key: cord-0719832-5lrpo7g6
authors: Schaal, Justin B.; Tran, Dat Q.; Subramanian, Akshay; Patel, Reshma; Laragione, Teresina; Roberts, Kevin D.; Trinh, Katie; Tongaonkar, Prasad; Tran, Patti A.; Minond, Dmitriy; Fields, Gregg B.; Beringer, Paul; Ouellette, André J.; Gulko, Percio S.; Selsted, Michael E.
title: Suppression and resolution of autoimmune arthritis by rhesus θ-defensin-1, an immunomodulatory macrocyclic peptide
date: 2017-11-16
journal: PLoS One
DOI: 10.1371/journal.pone.0187868
sha: 848c234214cbc5eeb1d64a043ff097e32e8ebff0
doc_id: 719832
cord_uid: 5lrpo7g6

θ-defensins constitute a family of macrocyclic peptides expressed exclusively in Old World monkeys. The peptides are pleiotropic effectors of innate immunity, possessing broad spectrum antimicrobial activities and immunoregulatory properties. Here we report that rhesus θ-defensin 1 (RTD-1) is highly effective in arresting and reversing joint disease in a rodent model of rheumatoid arthritis (RA). Parenteral RTD-1 treatment of DA/OlaHsd rats with established pristane-induced arthritis (PIA) rapidly suppressed joint disease progression, restored limb mobility, and preserved normal joint architecture. RTD-1 significantly reduced joint IL-1β levels compared with controls. RTD-1 dose-dependently inhibited fibroblast-like synoviocyte (FLS) invasiveness and FLS IL-6 production. Consistent with the inhibition of FLS invasiveness, RTD-1 was a potent inhibitor of arthritogenic proteases including ADAMs 17 and 10 which activate TNFα, and inhibited matrix metalloproteases, and cathepsin K. RTD-1 was non-toxic, non-immunogenic, and effective when administered as infrequently as once every five days. Thus θ-defensins, which are absent in humans, have potential as retroevolutionary biologics for the treatment of RA.

Rheumatoid arthritis (RA) is a systemic autoimmune disease that affects 0.5 to 1% the world population [1] . RA is typically characterized by symmetric polyarthritis involving the extremities with erosive joint changes mediated by hyperplastic synovium (pannus) and chronic a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 #AR068833 (DQT) https://www.niams.nih.gov/. The Wright Foundation (PT) https://research.usc. edu/for-investigators/funding/usc/#wright. The Southern California Clinical and Translational Science Institute #UL1TR000130 (MES) http:// www.sc-ctsi.org/. National Cancer Institute #P30CA014089 https://www.cancer.gov/. Seth MacFarlane Foundation. Oryn Therapeutics provided support in the form of salary for D.Q.T., but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the 'author contributions' section.

The following authors declare competing interests in relation to their affiliations with Oryn Therapeutics: Dat Q. Tran is Scientific Director), Percio Gulko is Chief Medical Officer, Michael E. Selsted is Chief Scientific Officer, Andre. J. Ouellette is a minority investor. Oryn Therapeutics has licensed technologies disclosed in US Patents 6,335,318; 6,514,727; 6,890,537; 7,119,070; 7,399,823 B1, 7,462,598, and 9,346,866 B2. Oryn has received an SBIR award from NIH to develop cyclic peptides as therapeutics for rheumatoid arthritis. The affiliation of the authors with Oryn Therapeutics does not alter our adherence to PLOS ONE policies on sharing data and materials. disease score of sixteen. Disease severity was assessed by at least two scorers and the inter-rater agreement (kappa) value was 0.806 as determined using MedCalc Statistical Software version 14.8.1 (MedCalc Software, Ostend, Belgium). In three separate trials, treatment was also blinded to scorers. There was no statistical difference between blinded and unblinded scoring of any treatment cohort. Treatment with saline (vehicle), RTD-1, or anti-arthritic agents was randomized and initiated when an animal first presented with an AS of 3 or more. RTD-1 was administered subcutaneously with doses ranging from 0.2 to 5.0 mg/kg in an injection volume of 0.5 ml at intervals ranging from once a day (qd) to once every seven days (q7d). MTX was administered at 0.25 mg/kg by intraperitoneal injection q3d [19] . Etanercept was administered at 0.4 mg/kg by subcutaneous injection q3d [20, 21] .

Limbs from euthanized animals were preserved in buffered formalin, decalcified, embedded, sectioned, and stained with hematoxylin and eosin (H & E) in the USC Research Model Pathology Core. Microscopic images were acquired with a Nikon-Microphot FXA with SPOT 5.1 image capture software.

Eight to 12-week old male Sprague Dawley rats (Harlan) were fitted with a surgically implanted jugular vein catheter for sequential blood collection (Culex NxT Automated In Vivo Sampling System; BASi, West Lafayette, IN), and monitored daily. RTD-1 dissolved in saline was injected subcutaneously at 5 mg/kg to 4 animals. Approximately 200 μl of blood was collected before peptide administration and at 0. 25, 0.5, 1, 2, 4, 6, 8, 12, 18, 24, 30, 36, 48, 60 , and 72 h after RTD-1 injection. Blood was collected into vials containing K 3 EDTA and plasma was collected after centrifugation. Plasma concentrations of RTD-1 were determined after solid phase extraction (SPE) by weak cation exchange (Oasis WCX, Waters, Milford, MA) and quantitation by C18 RP-UPLC, photodiode array (PDA) detection, and electrospray-ionization mass spectrometry (ESI-MS). Chromatography was performed on an Acquity H-class UPLC with an analytical PDA detector using Empower 3 software (Waters). Quantitative mass spectrometry was performed on post PDA eluent using a Micromass Quattro Ultima mass spectrometer with MassLynx 4.1 (Waters). Pharmacokinetic analysis was performed using parametric population modeling with maximum likelihood estimation via the EM algorithm with sampling (MLEM) as implemented in the ADAPT 5 (Version 5.0.49) PK/PD Systems Analysis Software (Biomedical Simulations Resource, University of Southern California).

Saline (control) and RTD-1 (5 mg/kg, s.c., qd) treated PIA rats were euthanized 4 days after disease onset (AS ! 3). The ankle joints and feet were snap frozen in liquid N 2 and pulverized with mortar and pestle as described [22] . One gram of tissue powder was extracted with 5 ml of ice cold 10 mM sodium phosphate, 150 mM NaCl, pH 7.4 (PBS) containing protease inhibitor tablets (cOmplete ULTRA with EDTA; Roche Diagnostics GmbH, Mannheim, Germany). Samples were vortexed for 30 seconds and extracts clarified by centrifugation. Supernatant IL-1β was quantified by ELISA (Life Technologies, Grand Island, NY).

Fibroblast-like synoviocytes (FLS) were isolated by enzymatic digestion of the synovial tissue from DA rats with PIA as previously described [23] . Cells were washed and suspended in DMEM supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY), 30 mg/ml Lglutamine (Sigma), 250 μg/ml amphotericin B (Sigma), and 10 mg/ml gentamicin (Gibco). After overnight culture, non-adherent cells were removed and adherent cells were cultured to approximately 70-90% confluence and passaged following detachment with 0.25% trypsin-EDTA for 3 min at 37˚C. All experiments were performed with FLS after at least four passages.

FLS invasiveness was assayed in vitro using Matrigel-coated inserts in a transwell system (Becton, Dickinson and Company, Franklin Lakes, NJ) as previously described [23] . Briefly, DA rat FLS (70-80% confluent) isolated as described above were suspended at 2.0 × 10 4 cells/well in serum-free DMEM. Cells were placed in the upper compartment of the Matrigel-coated inserts to which RTD-1 was added to final concentrations of 0.1, 1, or 10 μg/ml. The lower compartment contained complete medium with 10% FBS and plates were incubated at 37˚C. After 24 hours the upper surface of the insert was wiped with cotton swabs to remove non-invading cells and the Matrigel layer. The opposite side of the insert was stained with Crystal Violet (Sigma) and the total number of invading cells was counted at 100X magnification. Experiments were performed in duplicate on seven cell lines each generated from the synovial tissue of individual DA rats with PIA.

Expression of IL-6 by stimulated FLS FLS were cultured in 12-well plates at 4-6 × 10 4 cells/well for 24 h (viability > 95% by trypan blue staining). Medium was replaced and FLS were incubated with TNFα or IL-1β at final concentrations of 3 or 10 ng/ml plus RTD-1 (10 μg/ml in 0.01% acetic acid (HOAc)) or vehicle alone. After 24 h incubation, FLS culture supernatants were removed, clarified by centrifugation, and analyzed for IL-6 by ELISA (Invitrogen, Carlsbad, CA).

FLS were seeded at 3 × 10 3 cells/well in a 96-well TC treated plate (Greiner). Cells were allowed to adhere for 18 h and medium was replaced prior to addition of peptide or vehicle. RTD-1 was added to wells at final concentrations of 0-30 μg/ml in culture medium and incubated for 0, 24, or 48 h at 37˚C in 5% CO 2 . FLS cell number was determined by DNA staining using CyQuant Cell Proliferation Assay Kit (Invitrogen) per manufacturer's instructions. Cellular toxicity was determined by measuring the release of lactate dehydrogenase (LDH) using a CytoTox 96 Cytotoxicity Assay (Promega Bio Sciences, San Luis Obispo, CA) in FLS supernatants. All assays were performed with ! 7 replicates.

RTD-1 was analyzed for enzymatic inhibition against zinc metalloproteases ADAM17, ADAM10, matrix metalloproteases (MMPs) 1, 2, 3, 8, 9, 13, and 14, and cysteine cathepsins (Cats) B, C, H, K, L, S, and V (Table 1 ). ADAM10 and ADAM17 activities were measured by release of fluorogenic peptides cleaved from pro-TNF (R&D Systems, Minneapolis, MN). ADAM10 was diluted to 0.05 μg/ml and ADAM17 to 0.1 μg/ml with 25 mM TRIS, 2.5 μM ZnCl 2 , 0.005% Brij35, pH 9.0 and incubated with 0 to 3 μg/ml RTD-1 dissolved in 0.01% HOAc. Assay was initiated by addition of substrate (10 μM final) and enzymatic activity measured at 37˚C (ADAM10) or 22˚C (ADAM17) every 30 seconds for 1 h in a SpectraMax M5e fluorometer (Molecular Devices, Sunnyvale, CA) with 320 nm excitation and 405 nm emission.

RTD-1 inhibition of Cat B (BPS Bioscience, San Diego, CA) and Cats C, H, K, L, S, and V (R & D Systems) was analyzed in a 96 well format with enzyme-specific substrates per manufacturer's instructions and as described in Table 1 . The activity of each enzyme was determined in the presence of 0 to 50 μg/ml RTD-1.

Inhibition of MMPs by RTD-1 was performed as described [24] , incubating each enzyme with 0-100 μg/ml of RTD-1. IC 50 values were calculated by GraphPad Prism 5.01 (GraphPad Software, San Diego, CA) utilizing a non-linear curve fit of log [inhibitor] versus response with variable slope.

Adult DA or Sprague-Dawley rats were challenged s.c. with 5 mg/kg RTD-1 every day (15-42 injections) , or every other day (8 injections). Several animals were boosted 2 months later, prior to serum or plasma collection. Immune responses were analyzed by dot blot on nitrocellulose membranes as described in legend to the figure.

Statistical analyses were performed utilizing GraphPad Prism 5.01. All sample populations were tested for normality using a D'Agostino and Pearson omnibus normality test. P-values were generated in parametric sample means by student's t-test while non-parametric sample medians were compared by Mann-Whitney U-test. Number of replicates and/or total number of animals are shown in figure legends or within the figures. Anti-arthritic efficacy of θ-defensin RTD-1

The anti-arthritic effects of RTD-1 were tested in DA/OlaHsd rats with PIA, a well-characterized model of chronic RA [16, 22, [25] [26] [27] . Animals with evolving PIA (arthritis severity score, AS ! 3) were scored, randomly assigned to a treatment group, and treated daily with subcutaneous (s.c.) injections of saline or 5 mg/kg of RTD-1 for 11 days (Fig 2A) . RTD-1 treatment significantly reduced arthritis progression within 24 h of the first administered dose (P = 0.0064; Fig 2A) . RTD-1 treated rats had markedly lower (69% reduction of mean) AS at the end of the observation period compared with vehicle controls (P < 0.0001; Fig 2B) . Additionally, complete resolution of disease (AS = 0) occurred in 29% (9 of 31) of RTD-1 treated animals compared to 2.5% (1 of 40) of saline treated animals (P = 2.9 × 10 −17 , Chi-squared) ( Fig 2B) . Further, RTD-1 treatment resulted in restoration of limb function and mobility (typical response shown in Supplemental S1 Video).

Joints from vehicle-treated PIA rats showed extensive swelling, synovial hyperplasia, invasive pannus, and erosion of cartilage and bone ( Efficacy of RTD-1 in severe PIA and comparison with RA drugs RTD-1 treatment was also effective in resolving advanced PIA. Rats with near maximum disease (AS !12) received daily s.c. injections of saline or 5 mg/kg of RTD-1, peptide treatment was continued until disease resolution (AS = 0) was achieved (6-9 RTD-1 treatments). Peptide treatment produced a rapid reduction in arthritis severity, which reached statistical significance within 48 hours of the start of treatment (P < 0.044), with complete resolution of clinical disease in all RTD-1 treated rats by day 15 ( Fig 3A) . As with animals treated in early PIA (Fig  2A) , RTD-1 treatment of animals with severe PIA restored normal limb function and mobility (S2 Video). Of note, after a disease-free interval ranging from 4-12 weeks, three of five RTD-1 treated animals relapsed with joint disease (average AS score of~3). Efficacy of RTD-1 treatment was compared with etanercept (Etan) and methotrexate (MTX), first line RA drugs that have also been studied in rodent models of arthritis such as rat PIA [1, [19] [20] [21] [22] . While both Etan and MTX limited the progression of arthritis by day 3 of treatment compared to vehicle controls (Fig 3B) , treatment with RTD-1 had a more pronounced effect by treatment day 11 than either Etan or MTX, (P = 3.56 × 10 −5 , P = 0.0059, respectively). Additionally, RTD-1 achieved the highest rate of complete disease resolution (25%) compared to MTX (4.5%) or Etan (0%) (Fig 3C) .

The anti-arthritic effects of RTD-1 in PIA were dose-dependent, with AS reductions observed with doses as low as 1 mg/kg (Fig 4A) . Further, dosing of RTD-1 at 5 mg/kg was equally effective whether administered daily or once every 2, 3, or 5 days (Fig 4B) . Analysis of single dose pharmacokinetics (PK) of s.c. administered RTD-1 revealed that the peptide disposition is best Fig 4C) . These PK parameters are consistent with the finding that RTD-1 is highly stable in human [9] and rat plasma and that q5d dosing is as effective as daily dosing (Fig 4B) .

To evaluate the immunogenicity of RTD-1, we challenged DA and Sprague Dawley rats with serial (8-42 daily) s.c. injections of 5 mg/kg RTD-1. Serum or plasma samples from animals were analyzed for humoral response by dot immunoblot assay. No anti-RTD-1 Ig was detected in any peptide-challenged animal (Fig 5) , consistent with previous failed attempts to generate antibodies with unmodified RTD-1 [9] .

IL-1β plays a central role in the pathogenesis of RA [28] and rodent arthritis [29] [30] [31] . In previous studies, we showed that RTD-1 suppresses IL-1β expression and release by immune-stimulated monocytes [9, 15] . We therefore analyzed the effect of RTD-1 administration on joint levels of IL-1β in PIA rats. Consistent with other studies, IL-1β was not detectable in joints of naïve rats, but levels were markedly elevated in diseased joints of PIA rats and joint IL-1β levels correlated with arthritis severity score of the corresponding limb (Pearson r = 0.4923, Fig 2. Efficacy of RTD-1 in rat PIA. A. DA rats with established PIA were treated s.c. daily with RTD-1 (5 mg/kg) or vehicle (saline), with cohort randomization as described in Methods. RTD-1 treatment reduced arthritis severity scores within 24 h compared to control (* P = 0.0064), and therapeutic efficacy increased with duration of treatment (** P < 0.001; *** P < 0.0001, mean ± 95% CI, results of 6 independent experiments with n = 5-10 animals per treatment group per experiment). The larger number of saline-treated animals represented in the figure is a result of variation in randomization sequences across six experimental trials. B. Median AS (with interquartile ranges) of individual animals on day 11 from panel A. C. Representative images (10X magnification, bar = 500 μm) show gross and histologic features of hind extremities from naïve, saline-treated PIA rats (daily s.c. x 11 days), and RTD-1 treated PIA rats (daily s.c. 3 mg/kg for 9 or 11 days; see S1 Fig for additional examples of normal, PIA, and RTD-1-treated joint histology). Joints from saline-treated PIA rats show synovial hyperplasia, invasive pannus, and disruption of joint architecture (panels 6-8). RTD-1 treatment of PIA rats resulted in marked resolution of paw swelling as well as preservation of a normal joint architecture (panels 9-12).

https://doi.org/10.1371/journal.pone.0187868.g002 A. RTD-1 induces resolution of advanced PIA (arthritis severity score, AS ! 12 prior to treatment) with significant (* P < 0.05; ** P < 0.005) reduction in AS (mean AS ± 95% CI) by day 2 of daily treatment with s.c. peptide at 5 mg/kg (results of two independent experiments). B. Efficacy of RTD-1 in rat PIA was superior to treatment with MTX or etanercept (Etan). All three agents significantly (P < 0.05) reduced arthritis progression by day 3 following treatment initiation, (mean AS ± 95% CI). Anti-arthritic efficacy of θ-defensin RTD-1 P = 0.0013; Fig 6A) . PIA rats with established disease (AS ! 3) were treated s.c. once a day for four days with 5 mg/kg RTD-1 or vehicle after which joint tissue from euthanized animals was analyzed for IL-1β. Peptide treatment reduced joint IL-1β by an average of 44% (P = 0.0054, Fig 6B) , which coincided with the marked reduction in arthritis severity in RTD-1 treated animals after 4 days of treatment (P = 0.0044; Fig 2A) .

IL-6 is a pleiotropic cytokine, also implicated in RA pathogenesis by inducing differentiation of Th-17 cells, joint damaging proteases, angiogenesis, synoviocyte invasion [32, 33] , and bone loss [34] . Stimulation of FLS from arthritic DA rats with TNFα or IL-1β dose-dependently induced IL-6 production, consistent with previous studies [35, 36] . RTD-1 significantly inhibited IL-6 production (27.3-42.2% reduction) by FLS (Fig 6C) . 

Arthritogenic cytokines stimulate the proliferation and invasion of FLS, and the in vitro invasive properties of these cells mediate and correlate with erosive joint changes in RA and PIA [23, 25, 37, 38] . Therefore, we analyzed the effect of RTD-1 on FLS proliferation and invasiveness. RTD-1 dose-dependently suppressed FLS proliferation when analyzed after 24 or 48 hours in culture ( Fig 6D) and this effect was not a result of cytotoxicity as cell viability was ! 98% in all samples. RTD-1 also dose-dependently suppressed FLS invasion of Matrigelcoated inserts (Fig 6E) . The invasion-suppressive effects were observed at low concentrations (100 ng/ml) and were highly reproducible against seven independent DA rat FLS cell lines ( Fig  6F) .

RTD-1 is a potent and extremely rapid inhibitor of TNFα release by LPS-stimulated leukocytes [9] . Based on the kinetics of these effects, we hypothesized that RTD-1 inhibits TNFα release by inhibition of TNFα converting enzyme (TACE; ADAM17) [39] . RTD-1 was found to be a potent inhibitor of ADAM17 as well as ADAM10, a related proTNF sheddase (Table 1) . Additional studies, currently underway, disclose that RTD-1 rapidly and reversibly inhibits TACE (manuscript in preparation).

With the finding that RTD-1 inhibits ADAMs 17 and 10, both of which are zinc metalloproteinases, we tested for peptide effects on zinc matrix metalloproteinases (MMPs) implicated in cartilage degradation in RA. RTD-1 selectively inhibited pathogenic MMPs 2, 8, 9, 13 , and 14, with IC 50 values ranging from 2 to 20 μM (Table 1 ). In parallel we analyzed the effects of RTD-1 on cysteine proteases implicated in RA. Surprisingly, RTD-1 also inhibited cysteine cathepsins B, C, K, L, and V, with its greatest potency against cathepsin K (IC 50~2 7 nM; Table 1 ). Cathepsin K is the protease primarily responsible for degradation of bone matrix by osteoclasts [40] . These data indicate that RTD-1 is a cross class anti-protease that inhibits TNFα activation and matrix degrading proteinases implicated in RA joint damage.

Despite advances in RA treatment over the past 15 years, there persists an unmet need for more widely effective therapies given the significant fraction of patients who are partial responders or who are ineligible candidates for existing drugs [4] . Previous studies disclosed that RTD-1 modulates inflammatory cytokine expression in rodent infectious disease models [9, 14] and in immune-stimulated human leukocytes [9, 15] . In those studies, arthritogenic cytokines (TNFα, IL-1β, and IL-6) were among those down-regulated by RTD-1. Based on these findings, we hypothesized that RTD-1 would alter the course of disease in a rodent model of RA wherein disease onset and progression are driven by TNFα, IL-1β, and IL-6. The PIA rats after 4 daily 5 mg/kg s.c. treatments compared to vehicle controls (brackets indicate number of paw-ankle joints). Data are shown as IL-1β levels in individual joint samples with medians with interquartile ranges indicated. C. RTD-1 (10 μg/ml) suppressed (3 or 10 ng/ml) TNFα-and IL-1β-stimulated production of IL-6 by cultured FLS (* P < 0.05, ** P = 0.0059) (mean + SE, n = 12; results of 3 independent experiments). D. RTD-1 dose-dependently suppressed proliferation of cultured DA-PIA-FLS. FLS cell numbers were quantified after 0, 24, and 48 h incubation with RTD-1 at the indicated concentrations (* P = 0.0023, ** P = 0.0243, *** P < 0.0001, P-values relate differences in cell proliferation relative to peptide free controls at corresponding time points) (mean + SE, n = 7, results of 2 independent experiments). E. RTD-1 suppressed invasion of FLS from PIA rats in a dose-dependent manner, with minimal cytotoxicity (percent viable cells above bars, n = 2). F. The anti-invasive effects of RTD-1 on 7 independently isolated rat FLS cell lines were analyzed (as in panel E). Incubation with 0.1 μg/ml RTD-1 inhibited invasion by 50% (mean ± 95% CI) obtained using 7 different rat DA cell lines.

https://doi.org/10.1371/journal.pone.0187868.g006

Anti-arthritic efficacy of θ-defensin RTD-1 studies reported here employed rat PIA since this T-cell dependent disease model mimics human RA, including symmetrical involvement of peripheral joints, destruction of cartilage and bone, dysregulated pro-inflammatory cytokines/chemokines, rheumatoid factor positivity, and its chronic course. PIA is dependent on MHC class II-restricted T-cells [26] and influenced by non-MHC genes, producing cellular and humoral autoimmunity [16, 25, 41] . Non-MHC genes that influence rat PIA arthritis severity include TNFα, IL-1β [42] , IL-6, IL-17 [43] , and CCL-2, and joint damaging MMPs [25, 44] . Rat PIA is increasingly utilized as a model for preclinical evaluation of human RA drug candidates [16] .

Here we report that RTD-1, the prototype θ-defensin, rapidly arrests PIA in DA rats and peptide treatment resulted in high rates of disease resolution. Therapeutic effects were rapid, and dose level and dose interval dependent. RTD-1 induced disease resolution more effectively than a TNFα inhibitor or MTX, and at rates exceeding that reported using a mouse anti-TNFα antibody [45] , or the < 10% remission rate observed in treated RA patients [46] . Moreover, arrest of disease progression and resolution of arthritis was obtained with RTD-1 treatment in early and in severe stages of PIA, and therapeutic responses correlated with marked reductions in limb swelling, and recovery of limb function and mobility. In our studies on the effect of RTD-1 on evolving disease, resolution of clinical arthritis corresponded to histologic joint morphology which disclosed that treated animals had essentially normal joint structures whereas those from control animals showed severe pannus formation, chronic inflammation, and erosion ( Fig 2C and S1 Fig) . In this study, we did not perform histologic evaluation of joints from RTD-1 treated animals with maximum established disease (Fig 3A) . Of note, peptide treatment of severely arthritic animals induced a higher frequency of complete clinical resolution (Fig 3A; S2 Video) than occurred with treatment of animals with evolving disease (Fig  2B) , and this may be of translational significance since treatment of RA is usually initiated once disease is well established. We hypothesize that this somewhat paradoxical finding reflects pleiotropic anti-arthritic RTD-1 mechanisms which differentially affect the pathophysiology of evolving and established severe disease. Studies are underway to identify pathologic and/or resolution-promoting processes modulated by RTD-1 at different stages of PIA. Evaluation of the effect of RTD-1 in another animal model of RA, such as collagen induced arthritis, is likely to provide additional insights into anti-arthritic mechanisms mediated by θ-defensins.

Our data suggest that RTD-1 modulates arthritogenic pathways at more than one level. For example, down-regulation of inflammatory cytokine expression has been shown to occur at the transcriptional and post-transcriptional levels, the former revealed in studies on THP-1 macrophages wherein RTD-1 was shown to down-regulate MAP kinases and NF-B activation, inhibiting transcription of TNFα, IL-1β, IL-8, CCL-3, and CCL-4 [15] . Post-translational regulation of arthritogenic mediators is exemplified by the rapid and potent inhibition of ADAM17 by RTD-1 which blocks TNFα release [39] . Consistent with this effect, RTD-1 treatment of PIA rats suppressed expression of joint IL-1β, a central mediator of arthritis which is upregulated by TNFα [35, 47] . RTD-1 also acts directly on FLS to reduce the production of IL-6 and reduces cell invasiveness, two processes involved in joint damage. Studies are underway to further delineate the pathways that mediate RTD-1 modulation of other inflammatory/ arthritogenic cytokine networks implicated in RA.

The inhibitory action of RTD-1 on arthritogenic proteases, including MMPs and cathepsin K, suggests another anti-arthritic mechanism that may be operative in vivo. Results of the current study demonstrate that RTD-1 is a cross class anti-protease that inhibits zinc metalloproteinases (ADAMs and MMPs) and cysteine cathepsins. The inhibitory potency of RTD-1 against a panel of MMPs and cathepsins varies substantially, but plasma levels (C max ) achieved following s.c. treatment approximate or exceed the IC 50 s of ADAM17 and cathepsin K, potentially providing a rapid mechanism for regulating these arthritogenic proteases, in the former case down-regulating local TNFα levels and downstream effects (e.g., IL-1β, IL-6, MMP expression). This may also contribute to inhibition of FLS invasiveness mediated by low concentrations of RTD-1.

θ-defensins are expressed at high levels in leukocytes and other tissues of Old World monkeys [8, [11] [12] [13] . However, θ-defensins are absent in hominids as a result of pseudogenization of θ-defensin precursor genes during primate evolution [48] . Evidence suggests that the expression of θ-defensins in rhesus macaque neutrophils underlies the superior antimicrobial properties of azurophil granules from these cells, compared to those of humans [11] . Further, θ-defensins possess unique immunomodulatory properties lacking in human α-and β-defensins [9, 15] , potentially contributing to the marked resistance of θ-defensin expressing monkeys (e.g., macaques, vervets, and baboons) to endotoxin [49] .

The results of this study indicate that θ-defensin-like cyclic peptides have potential as new agents for treatment of RA. RTD-1 itself is a pharmaceutical candidate as this naturally occurring molecule is well tolerated when administered subcutaneously (this study and ref. [9] ), and by intravenous, intramuscular, and intraperitoneal routes (our unpublished results). RTD-1 was non-immunogenic (Fig 5) , circumventing the development of neutralizing antibodies, a problem encountered by a subset of patients in response to biologic therapeutics [50] . Further, RTD-1 is highly stable in biological matrices [9] , has an unusually long terminal half-life allowing for infrequent dosing (Fig 4) , and shows no evidence of being immuno-compromising [9, 14] . The concept of using a nonhuman primate peptide represents a novel retroevolutionary therapeutic paradigm whereby products of genes expressed in Old World monkeys may be used to treat a human autoimmune disorder.

Supporting information S1 Fig. PIA histology. Additional examples of joint histology of naïve, PIA, and RTD-1-treated PIA rats from which representative micrographs were selected in Fig 2C. Histology images were captured at either 10 or 4X magnification (organized by column), the top of each column contains a scale bar (10X = 500 μm, 4X = 1000 μm). The disease severity score of each individual sectioned limb is indicated as AS with possible scores of 0-4. (TIF) S1 Video. RTD-1 arrests PIA and protects limb function. Scene one shows a healthy DA rat before PIA induction. Scene two is a PIA DA rat treated with saline for 11 days after disease presentation. Scene three is a PIA DA rat treated with RTD-1 (5 mg/kg s.c. qd x 11 days) for 11 days after disease presentation. (MOV) S2 Video. RTD-1 treatment induces resolution of severe PIA and restores limb function. Scene one shows a healthy DA rat before PIA induction. Scene two is a PIA rat treated qd with saline 11 days after disease presentation. Scene three is the same rat in scene 2 after 8 days of RTD-1 treatment (5 mg/kg s.c. qd x 8 days). 

The pathogenesis of rheumatoid arthritis

State-of-the-art: rheumatoid arthritis

Evolving concepts of rheumatoid arthritis

Factors influencing response to disease modifying antirheumatic drugs in patients with rheumatoid arthritis

Treatment of rheumatoid arthritis with anti-TNF-alpha agents: a reappraisal

Theta-Defensins: cyclic peptides with endless potential

Defensins: endogenous antibiotic peptides of animal cells

A cyclic antimicrobial peptide produced in primate leukocytes by the ligation of two truncated alpha-defensins

Rhesus macaque theta defensins suppress inflammatory cytokines and enhance survival in mouse models of bacteremic sepsis

Structural characterization of the cyclic cystine ladder motif of theta-defensins

Rhesus macaque {theta}-defensin isoforms: expression, antimicrobial activities, and demonstration of a prominent role in neutrophil granule microbicidal activities

Isolation, synthesis, and antimicrobial activities of naturally occurring theta-defensin isoforms from baboon leukocytes

Homodimeric theta-defensins from rhesus macaque leukocytes: isolation, synthesis, antimicrobial activities, and bacterial binding properties of the cyclic peptides

Rhesus thetadefensin prevents death in a mouse model of severe acute respiratory syndrome coronavirus pulmonary disease

Rhesus macaque thetadefensin RTD-1 inhibits proinflammatory cytokine secretion and gene expression by inhibiting the activation of NF-kappaB and MAPK pathways

Animal Models of Rheumatoid Arthritis (I): Pristane-Induced Arthritis in the Rat

Pristane-induced arthritis in the rat

The mouse model of collagen-induced arthritis

Overexpression of toll-like receptor 3 in spleen is associated with experimental arthritis in rats

Interleukin-20 antibody is a potential therapeutic agent for experimental arthritis

Combined effects of bucillamine and etanercept on a rat type II collagen-induced arthritis model

Characterization of pristane-induced arthritis, a murine model of chronic disease: response to antirheumatic agents, expression of joint cytokines, and immunopathology

Cia5d regulates a new fibroblast-like synoviocyte invasionassociated gene expression signature

Matrix metalloproteinase inhibition by heterotrimeric triple-helical Peptide transition state analogues

Identification of two new arthritis severity loci that regulate levels of autoantibodies, interleukin-1beta, and joint damage in pristane-and collagen-induced arthritis

Pristane, a non-antigenic adjuvant, induces MHC class II-restricted, arthritogenic T cells in the rat

Pristane-induced arthritis in rats: a new model for rheumatoid arthritis with a chronic disease course influenced by both major histocompatibility complex and non-major histocompatibility complex genes

Balance of IL-1 receptor antagonist/IL-1 beta in rheumatoid synovium and its regulation by IL-4 and IL-10

IL-1 alpha beta blockade prevents cartilage and bone destruction in murine type II collagen-induced arthritis, whereas TNF-alpha blockade only ameliorates joint inflammation

Role of interleukin 1 in antigen-induced exacerbations of murine arthritis

Analyses of synovial tissues from arthritic and protected congenic rat strains reveals a new core set of genes associated with disease severity

The vitamin D receptor regulates rheumatoid arthritis synovial fibroblast invasion and morphology

Liver X receptor regulates rheumatoid arthritis fibroblast-like synoviocyte invasiveness, matrix metalloproteinase 2 activation, interleukin-6 and CXCL10

Interleukin 6 and rheumatoid arthritis

Anti-arthritic effects of magnolol in human interleukin 1beta-stimulated fibroblast-like synoviocytes and in a rat arthritis model

Notch signaling mediates TNF-alpha-induced IL-6 production in cultured fibroblast-like synoviocytes from rheumatoid arthritis

Duality of fibroblast-like synoviocytes in RA: passive responders and imprinted aggressors

mTOR regulates the invasive properties of synovial fibroblasts in rheumatoid arthritis

A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells

Cathepsin k is a critical protease in synovial fibroblast-mediated collagen degradation

Contribution of genetic studies in rodent models of autoimmune arthritis to understanding and treatment of rheumatoid arthritis

The non-major histocompatibility complex quantitative trait locus Cia10 contains a major arthritis gene and regulates disease severity, pannus formation, and joint damage

Synovial expression of Th17-related and cancer-associated genes is regulated by the arthritis severity locus Cia10

The arthritis severity locus Cia5d is a novel genetic regulator of the invasive properties of synovial fibroblasts

Anti-TNF-alpha antibody allows healing of joint damage in polyarthritic transgenic mice

Healing of erosive changes in rheumatoid arthritis

Lasker Clinical Medical Research Award. TNF defined as a therapeutic target for rheumatoid arthritis and other autoimmune diseases

Mammalian defensins in the antimicrobial immune response

Endotoxic shock model with fluid resuscitation in Macaca mulatta