forty-six patients with at least 3 months of continued antiviral therapy were examined thus far, including 30 with genotype 1 (Cl) and 16 with either genotypes 2 or 3 infection (C2). On-treatment response (OTR) was achieved in 29/46 (63%) patients overall, including 47% among C1 and 94% in C2 groups. Positive and negative control groups included 19 healthy HCV seropositive but nonviremic subjects without history of antiviral therapy ("Recovered") and 23 healthy HCV seronegative volunteers, respectively. HCV-specific CD4 T cell response was examined using recombinant HCV Core, NS3l4, NS5 and control proteins in standard proliferation and IFN-gamma (IFNg) Elispot assay at baseline (pre-treatment) and at 1, 3 andlor 6 months during antiviral therapy. Results: HCV-specific CD4 T cell response was significantly greater in the Recovered compared to the Chronic patients, consistent with its expected role in natural HCV clearance. Among the Chronics, C2 patients (genotype non-1) displayed a baseline T cell response to HCV NS3/4 that was modestly but significantly greater than C1 patients ( The progression of fibrosis in chronic hepatitis C virus (HCV) infection differs among individuals and determines the ultimate prognosis and thus the need for therapy. The molecular mechanisms associated with the progression of fibrosis are poorly understood. Gene expression profiling technologies allow the analysis of gene networks whose expression is associated with specific pathological conditions. We used real-time quantitative RT-PCR assays to compare the mRNA expression of 148 selected genes in liver biopsies of controls (n=10 normal liver samples) and of 57 untreated patients with chronic HCV infection and different stages of fibrosis according to Metavir, i.e. stage FlAl (n=11), F1A2 (n=9), F2A1 (n=10), F2A2 (n=lO), F3A2 ( n = l l ) and F4A2 (n=6). In order to limit the number of PCR experiments, we first studied total mRNA pools which were prepared by mixing amounts of individual liver biopsies mRNA of each group. This pooled sample analysis allowed the selection of genes displaying significant different expression (> 2-fold variation) in comparison to "normal livers". The selected genes were then studied at the individual level and their diagnostic performance was assessed using ROC curves. The most informative genes were used to construct specific gene expression signatures. We identified several genes specifically involved in various stages of fibrosis. The dysregulated genes mainly encoded extracellular matrix proteases, growth factors, cytokineslchemokines, and IFNcu-induced genes. We also observed a statistically significant association between HCV RNA amount (as determined with the same real-time quantitative RT-PCR technology) and expression of several genes, mainly IFN-a-induced genes including STAT1, IFI27, MIX1, OAS2 and GIP2. Understanding the correlates of protective immunity in this setting is an important first step in the development of potential vaccine candidates. Methods: Two recipients of frozen patellar allograft tissue procured from the same donor developed evidence of acute HCV within 6 months of surgery. In retrospect, the donor was confirmed to be HCV antibody negative but HCV RNA positive (genotype la). Both patients were enrolled in a prospective study of T cell immunity requiring whole unit blood draw at baseline, 2, 4, 6 and 12 months. Comprehensive HCV-genomewide analyses were determined by IFN-y ELISPOT responses to 33 overlapping peptide pools that spanned the entire HCV polypeptide (genotype la, 3010 aa, total 750 peptides). Peptide responses were defined as greater than mean plus 3 SD compared to 10 control wells. Results: Patient 1 (51 yo WF) cleared the virus spontaneously within 6 months; when first evaluated, the pt. demonstrated CD4+ T cell responses to 8 of 33 (24%) peptide pools, with effector frequencies as high as 1 in 5,000 (to NS3helicase-7, amino acids 1569-1667). IFN-y CD8+ T cells were detected following stimulation with 6 of 33 (18%) pools (highest frequency to pool NS5B-5, aa 2829-2927). Remarkably, 6 months later, the repertoire of the HCV-specific CD4 t T cell response had expanded further, and patient 1 demonstrated responses to 13 of 33 (39%) pools. In contrast, patient 2 (49 yo WM) demonstrated persistent viremia (4.4 million copieslml, Bayer assay) and lacked CD4+ T cell responses to any peptide pools when first evaluated; only one peptide pool (NS3-helicase-5, aa 1369 -1478) elicited responses in CD8+ T cells. Despite virologic clearance with antiviral treatment (pegylated interferon and ribavirin), ELISPOT screening failed to reveal emergence of new CD4+ or CD8+ T cell responses (6 months later). Conclusions: Comprehensive analyses of HCV-genome-wide CD4+ and CD8+ T cell responses reveal significant differences in patients exposed to the same HCV innoculum and correlate with spontaneous clearance versus chronicity. In HCV infection that resolves spontaneously, the repertoire and strength of the HCV-specific immune response may continue to expand in the first year after infection (and may target more than one-third of the HCV polypeptide). In contrast, the profile of T cell responses remains narrow, weak, and constant in the acute infection that becomes chronic, even after successful antiviral therapy. Alcohol consumption exacerbates liver injury in chronic hepatitis C and enhanced oxidative stress is one possible pathophysiological mechanism. We have previously developed a hepatoma cell line with conditional, stable expression of HCV core protein and constitutive expression of CYP2E1. These cells demonstrate dosedependent ethanol toxicity when core protein is expressed. The AIMS of this study were to determine whether reactive oxygen species (ROS) production and mitochondrial dysfunction are responsible for ethanol-induced cytoxicity. METHODS: Huh-7 cells with conditional expression of core protein and constitutive expression of CYP2E1 (L14 subclone) were exposed to 0.1 micromolar tBOOH and/or ethanol (25 mM) for up to 24 hrs. Cytotoxicity was measured by MTT assay or trypan blue exclusion. ROS production was assayed with the oxidation sensitive fluorescent dye DCFDA. Mitochondria1 membrane potential was analyzed by flow cytometry using the dye JC-1 as a probe. Apoptosis was detected by cellular DNA content analysis with flow cytometry. RESULTS: In the presence of 0.1 pM tBOOH, there was a progressive increase in cell death in Huh-7 cells expressing no heterologous proteins (2 2 2%), CYPZEl only (12 2 l%), core only(l5 2 5%), or core plus CYPZEl (3927 X). Addition of 25 mM ethanol to core/ CYP2E1 expressing cells further increased cell death to 5326%. DNA content analysis and nuclear morphology showed that this type of cell death represented necrosis and not apoptosis. Effects on mitochondrial membrane potential were also examined. Under control conditions, only 250.5% of cells had depolarized mitochondria. Expression of corelCYP2El and exposure to tBOOH depolarized mitochondria in 47211% of cells. Similar to cell death, ethanol (25 mM) addition increased depolarization to 7027% of cells. Compared to control cells, core protein increased ROS by a factor of 1.220.3, and the combination corelCYP2El by 2.150.3. Furthermore, cells expressing corelCYP2El ampIified the effect of exogenous tBOOH. Incubation with tBOOH (0.1 pM) had no effect on ROS content of control cells. It approximately doubled the ROS content of cells expressing either core or CYP2E1 and it increased ROS content of cells expressing both corelCYP2El by 4.120.5 fold. In all cases the increase in ROS was completely blocked by the antioxidant N-acetyl cysteine (20 mM). The antioxidant also completely blocked both mitochondrial depolarization and cytotoxicity. CONCLUSIONS: HCV core protein and CYP2El synergistically enhance ROS production in hepatoma cells, amplify the oxidative stress produced by extracellular ROS, and sensitize cells to mitochondrial depolarization and necrotic cell death caused by alcohol. Since all these effects are blocked by antioxidants, the formation of ROS is likely to be the primary event which subsequently produces mitochondrial dysfunction and cell death. sponse associated with different outcomes of HCV infection may provide important insights into our understanding of HCV pathogenesis. For this purpose, we compared the functional features of HCV-specific CD8 cells in patients with chronic hepatitis C (23 patients, CH), chronic asymptomatic carriers of HCV with persistently normal ALT and positive serum HCV-RNA (12 patients, AS) and in subjects with resolved HCV infection, either spontaneously (8 subjects, SP) or following anti-viral treatment (12 subjects, RT). Functional analysis was carried out with 5 highly immunogenic peptides corresponding to NS31073-1081 , NS31406-1415 , NS41812-1821 , NS41992-2000 , NS52627-2635 , containing previously identified HLA-A2 restricted epitopes. Since the different HCV genotypes were represented in the different groups of patients in different proportions, different sets of peptides designed on genotype 1, genotype 2 and genotype 3 sequences were synthesized and used to reproduce more closely the sequence of the viruses infecting each individual patient and responsible for in vivo priming of the CDS response. The immunological parameters tested were: a) frequency of HCV specific, CD8+ T cells by tetramer staining, both ex-vivo and after in vitro stimulation with synthetic peptides; b) IFN-7 production by intracellular cytokine staining; c) cytolytic activity by a standard %hromium release assay. The results of our study indicate that subjects recovered from HCV infection following treatment show lower ex-vivo frequencies of circulating HCV-specific CD8 cells compared to CH patients but their cells are able to expand and to produce IFN-7 more efficiently after in-vitro stimulation. Indeed, HCV-specific T cell lines were induced in 83% of RT subjects and in 74% of CH patients; moreover, IFN-7 production was induced in 100% of RT subjects compared to 67% of CH patients (p<0.05). In the group of subjects spontaneously recovered from infection, ex vivo CD8 frequencies were below the threshold of tetramer detection; however, peptide stimulation in vitro was able to expand tetramer+ CD8 cells and to induce IFN-7 production in 37% and 63% of the subjects, respectively. These lower levels of reactivity of SR subjects compared to TR patients are likely due to the different time elapsed from recovery, which was approximately 3 years in TR subjects but much longer in SP subjects, who had recovered even decades before the time of immunological analysis. Finally, the group of AS patients showed the lowest levels of response to HCV, in terms of both frequencies of HCV-specific circulating CD8 cells and cytokine secretion. In conclusion, the HCV-specific, CD8-mediated response has different features in patients with different outcomes of infection, showing a hierarchy of efficiency declining from subjects recovered after treatment who displayed the best responses, to chronic hepatitis patients, subjects recovered spontaneously and asymptomatic HCV-carriers, who appear to be the least responsive as a likely result of a deeper condition of tolerance to HCV. Hepatitis C virus (HCV) infection is the most frequent cause of chronic liver disease in Western countries and it has been involved in the development of cirrhosis with the risk of hepatocellular carcinoma. Cyclooxygenases (COX) are crucial enzymes in the biosynthesis of prostaglandins and COX-2, the inducible isoform, has been implicated in inflammation, fibrogenesis and carcinogenesis. To address whether COX-2 may play a role in HCVrelated hepatic inflammation and fibrosis, we determined the COX-2 expression pattern in the liver tissue from 27 patients with HCV-induced chronic hepatitis (low histological activity=22, high histological activity=5) and 5 with end-stage cirrhosis, searching for correlations between the expression level of this enzyme and the histological activity of liver disease as well as with the intrahepatic expression and activity of metalloproteinases (MMPs). We also investigated whether structural and non-structural HCV proteins may promote COX-2 expression in the human hepatocytederived cell line CCL13. Western-blot and RT-PCR analysis for COX-2 demonstrated that COX-2 expression levels were higher in mild chronic hepatitis (MCH, 2.6 fold), severe chronic hepatitis (SCH, 3 fold) and cirrhosis (2.4 fold) than in normal liver. Moreover, PGE2 levels were also higher in liver samples from patients with SCH (1.9 fold) and in those with cirrhosis (3.3 fold) than in normal liver. Besides other cell types, hepatocellular COX-2 immunoreactivity was markedly observed in HCV cirrhosis, and although some COX-2 positive hepatocytes were observed at the edge of hepatic lobules, the majority of them were mainly restricted to the regenerative nodules. In contrast, none or few COX-2 expressing hepatocytes located in periportal areas were observed in patients with MCH and SCH, respectively. Interestingly, we found a significant correlation between the intrahepatic expression and activity of COX-2 and MMP-2 and -9 in all the histological groups of patients analyzed. COX-2 mRNA, protein and activity was de novo induced in resting CCL13 cells stably transfected with HCV core and NS5A, and this effect was higher (1 fold and 2.5 fold) when activated with cytokines and phorbol esters, respectively. In conclusion, our results provide evidence that a virus-induced hepatocellular COX-2 upregulation could mediate important pathogenic events in the course of chronic HCV infection, suggesting that specific COX-2 inhibitors might be useful for chemoprevention and therapy of HCV-related liver disease. were nahe to HBV. Thl responses to HCV proteins NS3, NS4 and core were sought by ELIspot. As a control Thl responses were also sought to HBsAg, where results were analysed according to recovery from HBV infection or nayve to HBV. Thl responses to each of the 5 proteins was performed before and after depletion of T-regs using anti-CD25 (>80% depletion). Results: Following Treg depletion Thl responses were revealed de novo in 3/12 and enhanced in 8112 patients with NS3,2 and 9 with NS4 respectively and 4 and 7 with HCV core respectively. The overall increase in the Thl responses with T-reg depletion was significant for both HCV core (median increase 13-fold p = 0.004) and NS3 (5- Backaround: In both chronic viral hepatitis B and C viral persistence is thought to be due to an inadequate antiviral T cell response, which in turn may be caused by inappropriate priming of T cells by dendritic cells. An important subset of peripheral blood dendritic cells, the plasmacytoid dendritic cell, has been identified to be the major interferon-alpha producing cell in humans. Since both hepatitis B and hepatitis C infection can be successfully treated by exogenous interferon-alpha, an impairment of endogenous interferon-alpha producing cells is an attractive hypothesis for the pathogenesis of chronic viral persistence. Methods: Patients with acute symptomatic ( n = l l ) and chronic hepatitis C (n=25), sustained responders to interferon-alpha therapy (n=22) and patients with acute hepatitis B (n=lO) as well as healthy controls (n=20) were included in this study. Plasmacytoid dendritic cells were stained for FACS-analysis in peripheral blood mononuclear cells with antibodies against BDCA-2 and CD4 and by exclusion of lineage marker positive cells (CD3, CD14, CD16, CD19). In addition pDC were stained with various activation and maturation markers (e.g. CDSO, CD83, CD86). Production of interferon-alpha was measured by ELISA after stimulation with CpG. Results: In acute hepatitis C, IFN-alpha production was about 45-fold reduced as compared to healthy controls (15 pg/50000 PBMC vs. 701 pg150000 PBMC). This reduction was caused by both a significant reduction in absolute numbers of pDC (3.8/~1 vs. 8.4/pl, p=0.0032) as well as by a reduction of IFN-alpha production per cell (0.19 pg/PDC vs. 3.45 pg/PDC; p=0.0004). In chronic hepatitis C, IFN-alpha production was still reduced by over 50%, although absolute numbers of pDC were similar to healthy controls. Following spontaneous or treatment induced viral clearance both number and IFN-alpha secretion of pDC were not different from healthy controls. Importantly, in acute hepatitis B, which runs a self-limited course in the majority of cases, also the IFN-alpha production was reduced (21 pg150000PBMC). This reduction was also caused by both the reduction of absolute PDC count (5.291~1, p=O.l) as well as by the IFN-alpha secretion per PDC (0.32 pg1PDC; p=0.0007). Using a panel of DC activation and maturation markers we did not find any evidence of a more immature or more activated phenotype of pDC in acute hepatitis C. Exogenous IFN-alpha administration during acute hepatitis C led to a further reduction of IFN-production by pDC. Conclusions: Acute viral hepatitis has a dramatic impact on frequency and function of plasmacytoid dendritic cells in the peripheral blood, indicating that pDC play an important role during this phase of viral hepatitis. However, no differences were found between patients with self-limited vs. chronically evolving acute hepatitis C or patients with acute self-limited hepatitis B. Future studies have to clarify whether the reduced IFN-alpha secretion by pDC contributes to viral persistence or whether this is rather part of a physiological response to acute viral disease. It has been suggested that hepatocellular steatosis, a frequent histological feature of chronic hepatitis C, is more frequent in HCV genotype 3 infection, and disappearance of steatosis in patients who clear HCV genotype 3 infection after antiviral therapy suggests a possible direct role of this HCV genotype. In order to assess the direct causal role of HCV in steatosis, we studied the relationship between HCV RNA load and steatosis according to the HCV genotype. Methods: We studied 176 patients with chronic hepatitis C (genotype 1, n = 83 ; genotype 3, n = 93). The serum HCV RNA level was measured at the time of liver biopsy by means of a third-generation "branched DNA"-based assay (Versant HCV RNA 3.0 Quantitative Assay, Bayer Diagnostics). Steatosis was graded as absent, mild (< 10% of hepatocytes), moderate (10 to 30% of hepatocytes) or marked (> 30% of hepatocytes). Daily alcohol intake during the 6 months prior to liver biopsy, and the body mass index (BMI), were recorded. Results: Steatosis was more frequent in patients with genotype 3 infection than in those with genotype 1 infection (84.9% vs 74.7%, NS). Steatosis was significantly more severe in HCY genotype 3 than in genotype 1 infection (moderate or marked in 54.8% vs 25.3%, p=O.OOOl). In patients infected by genotype 3, the severity of steatosis was significantly related to HCV RNA load but not to alcohol intake or BMI. In contrast, in patients infected by HCV genotype 1, the severity of steatosis was not related to the HCV RNA level but was significantly influenced by alcohol intake and BMI. Bivariate analysis demonstrated the effect of the HCV genotype on the association between the severity of steatosis and viral load in genotype 3-infected patients, and between the severity of steatosis and exogenous metabolic factors in genotype 1-infected patients. Multivariate analysis showed that : (i, in patients infected by HCV genotype 3, only HCV RNA load was independently related to the severity of steatosis (odds ratio (OR) = 2.5, 95% confidence interval (CI): 1.4-4.4; p = 0.001), (ii) in patients infected by HCV genotype 1, BMI higher than 28.0 kglm2 (OR = 5.8, 95% CI: 1.3-26.9; p = 0.016), alcohol intake exceeding 30 glday (OR = 29.6, 95% CI: 3.2-272.0; p = 0.009), and histological grade (OR = 20.0, 95% CI: 2.6-152.0; p = 0.001) were independently related to the severity of steatosis. Conclusion: Our results suggest : (i) steatosis is a cytopathic lesion induced by HCV genotype 3 ; (ii) HCV genotype 1 is not steatogenic per se or at the usual in vivo expression levels, and liver steatosis is a feature of associated steatohepatitis in these patients. The effect of HCV genotype 3 sequences and the role of their expression levc4s must be tested in vitro, in order to better reflect the in vivo situation. In animal models, steatosis is associated with oxidative stress and lipid peroxidation which is known to promote fibrogenesis. This study was aimed to assess in patients with chronic hepatitis C whether steatosis contributes to fibrosis through oxidative stress. Methods: Markers of lipid peroxidation and antioxidant status were measured in blood from 192 chronic hepatitis C patients and age and sex matched healthy controls. Intrahepatic levels of these markers were also assessed in a subgroup of 23 patients and controls. Results: The lipid peroxidation marker malondialdehyde was significantly higher in the blood and in the liver of patients compared to controls (p=0.05 and p=O.O01 respectively) while the antioxidant glutathione peroxidase was significantly lower (p=0.05 and p =0.0001 respectively). The antioxidant superoxide dismutase was significantly higher in the blood and lower in the liver compared to controls (p=O.O01 and p=0.002 respectively). Autoantibodies to soluble liver antigen (SLA) are specific for autoimmune hepatitis. The molecular characterization of the SLA antigen has allowed the establishment of highly sensitive and specific radioligand assays. To determine serum anti-SLA a specific radioligand assay was used. Total RNA was isolated and reverse transcribed from HepG2 cells. The cDNA encoding SLA was amplified by PCR and used as a template to express SLA protein eukaryotically in a TNT coupled reticulocyte lysate system (Promega Corporation, Southampton, UK). Anti-SLA was measured retrospectively in 16 consecutive patients (10 girls, median age at transplant 52 months; range 8-160) with dn-AIH, in whom sera were available before transplant, 1,2,6,12 and 24 months post-transplant and at the time of the diagnosis of dn-AIH. Eight patients had biliary atresia, 2 Alagille syndrome, 2 cryptogenic cirrhosis, 2 alpha-1-antitrypsin deficiency, 1 druginduced acute liver failure and 1 BSEP deficiency. The patient with acute liver failure did not have pre-transplant serum specimen stored. Twelve patients had developed smooth muscle andlor antinuclear antibodies, two anti-liver kidney microsomal (one atypical) and two anti-mitochondria1 antibody. Before transplantation, anti-SLA was negative in 13/15 cases, the two positive patients having cryptogenic cirrhosis and biliary atresia. Post transplant, anti-SLA remained positive in these two and became positive in further 9 patients. Of these, in 7 patients anti-SLA became positive at a median of 17 months (range: 1-115) before the clinical diagnosis of dn-AIH. In 5 children anti-SLA became detectable as early as 2 months post-surgery. In 6 patients anti-SLA levels were highest at the time of the clinical diagnosis of dn-AIH. The presence of anti-SLA in dn-AIH supports the autoimmune nature of this condition; the frequent appearance of anti-SLA before the clinical manifestation of the disease makes it a potential predictive marker for development of dn-AIH. All 38 patients showed positive response to the therapy, with respect to remarkable release of severe meteorism, active diet, and significant improvement of liver and kidney functions. However, no difference was presented in the markers of electrolytes, blood routine and blood gas analysis before and after the MARS, while the effects on serum levels of alanine aminotransferase, aspartate aminotransferase and y-GT were remained uncertain since the ALT decreased from 941 ulL to 690 ulLduring a In conjunction with observations from other centres, our data show that MARS treatments resulted in a ramarkable removal of bilirubin, uric acid, BUN, Cr and ammonia, which could therefore decrease the toxic effects that higher concentrations of these compounds exert on liver and kidney function and could thus contribute to improvements in multiple organ dysfunctions. We found additionally that the higher concentration of serum toxin before detoxication therapy, the more efticacy removal could be achieved which presents MARS could be {of therapeutic results even in the very serious cases. It remains questionable whether some of useful substances characted in low molecular weight such as TRHlthyrotrophin-releasing factor), GnRH), ADH(Anti-diuretic hormone and Calcitonin will be also removed by albumin dialysis, whether the added synthesis function is necessary for an artificial liver support, and whether this encouraging survival rate applies to long term outcome remains open for further investigation. scoring system has recently been introduced to determine priority for organ allocation in liver transplantation (LT). There is limited available data on resource utilization in the post-MELD era.AIMS: 1. Evaluate the effect of the MELD system on LT-associated resource utilization and post-LT survival. 2. Compare the costs of LT in the periods before and after the implementation of MELD. METHODS : Patients undergoing LT at our center in the 6 months following MELD implementation(cases) were compared to those undergoing LT in the immediately preceding 9-month period (controZs). The outcome parameters studied were: 1) Resource utilization: defined as ( There were no significant differences between the two groups in terms of age, sex distribution or presence of hepatocellular carcinoma (HCC). The average MELD among cases was 25.9, compared to 24.1 in controls (p =0.19, NS). Overall LOS was similar in the two groups. However, pre-LT LOS, especially in the intensive care unit (ICU), was significantly higher in the pre-MELD or control group. The pre-LT cost was also higher in this group(p=0.05), translating into significantly higher total costs of LT in the pre-MELD period (p=O.Ol). The need for post-LT HD, PMV and pressors was no different between the two groups. This data is summarized in table 1. The MELD score was found to be significantly correlated with Bile d u d damage is a maior feature in liver graft rejection. Ductopenia (DP), defined as loss of more than 50%-of bile ducts, is a major hallmark of chronic liver graft rejection (CR), which usually leads to graft failure within the first post transplant year. In a previous study we have shown that in patients who developed DP and CR a deficient proliferative response of canals of Hering (CoH) was present in the preceding biopsies showing acute rejection (AR) when compared with a control group who experienced AR but did not progress to CR. These findings support the postulation that CoH represent a regenerative compartment of the biliary unit in the liver. Aim of the present studv: to analyze whether preservation of CoH might contribute to the reversibility of DP. Patients and Methods We studied 2 groups of patients. The index group (IG, n=5) developed loss of bile ducts without progression to graft failure due to CR during a follow up of 5 years after transplantation. The second group (CRG, n=12) were all CR patients, retransplanted at a median time of 7 months (range 2-19 months). Reperfusion (RB), 1 week (1-W), 1 month (1-M) biopsies were studied in both groups. In the CRG the last biopsies (LB) before retransplantation taken at a median time of 160 days (range 42-329 days) were also studied. Additionally, in the IG the 1& (1-Y), 2nd (2-Y) and 5* (5-Y) annual biopsies were also included. Bile ducts and CoH were identified by immunohistology using cytokeratin 7. Ki67 (MIB) was applied to investigate the proliferative activity. The number of bile ducts and CoH were counted per portal tract and the number of K67+ cells was counted as ratio of the total number of cells in these structures. Clinical follow-up of the IG group was studied based on liver function tests at similar time points as the biopsies. The IG group showed damaged bile ducts and decreasing numbers of bile ducts in the course of time, leading to DP at a median of 3 years. Contrastingly, the CRG developed DP at a median time of 160 days, observed in the last biopsies taken before retransplantation. When compared with the CRG, the IG showed significantly less proliferative activity in bile ducts at 1-W (p=O.O43) but not in RB, 1-M and 1-Y, the latter was compared with the LB of the CRG. CoH showed an initial increase at 1-W in the IG, but a progressive decrease in the subsequent biopsies, reaching significant loss at 5-Y (p=0.043). In the CRG, progressive loss of CoH started at 1-W, leading to a significant loss within 1 year. Numbers of CoH were consistently lower in the CRG at all time points except RB, although not statistically significant. There were also no significant differences in proliferative activity both within the IG in the course of time and when compared with CRG. All IG patients continuously showed abnormal liver function tests apart from the bilirubin levels, which were elevated during the first month after transplantation but were normal after 1-Y. At 5-Y, median serum level of alkaline phosphatase was 228 U/l, gamma glutamyl transpherase was 312 Ull, bilirubin was 26 micromoll1 and ASATlALAT were 96193 Ull. Conclusion: The IG showed a much slower course of loss of bile ducts when compared with CRG, leading to graft survival of more than 5 years. A higher proliferative activity in bile ducts at 1 week in the CRG did not prevent the progressive development of DPlCR in CRG. The initial increase of CoH at 1 week in the IG might be responsible for the prolonged preservation of CoH in the IG. As CoH are believed to represent a regenerative compartment of the biliary unit, our findings indicated that prolonged preservation of CoH might contribute to the delayed occurrence of DP but apparently not to reversibility of DP. Absence of reversibility of DP is supported by the abnormal liver function tests. (LTx) . Recurrent HCV poses a significant and possibly increasing threat to graft and patient survival. Diagnosis of acute rejection and alterations in immunosuppression (IS) may significantly impact the tempo of post-tx HCV. These associations motivated us to re-examine risk factors for early acute rejection (EAR) in a large and contemporary cohort of LTx recipients. Methods: The study cohort comprised of 282 consecutive adults undergoing primary LTx between 1/1/99 and 10/31/2002 for chronic liver disease with a minimum of 7 day graft and patient survival. Recipients received triple IS, typically with steroids, tacrolimus, and mycophenolate mofetil. EAR was defined as biopsyproven rejection or treatment for presumed rejection within 6 months of LTx. Risk factors for EAR were determined by Cox proportional hazard methods. Spearman rank and phi correlations were used to determine associations between factors. Results: 177 men (63%) and 105 women (37%) underwent deceased (n = 236; 84%) or living donor (n = 46; 16%) LTx. The etiologies of chronic liver disease were HCV (n=138; 49%), AIH I PBC / PSC (n=40; 14%), HBV (n= 34; 12%), cryptogenic (n=27; lo%), alcohol (n=24; 9%), and miscellaneous (n=19; 7%). Our overall incidence of EAR was 45% (126 I282 recipients); 118 recipients (94%) had biopsy-proven rejection while 8 recipients (6%) were treated for rejection without histologic confirmation. Cox univariate models showed that HCV, female gender, MELD 2 28, year 2000 compared to year 1999, and lower day 5-7 tacrolimus (TAC d5-7) level were positively associated with EAR while Asian compared to Caucasian ethnicity was negatively associated with EAR (Table la) . Factors such as recipient and donor age, recipient African American or Hispanic ethnicities, donor type (deceased or living), other years (2001 and 2002) , Child's score and class, pre-LTx ICU location, and cold and warm ischemia times were not significant risk factors. Cox multivariate models showed that HCV diagnosis and female gender remained independent and significant risk factors for EAR (Table lb) . HBV etiology and Asian ethnicity were significantly correlated (phi coefficient 0.53; p = <0.0001) and thus, did not remain independent risk factors in multivariate analysis. While lower tacrolimus level tended to predispose to EAR, MELD 2 28 and year 2000 were no longer associated with EAR. Lower tacrolimus level was significantly correlated with both higher MELD score (Spearman rank correlation -0.34; p = 0.0000) and later transplant year (Spearman rank correlation -0.17; p = 0.019). Conclusions: HCV etiology of liver disease is strongly associated with EAR after LTx. Risk of EAR is higher for HCV than EtOH, cryptogenic, and HBV etiologies and comparable to AIH / PSC I PBC etiologies; this effect is independent of gender, ethnicity, year, MELD, and post-LTx IS. Recipients with high MELD scores are also at increased risk for EAR, at least partly because of lower post-LTx IS. It is unclear whether the strong association of HCV etiology with EAR is because HCV infection results in an immunologic environment predisposing to EAR or whether we are unable to accurately diagnose rejection in the setting of HCV recurrence. Aim: Although the donor pool has expanded in response to increasing waiting lists the quality of donor livers has suffered as a result. It remains unclear whether such marginal organs can be safely used in high-risk recipients or whether they should be implanted solely into good recipients. We aimed to define recipient selection criteria for implantation of these grafts. Methods: A prospectively collected database containing 397 patients who underwent orthotopic liver transplantation between 1994 and 2002 was analysed. Donors were scored using a formula derived by logistic regression that identified 3 covariates (graft steatosis, donor age and cold ischaemia time) that independently correlated with primary graft dysfunction. This enabled them to be stratified into either marginal or non-marginal groups. Using logistic regression analysis, recipient factors independently correlated with 1-year post transplant survival in the marginal group (recipient age, plasma albumin and urea) were built into a mathematical model and each patient was given a score accordingly. Patients with scores higher than the defined optimum cut-off value were considered as high-risk and the others with lower scores as low-risk recipients. Outcomes were then evaluated in these subpopulations Results: Marginal and non-marginal donor groups consisted of 65 and 332 patients respectively. The recipient score derived from the multivariate analysis was as follows : Score = 1.989 x plasma albumin+ 1.22 x age group + 2.076 x urea, with the recipient albumin level coded as 1 for < 3.1 gldl and 0 for 2 3.1 gldl, the recipient age group coded as 2 for > 55 years, as 1 for 43-55 years, and as 0 for 22.4 mgldl, and as 0 for 5 22.4 mgldl. The ROC curve analysis showed that the score had a good discriminating power (area under the curve 0.78, p =.001) and the ideal cut-off value demarcating high-risk from low-risk recipients was 3.25. Therefore the recipients were classed as high-risk if they had a plasma albumin level below 3.1 gldl with either an age of over 55 or a urea level of above 22.4 mgldl or when they had a urea level over 22.4 mgldl with an age of over 42. Of the 65 recipients in marginal group, 28 (43%) were classified as high-risk and 37 (57%) were classified as low-risk. There was a huge 1-year survival difference between high-risk and low-risk recipients (60.7% and 91.9%, respectively, p <.0001 Background Single-center experience with pretransplant use of rabbit anti-human-thymocyte globulin suggests that despite early depletion of lymphocytes, a third of all pediatric liver recipients develop early rejection, while the remainder demonstrate clinical graft adaptation. PurposelMethods: To identify potential mechanisms, 31 pediatric liver recipients, median age 7.8 years, median followup 8 months, received pre-and post-transplant measurements of 1. whole blood concentrations of Tacrolimus (TAC), 2. interdose changes in mitogen-stimulated T-and B-cell responses (sLR), 3. dendritic cell (DC),and peripheral blood mononuclear cell subsets, and 4. CD8+28-suppressor effect on donor antigenpresenting cell types (CD34fmonocytes and CD19+ B-cells). All patients received rATG preconditioning with a total dose of 5 mglkg in two divided doses. The first dose was given before liver transplantation (LTx). Maintenance agent was TAC without steroids. In 9 patients this was replaced with Sirolimus (SRL). Mitogen-stimulated T and B-cell responses were compared with those seen in a historical population, who had not received rATG pretreatment. Results: All measurements were performed at a median interval of 47 days (22-57 days) after LTx. 1. In sLR, the frequency of T-cells expressing the cytokines IFN-g, TNF-a, and IL-2 decreased with increasing total exposure (AUC) to TAC in historical controls (n=6). This relationship was markedly dampened in rATG patients (n=15), as suggested by slopes of 0.0023, 0.032, and 0.0006 relating IFN-g, TNF-a and IL-2 on the y-axis to TAC AUC. 2. Significant decrease in CD4 absolute counts at 1 week and partial reconstitution at 2 months after rATG pretreatment (mean pretx-3120 vs 1 week-1920 vs month 2-2117 celllmm3). During this period, monocytes and B-cells remained stable. 3. DC2 remained unchanged, while DC1 frequency increased significantly, from 0.15 to 0.23, p=0.05. 3. In coculture experiments, purified CD8+28-subpopulations from two recipients receiving minimal TAC doses induced decreased CD86 expression in donor APC, but increased its expression in HLA-mismatched APC. Eleven of 31 subjects experienced rejection, while 18 subjects are being maintained on daily (n=13) or every other day (n=5) doses of TAC or SRL. Conclusions: Despite lymphocyte depletion, rejection in the setting of T-cell anergy can be explained by relative sparing of antigen-presenting cells, which can recruit alternative effector mediators. The appearance of donor-specific CD8 + suppressor cells in recipients on low immunosuppression suggests that these conditions may also foster a favorable immunomodulatory response toward the liver allografts. (dropout) . The objective of this study was to evaluate the impact of the HCC-adjusted Model for End-Stage Liver Disease (MELD) organ allocation scheme on the intention-to-treat outcome. Under the MELD scheme, patients with HCC meeting the United Network for Organ Sharing (UNOS) T1 (single lesion under 2 cm) and T2 (single lesion between 2 to 5 cm, or 2 to 3 lesions none exceeding 3 cm) criteria were eligible for an initial MELD priority score of 24 and 29, respectively. They were also entitled to an increase in MELD score, corresponding to a 10% increase in mortality, for every 3 months on the waiting list. METHODS: Excluding patients undergoing living-donor liver transplantation, we prospectively evaluated 102 consecutive patients with HCC listed for OLT since January 1998. The Kaplan-Meier probabilities of OLT and dropout among 44 patients with HCC listed under the MELD system between February 2002 and January 2003 were compared with 58 patients listed between January 1998 and January 2002 under the previous system of organ allocation. Follow-up in the pre-MELD group was censored on February 27,2002, when the MELD system for organ allocation was implemented by UNOS. For patients under MELD, follow-up was censored on February 27,2003, when further refinements of the MELD policy for HCC were made. RESULTS: The baseline characteristics were not significantly different between the two groups. Eighteen of 44 patients (41%) in the MELD group and 26 of 58 patients (45%) in the pre-MELD group received chemoembolization or various ablation treatments before OLT (p=0.89). All patients in the MELD group met T1 (5 patients) or T2 criteria (39 patients). In the pre-MELD group, 6 patients had HCC stage exceeding T2 but meeting our proposed expanded criteria (single lesion not exceeding 6.5 cm or no more than 3 lesions none greater than 4.5 cm with total tumor diameter not exceeding 8 cm) by the time of OLT. The Kaplan-Meier cumulative probabilities for OLT at 3, 6, and 8.5 months of longest followup were 22.5%, 64.0% and 88.0%, respectively, in the MELD group, versus 17.2%, 24.7%, 35.8%, and 47.2% at 3, 6, 9, and 12 months, respectively, in the pre-MELD group (p=0.0006). Under MELD, none of the 5 patients with T1 lesion had received OLT. The cumulative probabilities for OLT for the 39 patients with T2 HCC in the MELD group were 25.5%, 69.2%, and 89.2%, respectively at 3, 6, and 8.5 months (p=O.OOOl versus the pre-MELD group). The cumulative probability of dropout was 5.6% at 8.5 months of longest followup without OLT under MELD, whereas the cumulative probability of dropout increased from 7.2% at 6 months to 37.8% at 12 months in the pre-MELD era. The difference did not reach statistical significance (p=0.74) largely due to low dropout rates in the first 6 months for both groups. Among the patients who received OLT, 3 of 22 patients in the MELD group versus 9 of 25 patients in the pre-MELD group had pathologic HCC stage in the explant exceeding T2 criteria (p=0.78). Unfavorable histologic tumor features in the explant, including either poorly differentiated grade or microvascular invasion or both, were observed in 4 of 22 patients in the MELD group versus 5 of 25 patients in the pre-MELD group ( I " 1.0). The short duration of follow-up under MELD precluded comparison of intention-totreat survival or HCC recurrence between the two groups. CON-CLUSION: The HCC-adjusted MELD system significantly improved the probability of timely OLT, and was not associated with selection of a greater proportion of HCC with unfavorable explant tumor histology. Given the low probabilities for dropout in the first 6 months following listing for OLT even in the pre-MELD era, patients with HCC might have indeed received too high a priority score in the first year under MELD. Disclosures: Nancy L Ascher -No relationships to disclose Nathan M Bass -No relationships to disclose randomized at D7 to receive maintenance IS regimen with ciclosporine microemulsion + prednisone (group 1) or without steroids (ciclosporine microemulsion + placebo, group 2) after a 7 days blinded oral steroid tapering period. Results : 193 patients were recruited and a total of 174 were randomized at D7 (group 1 = 90, group = M). There was no difference between the 2 groups for baseline characteristics, proportion of patients with hepatitis C (18.9% and 20.2%), or cyclosporine blood levels. The incidence of treated biopsy confirmed acute rejection at 6 months was 24.4% in group 1 and 38.1% in group 2 (p= 0.03), with a trend for higher incidence of grade 213 rejection ( 18.9% vs 28.6% ; p=0.12). This difference was maintained in HCV pos and HCV neg patients. No difference was observed between the 2 groups in terms of adverse events, infections, incidence of hypertension and renal dysfunction. Changes from baseline were similar with regards to metabolic parameters. A trend towards a better glucose tolerability was observed, less patients receiving an antidiabetic treatment in the placebo group ( 2 vs 10). Conclusion : peritransplant immunosuppression with basiliximab, cyclosporine microemulsion, and steroids resulted in excellent low rejection rates in LT patients. Early steroid withdrawl strategy at day 14 is not supported by the results of this study, which showed an higher incidence of acute rejection and a trend to more severe acute rejection, only balanced by a trend to a lower need of antidiabetic treatment. To determine if an association exists between MELD score and post transplant graft loss within 3 months. METHODS From 2/28/02 through 10/31/02,2,745 patients underwent liver transplantation, 92% of whom were followed for a minimum of three months. Cox regression analysis was utilized to assess the relationship between MELD score and post transplant outcome. transplantation. Seventy one percent of patients were positive for autoantibodies: 52.9 % were positive for anti-nuclear antibody, 23.5% for anti-smooth muscle antibody and 0% for anti-liver kidney microsomal antibody. The average ALT elevation was 224 (range 61-486) and AST elevation 166 (range 52-346). An elevated GGTP was seen in 70.6% of patients. All patients were Hepatitis C PCR negative. At the time of diagnosis, 12 patients were being treated with cyclosporine, 5 with F'K506, and 7 with steroids in addition to calcineurin inhibitors. After diagnosis, all patients received standard therapy with l-Zmg/kg of steroids. In addition, 89% (15117) began hthioprine and 24% had their calcineurin inhibitor dose decreased. After a mean of 2.6 years of follow-up (range 0.53-5.93), 59% remain steroid dependent and 12% are off steroids. One patient was re-transplanted for biliary complications and 1 patient required conversion to Sirohus. Conclusion: Diagnosis of de novo AM requires a high index of suspicion as auto-antibodies are often negative. Steroid therapy, often with the addition of azathioprine, is effective in controlling disease. Many patients, however, become steroid dependent in order to remain in remission. Background. Liver allografts have an improved outcome compared with other solid organ grafts, and rodent studies have suggested that activation-associated lymphocyte death may play a key role. Studies from our laboratory have shown increased levels of apoptotic lymphocytes in human liver grafts compared with renal grafts and native liver. Although the levels of apoptosis did not differ significantly between those who developed acute rejection (REJ) and those who did not (NR), the earliest timepoint studied was 3 days post-OLT. We hypothesized that the very early postoperative period (within 48 hours) would reveal a relationship between leukocyte apoptosis and rejection. Aims. The aim of this study was to determine the amount of early leukocyte apoptosis and its relationship with the degree of lymphocyte activation, subsequent rejection status and degree of donor cell chimerism. Methods. Peripheral blood mononuclear cells were isolated from 76 patients undergoing OLT and were collected on the day prior, 5 hrs after reperfusion (Day 0) and 24 hrs post-OLT (Day 1). DNA and RNA were prepared and real-time PCR used to quantify apoptosis (ligation-mediated PCR), lymphocyte activation (IL-2, IFN-gamma, IL-10, CD40 ligand, using GAPDH as an internal standard) and donor cell chimerism (Y chromosome DYZ3 or donor-specific DRB1). Results. The mean level of circulating apoptotic cells in Day 1 recipient PBMC was higher than healthy controls (0.9% t-0.2 vs 0.2% ? 0.1, ~~0 . 0 1 3 ) . Apoptosis was greater in NR (1.1% 2 0.3) compared with REJ (0.3% t-0.1, p=O.O21). On Day 1 the PBMC from NR had increased expression of IFN-gamma (p=0.006), IL-10 (p=0.016), CD40 ligand (p=0.02) and IL-2 (trend) compared with REJ, despite no difference in lymphocyte counts. Donor cell chimerism on Day 1 did not differ between the groups indicating that this was unlikely to account for increased leukocyte apoptosis in the NR group. Interestingly, the level of chimerism 5 hrs postreperfusion (Day 0) was significantly higher in NR (3.8% f 0.6) compared with REJ (1.2% t-0.4, p=0.004) and there was a close correlation between chimerism on Day 0 and PBMC cytokine expression on Day 1 (r=0.58, p=0.006). Conclusions. Patients who did not experience rejection had a paradoxical increase in markers of lymphocyte activation at Day 1, in association with higher levels of leukocyte apoptosis. This suggests that recipient cell death may have a role in graft acceptance and reduced acute rejection in human OLT. The higher donor cell chimerism seen in Non-Rejectors implicates the passenger leukocytes in the process of heightened cell death. Disclosures: Andrew Clouston -No relationships to disclose Wenyi Gu -No relationships to disclose Julie R Jonsson -No relationships to disclose Elizabeth E Powell -No relationships to disclose Daina M Vanags -No relationships to disclose Amadeo Marcos, Bridget Flynn, Paulo Fontes, Thomas Cacciarelli, Wallis Marsh, Michael DeVera, Obaid Shakil, Noriko Murase, Anthony Demetris, John Fun% Thomas E Stanl, University of Pittsburgh, Pittsburgh, PA The seminal mechanism of organ engraftment is thought to be immune activation-dependent clonal exhaustion-deletion. The conventional use of heavy immunosuppression may depress the initial step of donor-specific activation to the extent that the treatment is anti-tolerogenic. To avoid this pitfall, we have applied 2 therapeutic principles in management of 78 adults cadaveric liver recipients transplanted between 9/2002-5/2003. The treatment principles were, first, host conditioning prior to transplantation, and second, minimum post-transplant immunosuppression. The host conditioning was done with one gram of methylprednisolon and a single infusion of 30 mg of alemtuzumab, completed before liver revascularization. There were 49 males and 29 females in the group with mean age 52.3110.3 years (32-73). Main causes of liver disease were hepatitis C (40%), cholestatic liver disease (18%), alcoholic liver disease (22%). Daily post-transplant monotherapy with tacrolimus (trough target 10 ng/ml) was started 12-24 hours postoperatively (starting tacrolimus dose: 10.5+5 mg, median 10). Because of suspected neurotoxicity, 5 patients were switched to cyclosporin. In addition sirolimus monotherapy was substituted for tacrolimus in one patient for management of nephrotoxicity. Immune activation diagnosed by liver function tests or by mild or equivocal rejection in liver biopsies frequently was not treated, and tended to resolve spontaneously. Clinically and pathologically significant rejection was treated with a bolus of methylprednisolone and/or a 30 mg dose of alemtuzumab. Nine patients (11.5%) died 5 sepsis; 2 PNF; one coagulation disorder (hypercoagulable state), and one congestive heart failure. Of the 69 surviving recipients four experienced rejection during the first two months which was readily reversed with a bolus of methylprednisolon. After demonstrating the absence of immune activation in liver biopsies at 120 days post-transplantation, weaning from monotherapy was initiated in 49 patients (with the aim of completely stopping treatment in selected cases). Seven patients experienced one episode of rejection that was reversed with single dose of methylprednisolone and/or infusion of 30 mg of alemtuzumab. Presently, 14 patients are on once a day monotherapy, 14 patients on every-other-day, 10 patients on three timedweek, 10 patients on twicelweek and one patient on once-a-week immunosuppressive regimen (latest tacrolimus dose 5.7t3.5 mg, median 5.5). The patients on cyclosporin and sirolimus are also following the same stepwise weaning pattern. No cases of new onset diabetes, renal failure or hypertension was found in these patients. Mean serum creatinine level in the study group at the time of transplantation was 1.020.4 mgldl and at the latest follow up 1.220.3 mgldl. CMV infection was seen in 22% of the patient population but was easily treatable with antivirals. No PTLD was seen during the follow-up period. Hepatitis C recurrence was seen in over 70% of the patients. Response to anti-HCV therapy in this group, after reduction of immunosuppression and especially during the weaning process was promissing. Conclusion: By applying the principles of immunosuppression outlined above, it has been possible to drastically reduce the amount of total immunosuppression relative to any (of our) pre-vious experience with liver transplantation. To see less side effects of chronic and high dose immunosuppressive therapy, and probably to have a chance to get a better response to treatment in our hepatitis C population. The results suggest the possibility of systematically achieving drug-free tolerance after liver transplantation. Background:End stage liver disease as a consequence of hepatic sarcoidosis is an uncommon indication for liver transplantation (LT). Consequently, there is a paucity of information on the pre-LT findings and postoperative course of individuals transplanted for hepatic sarcoidosis. The purpose of this study was to evaluate our experience with LT for sarcoidosis. Methods: Cases were identified by review of the Mount Sinai Hospital LT database. Patient records were reviewed; including the pathologic analysis of the hepatic explant, to confirm that sarcoidosis was responsible for liver failure. For each case, two control patients with other causes of liver failure matched for age, gender and date of transplant were selected. Data collected encompassed a mean follow up period of 5 years (range 1-8 years) post-LT. Two-tailed Student's t-test was used for comparison of continuous data, two-tailed Fisher's exact test for comparison of categorical data and log rank test for comparison of survival d,ita. Results: Hepatic sarcoidosis was the indication for LT in 7 of 2016 adult-LT (0.3%) performed from September 1988 -June 200.3. The mean age at transplant was 56 years and 4/7 (57%) were males. The diagnosis of sarcoidosis was established by findings of extensive, non-caseating granulomas in pre-LT biopsy specimens or in the native liver explant. In 2/7 cases, sarcoid had been previously diagnosed by biopsies of lung parenchyma or mediastinal lymph node. 6 of 7 patients were AMA negative. The sole exception was a patient with an AMA titer of 1:320 in whom sarcoid was still considered the most likely diagnosis based on liver biopsy findings of granulomas present in both portal and lobular areas and granulomas in a lung biopsy performed 10-years prior to LT. Extrahepatic disease was limited to pulmonary involvement in 4 patients with radiographic findings of either interstitial infiltrates (2) or interstitial infiltrates and hilar adenopathy (2). The mean age at transplant and gender distribution were identical in cases and controls. The indications for LT in the control group were: hepatitis C (7), PBC (2) and one patient each with hereditary hemochromatosis, cryptogenic cirrhosis, hepatitis B and Laennec cirrhosis. No statistically significant differences between the groups were present with respect to pre-LT levels of AST, ALT, alkaline phosphatase, total bilirubin, serum creatinine or prothrombin time. Cases and controls had a similar prevalence of ascites and anti-hepatitis B core antibodies. In contrast, sarcoid cases were much more likely to have a diagnosis of diabetes mellitus (100% vs. 21%, p=.002) and less likely to have antibodies to hepatitis C (0% vs. 50%, p=.05). Standard orthotopic LT was performed (two control patients received right trisegment grafts). Rates of acute cellular rejection were 71% in cases and 43% in controls (p=O.44) with cases experiencing a greater number of acute rejection episodes per patient (1.9 vs. 0.6 episodeslpatient). De novo autoimmune hepatitis developed in one case and one control patient. Both de novo autoimmune hepatitis and chronic rejection developed in one case patient. Recurrence of hepatic sarcoidosis was diagnosed in two patients at 1.5 and 5.6 years of follow-up. Among cases, the one-year graft and patient survival rates were 100% and five-year graft and patient survival rates were 86%. There was no statistically significant difference in five-year graft or patient survival between cases and controls. Conclusions: End-stage liver disease as a consequence of sarcoidosis is a rare indication for LT. These patients share many features in common with those transplanted for other indications. Despite the small number of patients, a relatively high rate of acute rejection and de novo autoimmune hepatitis was observed in patients with sarcoidosis. Recurrence of hepatic sarcoidosis was observed. Five-year graft and patient survival rates were comparable to those of patients transplanted for other indications. Disclosures: Sander Florman -No relationships to disclose Leona Kim Schluger -No relationships to disclose Kevin M Korenblat -No relationships to disclose Evan J Lipson -No relationships to disclose 440 TRANSPLANTATION. James D Eason, Ari J Cohen, Safheesh Nair, George Loss, Ochsner Clinic Foundation, New Orleans, LA Sirolimus has been used successfully in improving renal function in OLT recipients previously treated with tacrolimus and steroids. We report our experience with early sirolimus conversion in steroid-free OLT recipients to determine safety and efficacy in patients never treated with steroids. Methods: We performed 220 OLT over a threeyear period. Steroid-free immunosuppression with rabbit ATG induction and tacrolimus and MMF was used in 140 of these patients. Twenty-five of these steroid-free recipients were converted from tacrolimus to sirolimus within one week to ll months of transplant. Results of these patients receiving sirolimus conversion were reviewed. Results: Renal insufficiency was the reason for conversion in18 patients, while seven patients were converted because of neurotoxicity. Six patients were dialysis-dependent at the time of conversion. One-year patient survival in this high-risk group was 80% (20/25) compared to 88% in patients who remained on tacrolimus. The four deaths were in patients converted because of renal failure, three of whom were on dialysis. The incidence of rejection was 31% in sirolimus patients compared to 24% in patients who remained on tacrolimus. Rejection was treated by the addition of MMF or reintroduction of low-dose tacrolimus. Only one patient required steroids to reverse rejection. Conclusion: Early sirolimus conversion is safe and effective Lawal, Sandy Florman, Isabel Fiel, Ronald Gordon, Myron Schwartz, Charles Miller, Thomas D Schiano, The Mount Sinai Medical Center, New York, NY Introduction: Primary non-function (PNF) fter liver transplantation occurs in approximately 5% of cases and is fatal without timely retransplantation. PNF has been associated with many risk factors, however the etiology remains unknown. Some evidence suggests that ultrastructural changes in the liver may be causative. Aim: We sought to examine the hepatic ultrastructure of donor allografts in patients with and without PNF using electron microscopy. Medical Center, over 2000 adult orthotopic liver transplants were performed. Patients with the classic clinical presentation of PNF requiring retransplant were identified. Archived pre-and postreperfusion donor liver biopsies were examined by electron microscopy in 9 patients with PNF and in 9 matched controls. Each PNF case was matched by donor age t5years, gender, cold ischemic time klhour and the donor's cause of death. All biopsies were blindly reviewed by the same pathologist. Particular attention was paid to abnormalities of the mitochondria, endoplasmic reticulum and sinusoidal endothelial cells. In addition, the glycogen content of the cells was also assessed. The recipient age and creatinine, as well as the donor serum peak transaminases and bilirubin were compared.Non parametric tests with p values <0.05 were regarded as significant using the SPSS 11.5 program. Results: Overall, patients with PNF were older (55.7 t 10 vs. 48.99 years, p=0.03) and had higher peak ALT levels (141 ? 185 vs. 40 2 32 U/L, p=0.04). There was no significant difference in recipient peak serum creatinine, donor peak serum AST, sodium or donor peak serum bilirubin. In all cases, the endoplasmic reticulum and sinusoidal endothelial cells were ultrastructurally normal. The hepatocytes had variable degrees of glycogen pooling and moderate to severe fatty infiltration. In 7/9 (78%) PNF cases vs. 2lY (22%) control had intramitochondrial crystalline inclusions on pre-perfusion biopsy. Conclusion: Liver allografts from patients with primary non-function have significant mitochondria1 ultrastructural changes on pre-perfusion biopsies, and may thus have some intrinsic mitochondria abnormalities. Introduction: The ideal liver allocation system would allocate donated livers to patients who are most likeIy to die without a transplant, but who have the lowest predicted mortality once they have been transplanted. This mode of allocation assesses the condition of both the recipient and the donor at the time of transplantation. In this study we have constructed a self-organising map (SOM: this is a neural network or non-linear mode of decision analysis suitable for modelling complex multidimensional relationships) and then validated it by using the SOM to classify survival in an unrelated population. Patients and Methods: We have previously described a SOM consisting of 72 (50 recipient and 22 donor) factors (inputs) constructed from 827 (449 male; median age 52 years; median MELD score 16) consecutive primary liver transplants undertaken between 01/01/1993 and 31/07/2002 in The Queen Elizabeth Hospital, Birmingham, UK. Using a 3 neuron version of this SOM with three output functions (patient survival at 3 months; 1 year and 5 years) we used the SOM to classify the post-transplant survival of all primary graft recipients (n=200 patients) over a 5 year period from a North American centre (125 male; median age 48 years; median MELD score 16). The Birmingham (UK) patients were first classified by the SOM and then the probabilities of survival (p,) at 3 months, 1 year and 5 years calculated by dividing the number of surviving patients by the total number of patients in neuron i. Using this SOM, each of the Wisconsin (US) patients was classified to each of the neurons and the distribution of survival probes for each neuron compared between the two populations using the chi squared test. Results: There was no significant difference in the survival probabilities of patients in each neuron when the Wisconsin population was compared with the Birmingham population. Thus, the Birmingham SOM was able to classify successfully the survival of the Wisconsin patient population at 3 months, 1 year and 5 years following transplantation by using the same donor and recipient factors ( Table 1) . Analysis of the inter-neuronal survival probabilities demonstrated that for 3 month and 1 year survival, neuron 3 was associated with a significantly lower survival probability than neurons 1 and 2 (p=0.006 for 3 months; p= 0.006 for 12 months). Further, if recipients from neuron 1 received livers from patients classified to neuron 3 by "computer simulated transplantation", a proportion of patients then had a lower predicted survival probability (by SOM re-distribution to neuron 3). This analysis indicates that patient survival post-transplantation is significantly infuenced by both the condition of the recipient and donor factors at the time of transplantation. Conclusions: (1)SOM analysis provides an efficient and automated method for assessing the probability of survival of individual patients in an unrelated population at 3 months, 1 year and 5 years following liver transplantation. (2)The model not only assesses the pre-transplant condition of the patient, but also considers a wide range of donor factors when predicting the probability of survival post-transplant. (3)This unique resource can match a single liver to a population of recipients most likely to die without a transplant whilst ensuring the best possible post-transplant survival for the most suitable recipient (4)SOM analysis can also assess the probability of survival of individual recipents matched to a range of possible donated livers when they are being considered for transplant programmes. Purpose of study: To assess the efficacy of targeted prophylaxis in patients identified to be at highest risk of IF1 after orthotopic liver transplantation. Introduction: Historically, invasive fungal infection (In) has complicated up to 20% of cases of orthotopic liver transplantation (OLT). Retrospective analysis has identified a variety of risk factors associated with a high risk for the subsequent development of IFI. These include patients undergoing retransplantation, transplantation for fulminant hepatic failure (FHF), requiring haemodialysis and prolonged intensive care unit (ICU) stay. Prophylactic anti-fungal therapy is safe and can reduce the incidence of IFl in OLT recipientri. In recent years our unit has moved towards targeting prophylaxis to higher risk transplants. We assessed the efficacy of our policy in reducing the incidence of invasive fungal infection. Methods: A retrospective audit was conducted comparing two groups of adult OLT recipients over two 5 year time periods, 1990-1994 when targeted prophylaxis was not used (Group 1) and 1997-2001 when targeted prophylaxis was used (Group 2). Data were collected with respect to risk factors for IFI, anti-fungal prophylaxis use and development of IFI. Results: There was no difference in the overall number of risk factors for IF1 associated with OLT between the two groups. However, a higher proportion of patients in Group 2 had high risk factors for IF1 compared to Group 1 (p=0.07). There was a significant difference in the use of targeted anti-fungal prophylaxis given to high risk patients in Group 2 compared to Group 1 (52.3% cf 23.8%, p=0.03), and there was a significant reduction in the number of IFIs in Group2 compared with Group 1 (4.5% cf 28.6%, Conclusion: A policy of targeting anti-fungal prophylaxis to highest risk liver transplant recipients leads to a significant reduction in IF1 this group. There remains a background incidence of infection in low risk or long term recipients in whom prophylaxis is not given. Background: Only few studies have described risk factors associated with cellular rejection. The diagnosis of cellular rejection requires histology. There are no data evaluating the absence of rejection in a protocol biopsy population. Aim: To assess predictive factors associated for the absence of cellular rejection in a protocol liver biopsy population and to evaluate the usefulness of protocol liver biopsies. Method 449 consecutive patients transplanted on a data-base at our centre. Protocol liver biopsies were performed between 5 and 10 days post transplantation and then when clinically indicated, during the first 3 months. Rejection was scored prospectively. The following variables were examined with respect to absence of rejection over 3 months and in the first protocol liver biopsy: a) donor factors: age, gender, race, donorlrecipient gender match, ABO match. b) pre-transplantation recipient factors: age, sex, aetiology, ascites, oesophageal varices, TIPS, encephalopathy grade, renal support, ventilation, total bilirubin, albumin, INR, AST, ALT, creatinine, urea, c) graft and surgical factors: surgeon's visual assessment of graft, cold ischaemic time, use of venovenous by-pass, intraoperative blood transfusion and initial maintenance immunosuppression of either cyclosporin or tacrolimus in triple-dual or monotherapy. Standard treatment of rejection was 1 g iv daily of methylprednisolone x 3 days. Results: Absence of rejection over 3 months was in 69 (15%), mild in 103 (21%) and moderatelsevere in 277 (62%). In the first biopsy: absence of rejection in 92 (21%), mild in 178 (39%), moderate in 159 (35%) and severe in 20 (4.5%). Univariate analysis showed for both first protocol biopsies and 3 months, that 4 variables were significantly correlated with non rejection: pre-transplantation use of renal support (haemodialysis or hemofiltration), (p=0.042), cold ischaemic time >15 hours (p=0.030), suboptimal graft visually (p=0.042) and blood transfusion <= 7 units Cp=0.008). No correlation was found between aetiology of liver diseases, initial maintenance immunosuppression. Conclusion: a suboptimal graft with prolonged cold ischaemic time in recipients with previous need of renal support seems to be associated with absence of rejection. These data are in contrast with previous reports. These factors may be helpful in identifying patients who do not need to be submitted to protocol liver biopsy. Tables 1 and 2 1 child developed abnormal transaminases and was found to have chronic hepatitis on liver biopsy and was treated with steroids. 2 required adjusting Cyclosporine dosage to maintain trough levels in the identified range (range 65-90 microgramlL as per protocol). Summary: There was good correlation between the trough levels taken on 2 occasions in the stable paediatric post liver transplant group of patients. The range of C2 peak levels were also similar suggesting good bioavailablity Conclusion: This preliminary study on long term post transplant recipients suggests that a peak C2 level is within the range of 280-460 microgramll. Background Hepatocytes and cholangiocytes release substantial amounts of adenosine triphosphate into bile, where it is rapidly degraded by membrane-bound ecto-ATPase and 5'-nucleotidase. This degradation process leads to the generation of adenosine and inorganic phosphate (I?#. Whereas adenosine is reabsorbed in a sodium-dependent manner back into hepatocytes, little is known about the fate of biliary PI. In rat, biliary PI concentration is 0.01 mM, which is about 100 fold lower than in hepatocytes (1 mM) and 200 fold lower than in plasma (2.3 mM)9 indicating active reabsorption of PI from bile canaliculi andlor from the biliary tree. Aim: The purpose of the present study was to functionally characterize canalicular P, reabsorption in rat liver and to identify the involved P, transport system(s). Methods: P, transport was determined in isolated canalicular liver plasma membrane (cLPM) vesicles using a rapid filtration technique. Identification of putative P, transporters was performed with Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) from rat liver total mRNA using sodiumlphosphate cotransporter specific primers for NaPi-IIb (GGGATTGGGAAATTCATCCTI TTCCAACACAAGGTTGGTCA), NaPi-111 subtype PiT-1 (CATCTCGGTGGGATGTGCITGTTGCTCTCTCCTCCTTCA) and NaPi-I11 subtype PiT-2 (GCTCTACCATTGGCTTCTCGlACA-GAGGAAGTGCCTGGAGA)3. On the protein level, phosphate transporter expression was confirmed by Western blot analysis in isolated basolateral (bLPM) and canalicular (cLPM) rat liver plasma membrane vesicles. Specific polyclonal antibodies were raised in rabbits against antigenic peptides from mouse NaPi-IIb and human NaPi-IIIlPiT-2 showing >88% identity with rat homologues. Results: Transport studies in isolated cLPM vesicles demonstrated sodium-dependent P, uptake (NaJut > NaG 3877 pmollmin; K&t > KG 107 pmollmin). Initial Na+-dependent P, uptake was linear for at least 6 sec and exhibited a clear overshoot indicating transient intravesicular concentration of P, (secondary active P, transport). Initial P, uptake rates (4 sec) were saturable with increasing P, concentrations and exhibited an apparent K,,, value of -11 kM. Furthermore, sodium-dependent P, transport was higher at an acidic (pHout = 6.5; pH,, = 7.4) as compared to an alkaline extravesicular pH (8.0). In addition, an intravesicular negative membrane potential stimulated sodium-dependent P, transport indicating a Na:P, stoichiometry of > 1. These data are comparable with the transport characteristics of sodiumlphosphate cotransporters NaPi-IIb, NaPi-IIIlPiT-1 and NaPi-IIIlPiT-23. mRNAs of all these three NaPi's were found to be expressed in rat liver by RT-PCR. However, on the protein level only NaPi-IIb was found to be expressed selectively in cLPM, whereas NaPi-IIIlPiT-2 was detected in bLPM. No clearcut localization of NaPi-IIIlPiT-1 protein could be obtained. Conclusions: The canalicular membrane of rat hepatocytes localizes the sodiumlphosphate cotransporter NaPi-IIb, which can reabsorb P, from primary hepatic bile back into hepatocytes. In contrast, NaPi-IIIlPiT-2 was found to be expressed at the basolat-era1 hepatocyte plasma membrane. The results indicate that NaPi-IIb regulates P, concentration in bile and may play an important role in the overall P, homeostasis in rat liver BackgroundlAims. Organic anion transporting polypeptides (OATPs) are a family of transport proteins of the basolateral hepatocyte membrane. Human OATP8 (SLC21A8) is predominantly expressed in hepatocytes and is an uptake system for xenobiotics such as digoxin. The OATP8 gene promoter possesses an inverted repeat (IR1) element at nt -82l-70 relative to the transcription start site, that binds and is activated by the farnesoid X receptorlretinoid X receptor (FXRlRXR). Ligands of FXR include bile salts such as chenodeoxycholic acid (CDCA), previously shown to activate the OATP8 gene. However, because OATP8 is only a poor bile salt but an efficient xenobiotic transporter, we investigated whether xenobiotics such as rifampicin regulate the OATP8 gene. Rifampicin is a prototypic ligand of the xenobiotic receptor PXR (pregnane X receptor). Methods. Endogenous OATP8 mRNA levels in Caco2 cells were quantified by real-time PCR. An OATP8 gene promoter construct containing nucleotides -120l +38 relative to the transcription start site (Luc-120) was assayed for reporter activity in transiently transfected cells. The FXR binding site in the OATP8 gene (IR1 element) was characterized using the luciferase construct IR1-TK-Luc that contained a thymidine kinase promoter under the control of the IRl element. Results. Endogenous OATP8 mRNA levels in Caco2 cells were induced 7fold by CDCA (100 pmolll). Interestingly, incubation of cells with rifampicin (10 pmolll) resulted in a 4.5fold increase in OATP8 mRNA levels. To study the mechanism of rifampicinmediated induction of OATP8, the promoter sequence was searched for potential PXR binding sites. Because the IR1 element, previously shown to bind the FXRlRXR heterodimer, was the only nuclear receptor binding site found, we hypothesized that the induction by rifampicin could be mediated through the FXR element. We, therefore, studied the effect of CDCA and rifampicin on the IR1 element. In cells cotransfected with the IR1-TK-Luc construct and expression plasmids coding for FXRlRXR, PXRlRXR or CAR (constitutive androstane receptor)lRXR, FXRlRXR induced the IRl element llfold in the presence of CDCA and 2.5fold in the presence of rifampicin, indicating that rifampicin is capable of activating FXR. In contrast, PXRlRXR or CARlRXR did not activate the IR1 element in the presence of CDCA or rifampicin. To confirm that an OATP8 promoter construct is also activated by rifampicin, Huh7 cells cotransfected with the Luc-120 construct and FXRlRXR expression plasmids were incubated with CDCA or rifampicin. CDCA induced OATP8 promoter activity -Zfold, rifampicin -15fold, confirming that rifampicin transactivates the OATP8 promoter. To investigate whether other xenobiotics also activate the FXR element, Huh7 cells cotransfected with the IR1-TK-Luc construct and FXRlRXR expression plasmids were incubated with rifampicin, rifamycin, RU486, clotrimazol, phenobarbital or PCN. A 1.4fold (rifamycin) to 2.5fold (clotrima-201) induction of the IR1 element was seen in the presence of these xenobiotics. Conclusions. The human OATP8 gene is induced transcriptionally by xenobiotics that represent prototypic ligands of the xenobiotic receptor PXR. However, activation does not occur through PXR but through activation of FXR bound to the IR1 element in the OATP8 gene promoter. The data thus indicate that PXR ligands such as rifampicin and clotrimazol can also activate FXR and thereby induce the FXR-regulated transporter gene OATP8. Disclosures: May-Britt Becker -No relationships to disclose Michael Fried -No relationships to disclose Diana Jung -No relationships to disclose Gerd A Kullak-Ublick -No relationships to disclose Peter J Meier -No relationships to disclose Epidemiologie de Population EPI 106 Seema Sonnad -No relationships to disclose 428 PRESERVATION OF CANALS OF HERING MITIGATES BILE DUCT LOSS IN LIVER GRAFT REJECTION. Ivlarius C van den Michael Angelis -No relationships to disclose Jeffery Cooper -No relationships to disclose Erick Edwards -No relationships to disclose Richard B Freeman Jr -No relationships to disclose Ann Harper -No relationships to disclose Abigail Mithoefer -No relationships to disclose No relationships to disclose the data was not available. The CA 19-9, CEA, and AFP levels were obtained using standard commercially available assays Results: Out of a total of 108 patients with ESLD fifty-eight patients fulfilled the inclusion and exclusion criteria. Of these, thirty-three patients had evidence of ascites and twenty-five did not. The etiology of liver disease in the two groups was similar. The mean levels of CA 19-9, CEA and AFP levels in patients with ascites were not significantly different from those without ascites (CA 19-9 48.92 % 65 Furthermore, 15 out of these 58 patients with ESLD had levels -> 37 Ulml, the upper limit of normal in this assay Unitslml] vs 6.53 5 6.13 Conclusions: Ascites does not seem to make an impact on the serum levels of CA 19-9 Chhaya Hasyagar -No relationships to disclose Savant Mehta -No relationships to disclose 445 SEVERE MITOCHONDRIAL TOXICITY AFTER LIVER TRANSPLANTATION IN HIV-HCV COINFECTED PATIENTS Liver transplantation (LT) may be the only potentially curative treatment available to these patients at this stage. However its feasibility and benefit has still to be established. Severe recurrence of hepatitis C on the liver graft may complicate the post operative course as recently suggested by us and others. Mitochondrial toxicity of highly active antiretroviral therapy (HAART) may also play an active role on the liver graft of HtV-HCV co-infected patients. We aimed to study this complication on the liver graft of HIV-HCV coinfected patients. Patients and Methods: Between He had an history of episode of pancreatitis and HAART therapy was AZT-3TC-Nelfinavir. At M10 post LT, microvesicular steatosis was noted. At M18 a low content of liver mtDNA was found (mtDNAlnuclear DNA = 0.28). 3 other patients had microvesicular steatosis respectively at M2, M4 and M6 post LT. One of these 3 patients had low content of liver mtDNA (mtDNAlnuclear DNA = 0.20). Severe defect of complex IV of the respiratory chain was noted in 1 of these 3 patients. No deletion of mtDNA was observed by southern blot or long range PCR Conclusion: Mitochondrial toxicity on the liver graft may become a major problem during post LT c o m e and could worsen graft lesions related to HCV recurrence on the liver graft in HIV patients Daniel Azoulay -No relationships to disclose Henri Bismuth -No relationships to disclose Denis Castaing -No relationships to disclose Duclos-Vallee -No relationships to disclose Cyrille Feray -No relationships to disclose Michelle Gigou -No relationships to disclose Catherine Guettier -No relationships to disclose Philippe Ichai -No relationships to disclose Claude Jardel -No relationships to disclose Anne Lombes -No relationships to disclose Bruno Roche -No relationships to disclose Faouzi Saliba -No relationships to disclose Didier Samuel -No relationships to disclose Elina Teicher -No relationships to disclose Daniel Vittecoq -No relationships to disclose 381A 457 IDENTIFICATION OF SODIUMlPHOSPHATE COTRANSPORTER TYPE IIB Marek Nowicki, USC, Los Angeks, CA; Jorge Rakela, Tomasz Laskus, There is growing evidence that patients with chronic hepatitis C are more likely to have significant changes in their physical and mental well being, commonly manifested as fatigue and depression, than patients with liver disease of other etiology. Recently published studies demonstrated also that HCV infection is associated with cognitive dysfunction. Hepatitis C virus (HCV) was reported to replicate in monocyteslmacrophages and lymphoid cells. We have recently demonstrated that leukocytes carry HCV across the blood-brain barrier and we found HCV RNA in the central nervous system (CNS) in some infected patients (J Virol 2002, 76, 10064-10068; 76, 600-608) . However, biological basis for neurocognitive abnormalities observed in HCV-infected patients remains unclear. AIM: To determine the pattern of gene expression in CNS in HCV-positive patients as compared to HCV-negative controls. MATERIAL AND METHODS We analyzed samples of brain tissue obtained at autopsy from 3 HCV-positive patients and 3 HCV-negative control patients. All were men of similar age, none was infected with HIV. All 3 HCV+ patients and 2 out of 3 controls had liver cirrhosis. Only 2 deaths (one in each group) were liverrelated. The analysis of gene expression was conducted using three different techniques: differential display (GenHunter Inc), reverse Northern, and microarray analysis (BD Atlas Plastic Miroarray). Reverse Northern analysis was used for confirmation of differential display findings. Analysis of microarray data was done using Atlas Image 2.01 (BD) and Cluster 2.20 and TreeView 1.60 (M.Eisen; UCB). Only those genes that were up or downregulated 1.8 times in reverse Northern and/or microarray analysis were considered differentially expressed. RESULTS: The most striking finding was downregulation of mitochondria] oxidative phosphorylation genes in all HCV-infected patients as compared to controls; impairment of brain oxidative/ energy metabolism has been previously suggested to be the proximate cause of many disorders that impair mentation. Another consistent finding in differential display and microarray analysis was downregulation of multiple ribosomal proteins genes and several genes involved in transcription regulation. These could indicate reduced metabolic activities perhaps secondary to deficiencies in oxidative phosphorylation. We also observed upregulation of several genes involved in immune response. There was upregulation of MHC class I and Class I1 and several lymphokine receptors like interferon (alpha, beta), IL-6, IL-9, IL-17 as well as LFA-1 ligand intracellular adhesion molecule ICAM-2 and CEACAM1. Furthermore, upregulation of CD8O and allograft inflammatory factor-1 gene (AIF-1) suggested the presence of activated microglia cells and/or activated macrophages. CONCLUSIONS: We found downregulation of oxidative phosphorylation genes and upregulation of genes involved in immune response in brains from HCV-positive patients when compared to HCV negative patients. Our findings provide the likely substrate for neuropsychiatric symptoms and cognitive impairment associated with HCV infection. Medicine, Ehime, Japan; Emmett V Schmidt, Raymond T Chung, Massachusetts General Hospital, Boston , M A Background/Aims: We previously reported cell-based HCV replication using a novel binary expression system in which cells were transfected with a T7 polymerase-driven full length HCV cDNA plasmid (pT7-flHCV-Rz) and infected with vaccinia-T7 (PNAS 989847). To circumvent vaccinia-induced cytotoxicity, we sought to determine whether replication-defective adenoviral vectors expressing T7 or cell lines stably transfected with T7 could support HCV replication. Because vaccinia interferes with PKR function, we further sought to define the antiviral activity of interferon-ar (IFN) in these models. Methods: 24 hours after transient transfection of CV-1 and Huh7 cells with pT7-flHCV-Rz, cells were treated with recombinant replication-defective adenovirus vectors expressing T7 polymerase (Ad-T7pol). Medium with or without 1000 IUlml of IFN was changed at day 1 post-infection and every 2 days thereafter. Cells were harvested on day 1, 2, 3, 5, 7 and 9 post-infection. For T7-stably transfected cell lines (Huh-T7), we transiently transfected with pT7-flHCV-Rz, then added IFN to selected cells in analogous manner. HCV positive and negative strand were measured by strand specific real-time RT-PCR. HCV core protein expression was assessed by Western blotting. Results: In the HCV replication system with recombinant adenovirus vectors and with T7 stable cell lines, no cytotoxicity was observed at 9 days. With pT7-flHCV-Rz and Ad-T'lpol, HCV positive and negative strand RNA expression was strongest in the first 3 days post-infection and diminished thereafter, but were expressed throughout the 9 days. In the 2 days after adding IFN, HCV positive strand was significantly decreased (0.48520.213 vs. 0.162+0.017 in CV1, 0.123t0.051 vs. 0.035?0.012 in Huh7, HCV/ GAPDH copy ratio, pC0.05) than the samples without IFN. Sustained expression of HCV RNA and IFN inhibitory effect was also observed in T7-stable Huh-T7 cell lines. PKR expression was strongly induced by IFN, suggesting that PKR is appropriately induced by IFN. Conclusions: We have successfully substituted recombinant adenovirus vectors or T7 stable cell lines in our cell-based binary HCV replication system. In these systems, IFN inhibits HCV replication, and upregulates the PKR pathway. These improved binary systems are a durable and more authentic model for identification of host cellular processes critical to HCV replication. Disclosures:Jason Blackard -No relationships to disclose Raymond T Chung -No relationships to disclose Yoichi Hiasa -No relationships to disclose Norio Horiike -No relationships to disclose Yoshitaka Kamegaya -No relationships to disclose Morikazu Onji -No relationships to disclose Emmett V Schmidt -No relationships to disclose The MELD scoring system for the allocation of cadaveric donor liver for transplantation gives significant priority to cirrhotic patients with a small, T1 or T2, hepatocellular carcinoma. Patients on the wait list were given adjusted MELD scores equivalent to 3-month mortality rates of 15% and 40% for T1 and T2 lesions, respectively based on theoretical tumor doubling times. Frequently these patients have compensated cirrhosis and hence their intrinsic MELD score is lower, yet they are not candidates for surgical resection secondary to the underlying cirrhosis. To assess whether the additional MELD points for HCC is valid, we sought to examine the impact of the MELD scoring system on a single center's waiting list one year after implementation.Methods: Records of all patients undergoing liver transplantation (OLTx) at UTHSC-San Antonio from Feb 27,2002 through Feb 27,2003 were retrospectively reviewed. Pathology and radiology reports as well as data from UNET were collated for analysis and review. Results Over a 1-year time period following implementation of the MELD system, 126 patients underwent OLTx at our center. Of these 24 were excluded from analysis because of the following reasons: Status 1 (lo), living-related recipient (6), pediatric non-tumor (8). The table below summarizes the pertinent findings. Of the 102 remaining patients, 36 (35%) underwent OLTx with suspected HCC (4 Tl, 32 T2).None of the HCC patients had their MELD score downgraded because of tumor progression (T2-T3). At transplant, 2 patients were found to have extra hepatic disease(one with known HCC and the other unknown) and the cases were aborted. Sensitivity of radiological evaluation for the presence of HCC was 92.5% and specificity 94%. The calculated MELD score at the time of listing for OLTx and at the time of OLTx was significantly lower for those patients thought to have an HCC. Exception points for HCC gave those patients a significantly higher average MELD score at the time of OLTx than patients without HCC. Patients wlo HCC had a Aupper; True MELD of +3.1 at the time of OLTx compared to listing as opposed to HCC patients whose A-upper; True MELD was -0.3. During the study period 34 patients (none with suspected HCC) died on the waiting list or were too sick to transplant while 215 new patients were listed compared to 24 dying the prior year with 163 new registrants (15.8% vs 14.7%). Mean MELD score of those dying during the study period was 22.6 8 (median 22.5) and mean time on the list was 284 421 days (median 99 days). Of those patients dying. 16 had a MELD score of 1 2 4 . Conclusion: MELD has effectively prioritized cadaveric donor livers to the sicker patients. However, patients with HCC are given too much priority as evidenced by all of the wait list deaths occurring in patients without HCC and lack of tumor progression (to T3) sufficient to loose MELD exception points in those with HCC. Likewise, the mean MELD score of those dying was less than that given to patients with T1 lesions(22.6 vs. 24). The recent downward adjustments of MELD exceptions to 20 and 24 for TI and T2 lesions, respectively, are reasonable. Ongoing analysis will be needed to accurately place HCC patients in the current allocation system. There was no significant difference (p>0.05) in weight gain between the sexes, those with a BM1>30 pre-transplant and those with a BMI<30 pre-transplant or those on prolonged courses (>3 months) of steroids. Weight gain was significantly greater (p<0.05) in patients aged over 50 compared to those under 50 and those transplanted for chronic liver disease compared with fulminant liver failure. A pre-transplant BMI>30 was a strong indicator that the patient would still have a BMI>30 at 3 years. There was no effect of immunosuppression on weight gain: corticosteroids were discontinued in 59.3% by 3 months and long term use of steroids was not associated with weight gain. Weight gain was similar in those on cyclosporine and on tacrolimus.Conclusions: A BM1>3O is common in the liver transplant population, but seems to unrelated to specific immunosuppression and may be more closely linked to lifestyle. Most weight gain occurs after the first 6 months, and intervention with dietary and lifestyle advice at this point could be implemented to minimise the long-term morbidity and mortality risks associated with obesity. Serum tumor marker estimations are often performed as a part of evaluation for liver transplantation. There is a paucity of information on the effect of ESLD on the levels of these tumor markers making interpretation difficult One tumor marker, CA-125, has been consistently reported to be elevated in patients with ascites from liver disease. Purpose:-To compare the levels of serum CA 19-9, CEA and AFP in patients with ESLD with or without ascites.-To compare the serum CA 19-9 levels in patients with ESLD vs normal controls Methods: All patients referred for liver transplantation between Jan 2000 and Dec 2002 at our center were included in the study. Men with end stage liver disease (ESLD) develop a plethora of debilitating symptoms, including severe muscle wasting, that render many of these patients non-suitable for liver transplantation. ESLD-associated hypogonadism, i.e. testosterone deficiency, may, in part, be responsible for the development and progression of many of these incapacitating symptoms. In general, hypogonadism can be effectively treated with topical testosterone replacement (TTR), which avoids the first pass effect via the hepatic system in contrast to oral anabolic steroids. TTR has been shown to effectively improve muscle mass and overall well being in patients with HIV. However, little is known about the effects of TTR in men with ESLD. The Aim of this study was to determine the potential benefits and safety of TTR in patients with ESLD and muscle wasting. Method:The medical records of liver transplant patients treated for at least 3 months with testosterone gel 1% (5 grams per day) therapy for muscle wasting (MW) from January 2002 to March 2003 were reviewed. Information collected included; demographics, albumin, pre-albumin, transferring, testosterone, estrogen, LH and FSH. Drug safety results were also collected and included, acute cellular rejection, cholestasis, malignancy, and patient or graft loss. Tumor markers for AFP, CA19-9, CEA, PSA, prolactin levels and radiographic imaging was reviewed to rule out and potential malignancy. Patients not treated with testosterone but with a similar clinical setting were compared (Group 2). Results: Thirteen patients were identified with ESLD and MW, group 1 (n=8), group 2 (n=5). Demographics were as follows: (group 1) 6 males, 2 females, and mean age 56 years; (group 2) 4 males, 1 female, mean age 53 years. Disease etiology for group 1: (HCV (n=5), HCV/HIV (n=l), AIH (n=l), Cryptogenic (n=l)) group 2: HCV (n=4), HCVILaennec's (n=l). Most patients in group 1 stated a subjective improvement in muscle strength, and overall wellbeing. In group 1, serum albumin levels increased from day 1, 30, 90 (2.18/2.7875/3.45 g/dL, respectively) while those in group 2 did not (2.30/2.16/2.0 g/dL) (p= 0.0002). Similar results were seen for prealbumin (days 1, 30, 90) group 1 (10/18/22 mgldL) versus group 2 (10/8/6 mg/dL). In group 1, 7/8 patients received a liver transplant, while 1 remains listed for OLT. In contrast, 3/5 patients in group 2 were transplanted while 2/5 died. Two patients in each group were not listed at the time of evaluation due to extreme deconditioning. Of those in group 1, one patient was transplanted and another patient remains listed for OLT with an albumin greater than 3.5 gldL. This patient presently has no radiographical evidence of ascites. The two patients of group 2, who were not listed for LT due to extreme deconditioning, died on day 28 and 35, respectively. There was no rise in the tumor marker elevations in any of the patients. After transplantation, no patients in either group suffered from rejection or graft failure. Ascites accumulation, as assessed by abdominal ultrasound imaging, appeared to be less in patients receiving TTR (group 1) compared to all 3 living patients in group 2, who had persistent ascites. One patient of the treatment group developed marked cholestasis 10 days after starting TTR and received a liver transplant 14 days later. Prior to starting TTR, this patient underwent weekly large volume paracentesis and suffered from marked deconditioning and recurrent bouts of hepatic encephalopathy while maintaining a MELD score of 8-9 points over a 6-month period. Conclusion: Our preliminary data suggest that TTR increases muscle strength (subjective measures), stimulates albumin synthesis and improves the outcome of OLT. Thus, TTR (Testosterone gel 1%) appears to be of great benefit in patients with ESLD and muscle wasting. Further studies are needed to determine the efficacy and safety of TTR in patients with ESLD. Percutaneous ethanol injection (PEI), radiofrequency tumor ablation (RF), and laser thermal ablation (LTA) are percutaneous techniques used in the treatment of unresectable HCC. Percutaneous seeding of neoplastic cells along the needle track is a rare complication of these tecniques, and it has been recently suggested to consider with caution these techniques as a "bridge" treatment in cirrhotic patients with HCC candidates to LT. The aim of this study was to verify whether percutaneous ablation treatments represent a risk factor for HCC recurrence in these patients. During the period 1990 -2002,32 patients (mean age 53 yrs) with HCC complicating liver cirrhosis previously treated with percutaneous ablation treatments underwent LT in 3 Italian Centers. Among the 32 patients, 13 had monofocal HCC, 19 multifocal HCC. In the latter group, 5 had monolobar disease, 14 had bilobar disease. Considering the 13 patients with monofocal HCC, 7 were treated with RF (associated with transarterial chemoembolization [TACE] in one case), 4 with PEI (associated with TACE in one case), 2 with RF and PEI (associated with TACE in one case). Among the 19 patients with multifocal HCC, 9 were treated with RF (associated with TACE in four cases), 9 with PEI (associated with TACE in four cases), 1 with LTA associated with TACE. On the whole, 38 nodules were treated in the 32 patients. Furthermore, 31 additional untreated HCC nodules were found at pathological examination of the explanted liver. Among the 19 RF-treated nodules, 6 showed total necrosis, 12 partial necrosis, 1 absence of necrosis. Among the 15 PEI-treated nodules 4 showed total necrosis, 5 partial necrosis, 6 absence of necrosis. The 2 nodules treated with RF and PEI showed total necrosis in one case and partial necrosis in the other one while the 2 nodules treated with LTA, showed partial necrosis. The mean follow-up period was 35 months (range 3-130 months). Five patients died during the follow up and the actuarial survival rate was 94% at one year, 85% at 3 and 5 years. Three patients (9.4%) had post-transplant recurrence of HCC which was the cause of death in all cases; the diagnosis of tumor recurrence was made at 2 months from LT (peritoneal carcinomatosis), at 8 months from LT (bone), and at 9 months from LT (abdominal liymphnodes and bone), respectively. The 3 patients had all been treated with PEI before LT and no one of them showed recurrence of HCC at the abdominal wall level. According to our results, percutaneous ablation techniques performed before LT do not seem significantly affect both survival and tumor recurrence rate in patients transplanted because of HCC complicating liver cirrhosis. In particular, no cases of recurrence at the abdominal wall level were observed in the overall series and no one case of HCC recurrence was observed in patients treated with RF alone or in combination. Background: Transjugular intrahepatic portosystemic shunt (TIPS) is valuable in the management of portal hypertension (PHTN) in patients awaiting liver transplantation (OLT). Recurrent PHTN after OLT can be refractory to medical management and portosystemic shunting may be considered in rare situations, either as a bridge to retransplantation or as definitive therapy. In this report we review our experience and outcomes with TIPS after OLT. Methods: Records of 309 primary adult OLT recipients, between January 1995 and December 2003, were retrospectively reviewed. Evidence of refratory post-OLT PHTN was noted. Those who required TIPS were the subject of this review. Demoraphics, indications for OLT and TIPS, evidence of allograft dysfunction and outcomes after TIPS were described. Results: During the study period 6 TIPS were placed in 6 patients at a mean of 13.5 months after OLT (range 2-36 months). There were 3 males and 3 females, age 53.0L7.8 years. Hepatitis C was the primary indication for OLT in 5 and primary biliary cirrhosis in 1. Indications for TIPS included refractory ascites (6), variceal bleeding (2) and various degree of associated hepatic vein outflow stenosis (3). Five patients had resolution of PHTN and 1 patient with refractory ascites had severe hepatic vein outflow stenosis and associated hepatitis C in the allograft. Two patients required re-OLT for recurrent hepatitis C. There were 3 deaths: liver failure 1 month after TIPS done in the setting of allograft dysfunction, 3 months after TIPS with subsequent re-OLT and organ failure, and lung cancer 5 months after TIPS. Bridging fibrosis was present in 3 patients, 2 needed re-OLT and 1 died from liver failure while waiting for OLT. Currently, 3 patients are alive without evidence of PHTN 3, 9 and 58 months after TIPS. Conclusions: 1-TIPS effectively and safely controls PHTN after OLT. 2-TIPS in setting of a moderate to severe allograft dysfunction does not abrogate the need for re-OLT and can be associated with a high mortality. 3-Timely re-OLT should be considered in the presence of fibrosis from recurrent HCV and PHTN. 4-TIPS has been successful in the setting of mild-moderate hepatic vein outflow stenosis and PHTN. Introduction: Biliary tract complications after liver transplantation (LT) are reported to occur in 13% to 35% of patients. The frequency of biliary infections is not well documented. Therefore, we prospectivly obtained bile samples during diagnostic and or therapeutic endoscopic retrograde cholangiography (ERCP) in all liver transplant patients from 0112001 to 1112002. Methods: In 67 patients (mean age 53.6 years, range 24-73 years) a total of 172 ERCP's were performed (2.57 ERCFlpatient). All patients had a lumen adapted, end-to-end biliary anastomosis. LT was performed between 3.2 months and 2.8 years before the intervention. Cholangitis was defined as the presence of cholestasis with clinical and biochemical signs of infection not explained otherwise. Only in 19 cases of 172 interventions, positive culture results were combined with clinical signs of cholangitis. The follwoing risk factors were identified stenosis of any type, plastic endoprothesis, choledocholithiasis and previous papillotomy. Patients with these risk faktors had significantly higher incidence of positive bile cultures (77.6% vs 40%, p