CD31 peptides for the treatment of thrombotic or autoimmune disorders

Abstract

The present invention stems from the finding that the extracellular domain of CD31 proteins present on blood leukocytes is shed and released in the circulation as a soluble form of CD31. The invention relates to peptides corresponding to fragments of CD31 that inhibit T-cell response, and to their use in the treatment of thrombotic disorders such as atherothrombosis and autoimmune disorders.

Claims

1. An isolated peptide of 6 to 15 amino acids consisting of: a) a fragment of 6 to 15 amino acids of sequence SEQ ID NO: 6; or b) a fragment of 6 to 15 amino acids of sequence SEQ ID NO: 5 corresponding to (a) in murine CD31; and wherein said peptide comprises or consists of an amino acid sequence selected in the group consisting of SEQ ID Nos 2, 3, 4, 22, 23, 24, 25, 26, 27, 28, 29, 39, 40, 41, 42, 43, 44, 45, 46, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 87, 88, 89, 90, 91, 96, 97, 98, 99 and 100.

2. The peptide according to claim 1, wherein said peptide consists of any one of SEQ ID NOs: 2, 3 or 4.

3. The peptide according to claim 1, wherein said peptide comprises at least one chemical modification improving its stability and/or its bioavailability and wherein said at least one chemical modification improving its stability and/or its bioavailability is selected from the group consisting of: (a) modification to the N-terminal and/or C-terminal end of the peptide by N-terminal acylation or deamination; (b) modification of the C-terminal carboxyl group into an amide or an alcohol group; (c) modification at the amide bond between two amino acids by acylation or alkylation at the nitrogen atom or the alpha carbon of the amide bond linking said two amino acids; (d) modification at the alpha carbon of the amide bond linking two amino acids by acylation or alkylation at the alpha carbon of the amide bond linking said two amino acids; (e) replacement of one or more naturally occurring L-enantiomeric amino acids with a corresponding D-enantiomer; (f) retro-inversion, in which one or more naturally occurring L-enantiomeric amino acids is replaced with a corresponding D-enantiomer, together with inversion of the amino acid chain; (g) replacement of one or more alpha carbons with nitrogen atoms; and (h) binding of the amino group of one or more amino acid to the carbon instead of the carbon.

4. A pharmaceutical composition comprising one or more of a peptide of 6 to 15 amino acids consisting of a) a fragment of 6 to 15 amino acids of sequence SEQ ID NO: 6; or b) a fragment of 6 to 15 amino acids of sequence SEQ ID NO: 5 corresponding to (a) in murine CD31; and wherein said peptide comprises or consists of an amino acid sequence selected in the group consisting of SEQ ID Nos 2, 3, 4, 22, 23, 24, 25, 26, 27, 28, 29, 39, 40, 41, 42, 43, 44, 45, 46, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 87, 88, 89, 90, 91, 96, 97, 98, 99 and 100, a nucleic acid encoding said peptide; and a physiologically acceptable carrier.

5. A method of activating or restoring CD31-mediated signaling in an individual in need thereof, comprising administering to the individual a peptide of 6 to 15 amino acids consisting of: a) a fragment of 6 to 15 amino acids of sequence SEQ ID NO: 6; or b) a fragment of 6 to 15 amino acids of sequence SEQ ID NO: 5 corresponding to (a) in murine CD31; and wherein said peptide comprises or consists of an amino acid sequence selected in the group consisting of SEQ ID Nos 2, 3, 4, 22, 23, 24, 25, 26, 27, 28, 29, 39, 40, 41, 42, 43, 44, 45, 46, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 87, 88, 89, 90, 91, 96, 97, 98, 99 and 100.

6. A method of claim 5, wherein said method is a method for treating a thrombotic or an autoimmune disorder in said individual and wherein said individual has a thrombotic or an autoimmune disorder.

7. The method according to claim 5, wherein said individual has a CD31.sup.T lymphocytes phenotype.

8. The method according to claim 5, wherein said peptide comprises at least one chemical modification improving its stability and/or its bioavailability and wherein said at least one chemical modification is selected from the group consisting of: (a) modification to the N-terminal and/or C-terminal end of the peptide by N-terminal acylation or deamination; (b) modification of the C-terminal carboxyl group into an amide or an alcohol group; (c) modification at the amide bond between two amino acids by acylation or alkylation at the nitrogen atom or the alpha carbon of the amide bond linking said two amino acids; (d) modification at the alpha carbon of the amide bond linking two amino acids by acylation or alkylation at the alpha carbon of the amide bond linking said two amino acids; (e) replacement of one or more naturally occurring L-enantiomeric amino acids with a corresponding D-enantiomer; (f) retro-inversion, in which one or more naturally occurring L-enantiomeric amino acids is replaced with a corresponding D-enantiomer, together with inversion of the amino acid chain; (g) replacement of one or more alpha carbons with nitrogen atoms; and (h) binding of the amino group of one or more amino acid to the carbon instead of the carbon.

9. The method according to claim 6, wherein said disorder is a thrombotic disorder selected from the group consisting of atherothrombosis, atherosclerosis, acute coronary syndrome, ischemic stroke, peripheral arterial disease and abdominal aortic aneurysm.

10. The method according to claim 6, wherein said disorder is an autoimmune disorder selected from the group consisting of rheumatoid arthritis, multiple sclerosis, inflammatory bowel disease, systemic lupus erythematosus, Graves' disease and diabetes mellitus.

11. The method according to claim 6, wherein said individual has a CD31.sup.T lymphocytes phenotype.

12. The method according to claim 6, wherein said peptide comprises at least one chemical modification improving its stability and/or its bioavailability and wherein said at least one chemical modification is selected from the group consisting of: (a) modification to the N-terminal and/or C-terminal end of the peptide by N-terminal acylation or deamination; (b) modification of the C-terminal carboxyl group into an amide or an alcohol group; (c) modification at the amide bond between two amino acids by acylation or alkylation at the nitrogen atom or the alpha carbon of the amide bond linking said two amino acids; (d) modification at the alpha carbon of the amide bond linking two amino acids by acylation or alkylation at the alpha carbon of the amide bond linking said two amino acids; (e) replacement of one or more naturally occurring L-enantiomeric amino acids with a corresponding D-enantiomer; (f) retro-inversion, in which one or more naturally occurring L-enantiomeric amino acids is replaced with a corresponding D-enantiomer, together with inversion of the amino acid chain; (g) replacement of one or more alpha carbons with nitrogen atoms; and (h) binding of the amino group of one or more amino acid to the carbon instead of the carbon.

13. The pharmaceutical composition according to claim 4, wherein said peptide comprises at least one chemical modification improving its stability and/or its bioavailability and wherein said at least one chemical modification is selected from the group consisting of: (a) modification to the N-terminal and/or C-terminal end of the peptide by N-terminal acylation or deamination; (b) modification of the C-terminal carboxyl group into an amide or an alcohol group; (c) modification at the amide bond between two amino acids by acylation or alkylation at the nitrogen atom or the alpha carbon of the amide bond linking said two amino acids; (d) modification at the alpha carbon of the amide bond linking two amino acids by acylation or alkylation at the alpha carbon of the amide bond linking said two amino acids; (e) replacement of one or more naturally occurring L-enantiomeric amino acids with a corresponding D-enantiomer; (f) retro-inversion, in which one or more naturally occurring L-enantiomeric amino acids is replaced with a corresponding D-enantiomer, together with inversion of the amino acid chain; (g) replacement of one or more alpha carbons with nitrogen atoms; and (h) binding of the amino group of one or more amino acid to the carbon instead of the carbon.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1A-B shows an alignment between the sequences of human, mouse, pig and bovine CD31. The sequence defined by amino acids 579 to 601 of SEQ ID NO: 4 (human CD31) and the corresponding sequence in mouse, pig and bovine CD31 is highlighted by a box.

(2) FIG. 2 shows a representative example of 10-color flow-cytometry analysis of human peripheral blood cells from a healthy donor. Isotype controls of antibodies anti-CD31 dom1 and anti-CD31 dom6 are shown in the insets. Lymphocytes, Monocytes and Granulocytes were gated within the FSC/SSC scatter. B (CD20 A700+) and T (CD3 PE-TR+) lymphocytes were identified and gated within the Lymphocytes and CD8+ (PerCP) and CD4+ (APC) subpopulations were gated within T lymphocytes. CD8+ and CD4+ T cells were further analyzed for the expression of HLA-DR and CD45RA and accordingly subdivided in activated (1), memory (2) and nave (3) cells. All leukocytes were positive for CD31 dom6. Lack of dom1 increased from nave (3) to memory (2) to activated (1) T cells.

(3) FIG. 3A-B shows that the apparent loss of CD31 on lymphocytes is due to its extracellular shedding. a. Solubilized cell membrane-bound CD31 molecules were extracted from cultured Jurkat CD4+ T cells and coupled to fluorescent beads. The percentage of dom1-bead-bound molecules is <6% in resting conditions and >99% 5 after TCR engagement. b. Most soluble CD31 in culture supernatant () of TCR-activated T cells and in human plasma (.square-solid.) consists of a single truncated fragment comprising dom1-dom5 and lacking dom6. Negligible levels of truncated CD31 lacking both dom5 and dom6 could be detected only in plasma.

(4) FIG. 4A-B shows that a peptide homotypic of the residual extracellular fragment on CD31.sup.shedT induces CD31-ITIM phosphorylation. a. Proliferative response to TCR engagement of human peripheral blood mononuclear cells in the presence of increasing doses of CD31 peptide 551-574. *p<0.05 vs dose 0. b. Flow cytometry assessment of 686ITIM phosphorylation on solubilized membrane-bound CD31 from cultured Jurkat CD4+ T cells. Solubilized proteins were captured by E9-PECAM-1.2 (dom6) functional CBA beads and detection was carried out by anti-pY686 rabbit sera followed by AlexaFluor488-anti-rabbit secondary antibody. The histogram shows the Median Fluorescent Intensity (MFI)the % of the variability coefficient (CV %) of Alexafluor488 (pY686) over 2000 E9-PECAM-1.2 acquired beads. Pervan=positive control (pervanadate); CD3/CD28=anti-CD3 and anti-CD28 antibodies (1 g/ml each); peptide=CD31 peptide 551-574 (100 M).

(5) FIG. 5A-D shows that CD31 homotypic peptide 551-574 inhibits T-cell responses. a. BIAcore analysis of mouse CD31 peptide 551-574 homophilic binding at two-fold stepwise dilutions of the analyte (12.5, 25, 50 and 100 M). Data are normalised against the control channels and expressed as RU (Resonance Units). b. TCR-induced intracellular calcium mobilisation determined by flow cytometry in Fluo-3AM-loaded spleen cells. Data are expressed as MFI (530/30 nm). Grey arrow=addition of either anti-CD3/CD28 antibodies and crosslinker alone (.square-solid.=control) or with peptide 551-554 at 100 M (.circle-solid.) or with anti-mouse CD31 antibodies (). c. Proliferation in response to TCR-stimulation in CD31+/+ (black columns) and CD31/ (white columns) spleen cells. A dose-dependent inhibition is observed in CD31+/+ cells while only the highest dose of the peptide affect proliferation of CD31/ splenocytes. No effect was observed with 100 M dose of the scramble peptide on CD31+/+ cells (crisscross column). *p<0.05 vs previous peptide dose. d. Immunosuppressive effect of the peptide in the DTH model. *p<0.01 vs scramble (crisscross column) and 10 M dose of peptide 551-574. Data are expressed as meanSEM.

(6) FIG. 6A-C shows that CD31 peptide biotherapy prevents acceleration of atherosclerosis and aneurysm formation. a. MeanSEM of Atherosclerotic lesion surface area in serial cross-sections of the aortic root at 200, 400, 600, 800 m from the appearance of the first cusp in control (.square-solid.) and peptide 551-574 () treated mice. *p<0.05 vs control, ANOVA, repeated measure. b. The presence of an abdominal aortic aneurysm (AAA) was macroscopically, blindly evaluated by A.G. and A. N. after careful dissection of the adventitial tissue. An aneurysm was present in 8/10 (Exp #1) or 6/8 (Exp #2) control mice as opposed to only 1/8 peptide-treated mice in both experiments (p<0.001 by Chi squared test). The images below show an example of aneurysm (AAA; arrows) as compared to the absence of aneurysm (no AAA) of the abdominal aorta. c. Flow cytometry analysis of the effect of the peptide (50 M) on CD8+ T cell (% cytolysis) and macrophage (MMP-2/9 activity) functions. Data represent meanSEM of cultures from 3-4 mous/group. *p<0.05 vs control.

(7) FIG. 7 shows the clinical score of EAE activity (paralysis level). pepREG refers to the peptide of SEQ ID NO: 3. pepSCRA refers to the peptide of SEQ ID NO: 14.

BRIEF DESCRIPTION OF THE SEQUENCES

(8) SEQ ID NO: 1 corresponds to the sequence of human CD31.

(9) SEQ ID NOs: 2, 3, 4, 10, 11 and 12 correspond to CD31 peptides according to the invention.

(10) SEQ ID NOs: 5 and 6 correspond to CD31 peptides for use in the methods according to the invention.

(11) SEQ ID NO: 7 corresponds to the sequence of murine CD31.

(12) SEQ ID NO: 8 corresponds to the sequence of bovine CD31.

(13) SEQ ID NO: 9 corresponds to the sequence of pig CD31.

(14) SEQ ID NOs: 13 and 14 correspond to scramble peptides used as controls.

(15) SEQ ID Nos: 15 to 101 correspond to CD31 peptides according to the invention.

EXAMPLES

Example 1: Material and Methods

(16) Assessment of CD31.sup.+ and CD31.sup.shed blood leukocytes. Ten-color flow cytometry was performed on peripheral blood leukocytes from 5 healthy individuals either in basal conditions or after overnight stimulation with soluble 1 g/ml of purified anti-CD3 antibody (R&D Systems). Ten-color flow cytometry was performed after erythrocyte hypotonic lysis (10 minutes at 37 C. 1:10 v:v in Tris 10 mM, NH.sub.4Cl 155 mM, KHCO.sub.3 10 mM, pH 7.4) on heparinized peripheral blood leukocytes from 5 healthy individuals, fixed in PBS/Formaldehyde 1%/FCS 1% for 4 minutes at 37 C. prior to processing. All experiments on human blood were approved by the International Ethical committee (see world wide web page clinicaltrials.gov; Identifier: NCT00430820). Pelleted cells were incubated for 30 minutes at room temperature and protected from light with a cocktail of fluorescent monoclonal antibodies directed to CD3, CD4, CD8, HLA-DR, CD45RA, and CD31 (WM59) from BD Biosciences and anti-CD20 and anti-CD31 (PECAM 1.2) from Invitrogen (1 l of each). At least 50,000 events were acquired in the lymphocyte gate using a BD LSRII equipped with 3 lasers (405, 488 and 633 nm) and analysed with BD DIVA 6.0 software.

(17) Subtractive Measurement of Soluble CD31.

(18) To detect the splice variant and truncated CD31 in plasma and the culture supernatant, a cytokine bead array (CBA, BD) has been customized. Three differently functional CBA beads (A9, D5 and E9) were coupled with either one of the following purified monoclonal anti-CD31 antibodies JC70A (domain 1, DAKO), MEM-05 (domain 5, Zymed) and PECAM 1.2 (domain 6, Invitrogen). The coupled beads were then incubated with the plasma of the same 5 healthy controls (FIG. 2) or the culture supernatant and positive binding of circulating CD31 was detected by a fourth anti-CD31 monoclonal antibody, WM-59 (domains 1-2) coupled to PE (BD). The concentration of plasma CD31 including at least domain 1 (JC70A), or domains 1 to 5 (MEM-05) or all the extracellular domains 1 to 6 of CD31 (PECAM 1.2) was determined by analysing the median fluorescent intensity of the detecting antibody on 1000 gated beads on samples and serial dilutions of the same standard (recombinant, full length extracellular CD31, R&D Systems). The standard curve was obtained for each of the beads using the same known concentrations of the recombinant CD31 in order to overcome any bias due to differences in binding affinity of the diverse antibodies. The concentration in ng/ml of CD31 determined with PECAM 1.2 coupled beads (dom 1-6) was subtracted from the one obtained using MEM-05 coupled beads to obtain the amount of circulating CD31 lacking dom6 (dom 1-5). The latter was subtracted from the concentration of CD31 obtained using the JC70A-coupled beads to calculate the value of soluble CD31 lacking both dom 5 and 6 but containing at least domains 1 and 2 (dom 1-2).

(19) Assessment of CD31-ITIM Phosphorylation.

(20) Log-phase Jurkat cells (10.sup.7 cells/condition) were either left unstimulated (negative control) or incubated with pervanadate (positive control) or stimulated with anti-CD3 and anti-CD28 antibodies (R&D Systems, 1 g/ml each) in the presence or absence of peptide 551-574 (100 M), or incubated with the peptide alone during 20 minutes. Cells were then lysed with 1 ml of RIPA buffer on ice for 30 minutes, ultracentrifuged and 16 l of the supernatant was incubated with PECAM 1.2-coated Functional E9 CBA beads (BD) for 2 hours at room temperature. Beads were subsequently washed with CBA washing buffer and incubated with 2 l of undiluted rabbit anti-CD31 phospho-tyrosine 686 (pY686) sera followed by two washings and incubation with AlexaFluor488-conjugated (Fab).sub.2 fragments (1:100 in CBA whashing buffer) of goat-anti-rabbit IgG (Invitrogen). The beads (2000/condition) were finally analysed by flow cytometry in the FITC channel (530/30 nm) and data are expressed as Median fluorescence intensity (MFI)the percentage of the coefficient of variability (% CV) calculated with the DIVA 6.0 software (BD). Duplicate lysate aliquots and serial dilutions of recombinant CD31 were incubated with the PECAM 1.2-coated beads and the amount of dom1+ cell-bound CD31 was revealed using anti-CD31 WM59-R-PE (dom1) and PECAM 1.2-FITC (dom6) antibodies (data shown in FIG. 3a).

(21) Fluorescent Peptide Binding.

(22) For visualisation of peptide binding to CD31.sup.+ and CD31.sup.shed CD4.sup.+ T cells, freshly purified peripheral blood leukocytes prepared as above were washed with a buffered solution containing 2 mM EDTA (to avoid endocytosis of the peptide) and incubated overnight at room temperature in a dark humidified chamber with 50 M FITC-labelled CD31 peptide 551-574 and 1:10 dilution of fluorescent monoclonal anti-CD31 (PE) and anti-CD4 (APC) antibodies (BD Biosciences) in a poly-D-Lysine coated Ibidi 8-well culture chamber (Biovalley). Cells were then washed twice, nuclei counterstained with DAPI and digital images of a 0.3 m intracellular section were acquired on a Zeiss Axiovert M200 microscope (x63 immersion objective) equipped with the ApoTome and a cooled monochromatic digital camera (Zeiss).

(23) Calcium Mobilisation Assay.

(24) Spleen cells from C57Bl/6 mice were prepared as described in Caligiuri et al. (2005 Arterioscler Thromb Vasc Biol 25:1659-1664). Cells were incubated with Fluo-3AM (Invitrogen, # F1242) as per the instructions of the manufacturer. Fluorescence of calcium-bound tracer was measured in the FITC channel on an LSRII cytometer (BD Biosciences) prior to and during 60 seconds following the addition of hamster anti-mouse CD3/CD28 monoclonal antibodies (40 g/ml each) and rat/hamster compBead (1:50) either alone or in the presence of rat anti-mouse CD31 antibody (clone 390, 10 g/ml) or in the presence of CD31 peptide 551-574 (100 M). Negative controls included rat IgG isotype control and scramble peptide. Antibodies and compBeads were from BD Biosciences.

(25) Plasmon Surface Resonance.

(26) Homophilic binding association and dissociation constants were calculated by surface plasmon resonance (BIAcore 2000, GE). In brief, peptide 551-574 was coated at 3400 resonance units (RU) on CM5 chips according to the manufacturer's instructions. Soluble peptide 551-574 (12.5, 25, 50 and 100 M in 200 l of 10 mM HEPES pH 7.4, 150 mM NaCl, 0.005% Tween 20) was injected at 20 l/min at 25 C., on the peptide-coated channel and on an uncoated channel. Dissociation was monitored for 300 seconds. Association (kon) and dissociation (koff) constants were calculated using the BlAevaluation 3.0 Software (GE). Injection of peptide 551-574 on a channel coated with the scramble peptide yielded negligible signal.

(27) Evaluation of Immunoregulation In Vitro.

(28) CD8.sup.+ T cell-mediated cytolytic activity against allogeneic mouse aortic smooth muscle cells and measurement of macrophage gelatinase (MMP-2/9) activity were performed as previously described for human cells in Caligiuri et al. (2006 Arterioscler Thromb Vasc Biol 26:618-623) using kits and reagents from Invitrogen. Briefly, primary cultures of FVB/N mouse aorta smooth muscle cells were labelled with the lipophylic tracer DIO (green) and co-cultured for 3 hours with CD8+ T cell-enriched spleen cells from C57Bl/6 mice (n=3 scramble peptide and n=3 peptide 551-574, 50 M). Cytolysis was evaluated by intracellular accumulation of propidium iodide (PI). Cells were analysed by flow cytometry and the % of cytolysis was calculated by expressing the number of dead (PI+) cells among the target (DIO+) cells. Intracellular MMP-2/9 (gelatinase) activity was measured by flow cytometry in 7-day bone-marrow derived macrophages from C57Bl/6 mice (n=3 scramble peptide and n=3 peptide 551-574, 50 M) three hours after the incorporation of OregonGreen gelatine (MFI). T-cell proliferation was performed using either human peripheral blood mononuclear cells of spleen cells from C57Bl/6 (CD31.sup.+/+) and CD31.sup./ mice (Charles River France) as previously described (Caligiuri et al. 2005 Arterioscler Thromb Vasc Biol 25:1659-1664). Briefly, cells were plated in triplicates at 0.210.sup.6 cells/well in a U bottom 96-well plate in complete medium (RPMI 1640, 1% pyruvate, 1% glutamine, 1% penicillin-streptomicyne-fungizone, 10% decomplemented fetal calf serum, all from Invitrogen) containing 1 g/ml anti-mouse CD3/CD28 or 5 g/ml anti-human CD3 antibodies (BD) as appropriate. CD31 (551-574) and scramble peptide at 25, 50 and 100 M final concentration were deposited in the wells just before cell plating. Plated cells were cultured for 72 hours in 5% CO2 at 37 C. (.sup.3H) thymidine (0.5 Ci/well) was added for the last 16 hours and proliferation evaluated using a Tomtec harvester and analysis on a Wallac micro beta counter. Data are expressed as meanSEM of cpm in triplicates.

(29) Evaluation of Immunoregulation In Vivo.

(30) Delayed type hypersensitivity (DTH) suppression was evaluated as described in the Current Protocols in Immunology (2001) 4.0.1-4.0.2 Unit 4.2. Briefly, TNCB (2-chloro-1,3,5-trinitrobenzene, Fluka #79874) was dissolved in acetone/olive oil (1:1 v/v) at a concentration of 10 mg/ml. BALB/c mice (n=6/group) were primed by painting the shaved regions of the abdomen a with a total 0.2 ml of the preparation (n=6/group). The experiment included 3 groups for peptide 551-574 (10, 50, 100 M) and 1 group treated with scramble peptide at 100 M). Five days after priming, 10 l of the TNCB-solvent mixture was painted on the right pinna, 30 minutes after subcutaneous (interscapolar) administration of the peptide 551-574 or the scramble peptide. Ear thickness increases were calculated by subtracting the thickness of the right and the left pinna of each mouse (right-left/left100), measured at 24 h with a dial caliper (Pocotest, Kroeplin Lngenmesstechnick). The measure was performed 5 times on each ear and averaged for further analysis. The immunosuppressive effect of the peptide was calculated as % suppression=(1TE/TS)100, where T=(ear thickness 24 hr after elicitation)(baseline ear thickness), E=sensitised animals, S=treated animals. Data are expressed as meanSEM.

(31) Detection of Atherosclerotic Lesion Size and Aneurysm Formation.

(32) Male 28-week old apolipoprotein E.sup./ mice (n=8-10 mice/group) from our breeding facility were maintained on a regular chow diet and kept under standard conditions. Acceleration of atherosclerosis and aneurysm formation was induced by subcutaneous angiotensin II (Sigma, #A9525) infusion (1 mg/kg/d) for 28 days using osmotic minipumps (Alzet, #2004) as previously described (Daugherty et al. J Clin Invest 105:1605-1612). All experiments were approved by our institutional Ethical committee. Atherosclerotic lesion size was measured as previously described (Caligiuri et al. 2005 Arterioscler Thromb Vasc Biol 25:1659-1664). These experiments were repeated twice with similar results.

(33) Peptides.

(34) All experiments on human material were carried out using the human peptide sequence while the mouse equivalent was used in all mouse experiments. The sequences of the peptides are shown in the table below.

(35) TABLE-US-00004 Human NH2-NHASSVPRSKILTVRVILAPWKK-COOH SEQID NO:6 Mouse NH2-SSMRTSPRSSTLAVRVFLAPWKK-COOH SEQID NO:5 Scramble NH2-SMPAVRSRFSATSLVTLKSRWPK- SEQID COOH NO:13

Example 2: The Apparent Loss of CD31 at the Surface of Blood Lymphocytes is Due to its Shedding Between the 5th and 6th Extracellular Ig-Like Domains

(36) In order to establish whether the loss CD31 was restricted to part or extended to the totality of its 6 extracellular Ig-like domains, a multicolor flow cytometry analysis of whole blood leukocytes from 5 healthy donors using two different antibodies specifically recognizing the membrane-distal and membrane-proximal Ig-like domains of the molecule was performed. To be able to discriminate between the different leukocyte populations and assess their state of maturation and activation, a panel of lineage markers as well as the expression of CD45RA and HLA-DR were simultaneously used. While the expression of CD31, as detected by a monoclonal antibody specific for the first domains of CD31 (clone WM-59, dom1-2) was recognized on nave but not on activated/memory blood T cells, all cells expressed the membrane-proximal extracellular fragment of the molecule detected by another monoclonal antibody specific for the 6.sup.th Ig-like domain of CD31 (clone PECAM 1.2, dom6), irrespective of their state of maturation/activation (FIG. 2).

(37) Flow cell cytometry showed that T-cell receptor (TCR) engagement induces a shift of >80% of blood resting T cells from a CD31 dom1.sup.+/dom6.sup.+ to a dom1.sup./dom6.sup.+ (CD31.sup.shed) phenotype. Molecular analysis of the membrane proteins from cultured T-cell lysates demonstrated that >99% of the T cell-bound CD31 molecules drop the distal portion containing dom1 already 5 minutes after TCR stimulation in vitro (FIG. 3a). Analysis of the supernatant showed that, simultaneously, a single truncated soluble protein limited to the first 5 Ig-like domains of CD31, accumulates in the culture supernatant (FIG. 3b). Furthermore, the analysis of the plasma of the same healthy donors showed that major part of soluble CD31 in plasma was constituted of a truncated molecule comprising Ig-like domains 1 to 5 and specifically lacking the membrane-proximal 6.sup.th domain (FIG. 3b) that always remains anchored to the apparently CD31-negative (CD31 dom1.sup.) lymphocytes both in vitro and in vivo. Only a minimal fraction of soluble CD31 contained all 6 extracellular domains predicted in the previously reported (Goldberger et al. J Biol Chem 269:17183-17191) variant spliced form both in culture supernatant and in plasma (FIG. 3b). No significant other cleavage of the molecule occurs upstream of the 5.sup.th domain since the latter was virtually always present concomitantly with the first domain in the truncated soluble CD31 proteins (FIG. 3b).

Example 3: A Peptide Contained in the Residual Extracellular CD31 Fragment on CD31shed T Cells Enhances Phosphorylation of CD31-ITIM

(38) A CD31 dom6-derived synthetic peptide corresponding to the juxta-membrane 23 aminoacids (551-574) of the ectodomain of the human molecule binds both to CD31 dom1.sup.+ and to CD31 dom1.sup. (CD31.sup.shed) CD4.sup.+ T lymphocytes ex vivo. Importantly, the binding of this peptide on T cells has functional consequences on immune cell control since it exerted dose-dependent inhibition of human peripheral blood T-cell proliferation in vitro (FIG. 4a). To assess whether the inhibitory effect of the peptide could be mediated by homophilic binding and engagement of the CD31 signaling, the level of phosphorylation of the CD31 ITIM tyrosine at position 686 (.sub.686ITIM) in cultured T cells was evaluated. Stimulation of the TCR by anti-CD3 and anti-CD28 antibodies alone or the sole presence of the peptide induced a minor increase of CD31 pY686 (FIG. 4b) but concomitant TCR-stimulation in the presence of the peptide boosted the phosphorylation the CD31.sub.686ITIM by a factor of >23 (FIG. 4b).

Example 4: CD31 Homotypic Peptide 551-574 Inhibits a Large Array of Cell-Mediated Immune Responses In Vitro and In Vivo

(39) To further test our hypothesis, in vivo experiments were performed, and the murine equivalent of CD31 (551-574) peptide was therefore synthesized. Its properties were evaluated in vitro and in vivo. Its homophilic interaction by surface plasmon resonance analysis was first established (FIG. 5a). The association and dissociation constants were of 134120 M.sup.1s.sup.1 and 1.58E-031.07E-03 s.sup.1, respectively and the affinity at equilibrium was 17.89.7 M, in agreement with previous measurement done on recombinant human CD31 homo-dimerization using other systems. It was next assessed whether the mouse CD31 (551-574) peptide was able to inhibit calcium mobilization in response to the co-engagement of the CD3 and of the co-stimulatory molecule CD28 in spleen lymphocytes. The data showed that homophilic targeting of the juxta-membrane portion of CD31 whit this small peptide is as an efficient immunoregulatory strategy as it is the co-ligation of CD3 and CD28 with the distal portion of CD31 by cross-linking the molecules using specific monoclonal antibodies (FIG. 5b). The engagement of the CD31 molecular pathway by this peptide could attain effective suppression of antigen-driven lymphocyte responses in vitro and in vivo. It was found that peptide 551-574 inhibits in a dose-dependent manner the proliferation of spleen cells in response to stimulation of the T cell-receptor in vitro (FIG. 5c). At low dose, the effect of the peptide was exclusively due to its homophilic binding with CD31 molecules since no effect was observed on spleen cells from CD31.sup./ mice (FIG. 3c). However, at high concentrations the peptide can also bind to other (low affinity) heterophilic ligands as suggested by its effect at 100 M dose also on cells lacking CD31 (FIG. 5c). The scramble peptide, used at the highest dose, was ineffective (FIG. 5c). Finally, the therapeutic immunosuppressive potential of this peptide in vivo in the context of matched, histocompatible antigen-driven immune responses was evaluated using a model of delayed type of hypersensitivity. Distal recall of an hapten-elicited specific immune response was suppressed in a dose-dependent manner by peptide 551-574 in this model (FIG. 5d). Effective immunosuppression was achieved by a single subcutaneous administration of 50 M of the peptide 551-574 while the scramble peptide was ineffective (FIG. 5d).

Example 5: CD31shed Cell-Targeting Peptide Biotherapy Prevents Disease Progression and Aneurysm Formation in Atherosclerotic Mice

(40) The model of angiotensin II infusion into aged apolipoprotein E/ mice (25) was used, this model closely mimicking atherothrombosis in humans. Daily subcutaneous administration of 50 M of the peptide prevented both the acceleration of plaque growth in the aortic root (FIG. 6a) and the formation of atherothrombotic abdominal aortic aneurysms (AAA) in this model (FIG. 6b). in vitro, the CD31-derived peptide 551-574 inhibited both the activity of matrix degrading enzymes in bone-marrow derived macrophages and CD8.sup.+ T cell-dependent cytolysis of murine aortic smooth muscle cells (FIG. 6c).

Example 6: In Silico and In Vitro Validation of the 10 Amino Acid-Long Peptide Candidate Peptide PepReg

(41) A shorter nested peptide corresponding to the ten COOH terminal amino acids was tested. This peptide is shown as SEQ ID NO: 3 and referred to as PepReg. As shown in the table below, this peptide was expected to be more stable than the CD31 551-574 peptide of 23 amino acids.

(42) TABLE-US-00005 TABLE CharacteristicsofPepRegvsthe 23aaCD31peptideinsilico PepRegCD31564- CD31551-574 574(10aa) (23aa) sequence VRVFLAPWKK SSMRTSPRSSTLAVRV (SEQID FLAPWKK NO:3) (SEQID NO:5) Numberof 10 23 aminoacids Molecularweight 1243.5 2606.0 Theoreticalpl 11.17 12.31 Formula C62H98N16O11 C116H193N35O31S1 -(Asp,Glu)/+ 0/3 0/5 (Arg,Lys) charges Instabilityindex 25.38 65.09 Estimatedhalf- 100hours 1.9hours life(mammalian reticulocytes, invitro) Aliphaticindex 107.00 68.83

(43) The immunosuppressive properties of PepReg (10aa) vs the 23 aa parent peptide were evaluated in vitro. Negatively purified CD4+ cells from C57Bl6 mice were stimulated by soluble anti-CD3 purified antibodies and bone marrow derived dendritic cells. Cells from triplicate wells were analyzed for the expression of the early activation marker CD69 after 18 hours culture in complete RPMI medium supplemented with 10% fetal calf serum. Flow cytometry showed that PepReg was at least as efficient as the 23aa peptide in suppressing T cell activation as determined by the percentage of CD4 cells expressing CD69. Interestingly, the effect was more reproducible (smaller standard deviation) and was observed with lower doses (50 g/ml vs 100 g/ml) of PepReg (10aa) as compared to the parent (23aa) peptide. The suppression of T cell proliferation by PepReg was analyzed by [H3] thymidin incorporation and persisted up to 7 days of culture at 37 C. in 10% serum demonstrating that the data regarding the stability of the peptide obtained in sifico were validated in vitro.

Example 7: In Vivo Validation of the 10 Amino Acid-Long Peptide Candidate Peptide PepReg in the EAE Model

(44) Experimental Autoimmune Encephalomyelitis (EAE), also called Experimental Allergic Encephalomyelitis, is an animal model of Multiple Sclerosis.

(45) Twelve-week old female C57BL/6J mice were immunized with 300 g of MOG35-55 peptide emulsified in Complete Freund's Adjuvant 1:1 by volume containing 800 g of nonviable desiccated Mycobacterium tuberculosis H37RA. A final volume of 200 l of the emulsion was injected subcutaneously at 4 sites (50 l/site) over the flanks. In addition, 300 ng of Pertussis toxin was injected intravenously (retro-orbital plexus) on the same day and 2 days later. Clinical signs of EAE ware assessed daily by the following scoring system: 0, no signs; 1, hindlimb weakness; 2, hindlimb weakness and tail paralysis; 3, hindlimb and tail paralysis; 4, hindlimb and tail paralysis and forelimb weakness; 5, moribund state; and 6, death. The peak (waxing phase) occured around day 21. In this C57BL/6J mouse model, there was no waning phase as assessed in our laboratory up to day 41.

(46) The experiment was carried out with ten mice per group. The mice of each group were treating with either of: PBS; PepReg (SEQ ID NO: 3); or PepScra (SEQ ID NO: 14).

(47) The dosing was of 50 g of peptide per mice and per day (i.e. about 2 mg/Kg per day). The peptide was administered by a subcutaneous injection.

(48) Disease protection was associated with reduced infiltration of IL17+ and IFNg+ T helper CD4+ cells and increased proportion of regulatory CD25+/foxP3 CD4+ cells in the central nervous system of the mice.

(49) As shown on FIG. 7, PepReg is capable of arresting disease development in the waxing phase (Day 15) and reducing disease extension in the plateau phase (Days 21 through to 35).

(50) The scrambled peptide was also beneficial, although less than PepReg. This result suggests that the amino acid composition rather than the sequence per se is important for the beneficial effect. This result has important implications for the development of peptidomimetics.

Example 8: Detection of Shed CD31 in Plasma from Patients Suffering from Atherothrombosis and in Unaffected Individuals

(51) The total amount of CD31, the amount of shed CD31 and the amount of spliced CD31 has been measured both in eleven individuals suffering from atherothrombosis and in twenty-three unaffected individuals.

(52) The group Atherothrombosis comprised eleven individuals suffering from chest pain even at rest and presenting an abnormal coronarography.

(53) The group No Atherothrombosis comprised twenty-three individuals. A sub-analysis was carried out on the group No Atherothrombosis, which was found to comprise: eight individuals presenting a normal coronarography and a normal carotid echodoppler in spite of chest pain; four individuals presenting a normal coronarography in spite of chest pain, but in whom atherosclerosis was detected by carotid echodoppler; and eleven individuals suffering from chest pain only on effort and presenting an abnormal coronarography (i.e. suffering from coronary atherosclerosis without thrombosis).

(54) The total amount of CD31, the amount of shed CD31 and the amount of spliced CD31 was determined as follows.

(55) 1. The total amount (1 l/test) of beads (E9, coupled with clone JC70A, DAKO) was transferred to a conical tube and centrifuged at 200 g for 5 minutes. The supernatant was carefully discarded and replaced with same amount of serum enhancement buffer (BD #51-9002150), and incubated at room temperature for 15 minutes.

(56) 2. The fluorescently-labeled antibody antibody mix (PE-WM59; FITC-HC1/6; PB-PECAM1.2) was prepared, each at 1 g/ml, 1 l each/condition.

(57) 3. 1 tube precondition was prepared, each containing 3 l of a standard dilution or a plasma sample. The reconstituted beads were centrifuged at 200 g for 5 minutes, the supernatant was discarded and the serum enhancement buffer was replaced with the fluorescently-labeled antibody mix. 3 l of this solution was distributed in each of the tubes containing the standard dilution and samples, and the solution incubated for 1 hour at 4 C. in the dark.

(58) 4. 150 l of Washing buffer (BD #51-9003797) were added to each tube, and the signal was acquired.

(59) As shown in the table below, the percentage of shed CD31 was higher in individuals suffering from atherothrombosis than in unaffected individuals, in spite of the fact that all individuals were suffering from chest pain.

(60) TABLE-US-00006 CD31 Plasma Level (ng/ml) total splice shed Atherothrombosis (n = 11) 11.55 0.7 7.02 2.69 18.57 2.67 No Atherothrombosis 11.58 0.49 5.26 1.850 6.31 1.85 (N = 23) T-test Prob >F 0.9756 0.0007 0.0006

(61) Total CD31 amounts were similar in the four groups, while the amount of shed CD31 and the amount of spliced CD31 were significantly different in each paired group comparison. Shed CD31 was increased in individuals with abnormal coronarography, with highest values in those suffering from atherothrombosis. Splice CD31 was still present in patients suffering from atherosclerosis without atherothrombosis, while it was almost undetectable in patients suffering from atherothrombosis.

(62) These results demonstrate that high levels of CD31 soluble splice variants associated with low levels of shed CD31 indicates that the patient suffers from non specific chest pain, eventually associated with carotid plaques. A slight increase of shed CD31 levels associated with normal or reduced levels of CD31 soluble splice variants indicates that the patient suffers from atherosclerosis. An important increase of shed CD31 levels associated with undetectable amounts of CD31 soluble splice variants indicates that the patient suffers from atherothrombosis.

Example 9: Discussion of the Results

(63) Dysimmune diseases are linked to lack of appropriate control of immune responses. Atherosclerosis and its complications are not only due to metabolic disturbances but are increasingly recognized as a dysimmune disease and an important current issue is the identification of interventional tools able to restore immunoregulation. It has been previously shown that atherothrombotic manifestations such as plaque rupture and thrombosis (Caligiuri et al. 2005 Arterioscler Thromb Vasc Biol 25:1659-1664) or aneurysm complication (Caligiuri et al. 2006 Arterioscler Thromb Vasc Biol 26:618-623) are associated with a significant reduction of CD31+ T lymphocytes in the peripheral blood. In these previous works, we documented that lack of CD31 signaling on lymphocytes elicited pro-atherothrombotic immune responses whilst the presence of CD31 on T cells was able to inhibit them.

(64) Here it is demonstrated that the assumed loss of the molecule on activated/memory T lymphocytes is actually incomplete and results from shedding of CD31 between the 5th and 6th extracellular Ig-like domains. CD31 shedding occurred immediately after cell activation on T lymphocytes and was accompanied by the accumulation of the truncated molecule in the supernatant together with trace levels of the spliced variant produced by the cells. This finding was unsuspected because all commercially available tests to detect plasma CD31 use antibodies directed to CD31 domains 1 to 5, and therefore cannot discriminate between the spliced variant (containing all the 6 extracellular domains) and the truncated (domains 1 to 5) forms of CD31. The subtractive immunosorbent assay described herein is able to discriminate between the two forms of soluble CD31 and precisely quantify the proportion of each of them in the plasma. This assay showed that the major part of plasma CD31 comprises domains 1 to 5 but lacks the membrane-proximal 6th domain, which remains anchored to blood CD31 dom1-lymphocytes. Therefore, it is proposed to refer to these lymphocytes as CD31shed rather that CD31 negative cells. Previous work in vitro had indicated that CD31 shedding at an unidentified position N-terminal from the transmembrane segment of the molecule can occur in endothelial cells undergoing apoptosis (Ilan et al. 2001. Faseb J 15:362-372). For the first time, it is shown herein that shedding is responsible for the CD31 (incomplete) loss on blood lymphocytes and that the circulating CD31 consists mainly of a truncated form derived from its cleavage between the Ig-like domains 5 and 6, rather than of the secreted spliced variant form. Genetic polymorphisms for CD31 have been described, but the predictive value of soluble CD31 levels was conflicting either in atherothrombosis or other dysimmune diseases. In fact, while the amount of the spliced form can be predicted by different genetic variants, the proportion of the form resulting from protein shedding is not determined by CD31 gene polymorphism. It is proposed that the disparity between the different studies was due to fact that circulating CD31 is a mixture of the genetic variant and of the truncated form and discrimination between the two forms of CD31 was not possible. The subtractive method described herein will allow the differentiation of the prognostic value determined by genetic variants of CD31 independently of that linked to CD31 shedding.

(65) The fact that CD31 is not completely lost on blood lymphocytes offers a unique opportunity to rescue its physiological immunoregulatory function by targeting the residual extracellular portion of the molecule. Indeed, it has been documented herein that this can be achieved by a homotypic peptide-based therapy, both in vitro and in vivo. Homophilic binding of this peptide dramatically enhanced the phosphorylation the CD31 686ITIM and inhibited their TCR-induced proliferation. Induction of CD31 ITIM phosphorylation by antibody-mediated cross-linking of CD31 and CD3 surface molecules was previously known to inhibit calcium mobilization induced by anti-CD3 antibodies in human T-cell lines. Remarkably, it has been found that targeting the juxta-membrane portion of CD31 whit the small homotypic peptide is as an efficient immunoregulatory strategy as it is the co-ligation of CD3 and CD28 with the distal portion of CD31 by cross-linking the molecules with antibodies. A selective small synthetic peptide strategy is obviously simpler and might also be safer than using large proteins, such as monoclonal antibodies and cross-linkers, for the biotherapy of immunological disorders (Isaacs 2007. Curr Opin Pharmacol 7:418-425).

(66) With this idea in mind, it was assessed whether the engagement of the CD31 molecular pathway by this peptide could attain effective suppression of antigen-driven lymphocyte responses in vitro and in vivo in the context of matched, histocompatible antigen-driven immune responses. Distal recall of an hapten-elicited specific immune response was suppressed in a dose-dependent manner by a single subcutaneous administration of the peptide. A similar protective effect of a single peptide shot was also observed in experimental autoimmune encephalomyelitis (a mouse model of multiple sclerosis) and lasted for up to 5 days.

(67) Consequently, it was evaluated whether rescuing of CD31-mediated immunoregulation by the synthetic peptide 551-574 could be employed in a biotherapy to fight atherothrombosis since CD31-mediated immunoregulation is typically lost in this disease. It was chosen to use the angiotensin-induced model of atherothrombosis because the abrupt acceleration of atherosclerotic plaque growth and the development of abdominal aortic aneurysms complicated by a thrombus in this model are produced simply by excess bioavailability of a physiological peptideangiotensin IIand does not require the use of surgery, gene transfer or high fat diet, each of which could considerably bias the interpretation of the results. The CD31-peptide prevented both the acceleration of plaque growth in the aortic root and the formation of atherothrombotic abdominal aortic aneurysms in this model. Such a dramatic protective effect was superior to any therapeutic molecule ever tested and equivalent to that achieved by genetic manipulation.

(68) Macrophages and lymphocytes represent the most important immune cells involved in the development of atherothrombosis. The local function of these cells injure the cellular and extracellular components of the arterial layers resulting in either plaque rupture and luminal thrombosis, when occurring in the fibrous cap of atherosclerotic plaques, or aneurysm formation and eventually rupture, when happening in the outer layers of the artery. Degradation of the extracellular matrix is essentially due to the activity of macrophage-derived matrix metalloproteases while death of arterial smooth muscle cells is putatively caused by T cell-mediated cytolysis. Remarkably, CD31+ T lymphocytes exert an important immunosuppressive function on both these phenomena which are conversely aggravated by CD31shed T cells. An aberrant reduction of CD31+ cells in patients with atherothrombosis underlies the defective immunoregulation observed in the disease. Here it is shown that the CD31-derived peptide 551-574 inhibits both the activity of matrix degrading enzymes and T cell-dependent cytolysis of the arterial wall cells. The immunoregulation conveyed by this peptide is as efficient as that exerted by immunoregulatory CD31+ T cells and hence may counterweight the loss of the physiologic CD31-dependent immunoregulation in human atherothrombosis.

(69) This is the first time that a peptide-based biotherapy is envisaged to correct the defective immunoregulation characteristic of atherothrombosis and prevent development of the disease in patients. In addition, such biotherapy may broaden over to other debilitating dysimmune diseases. In particular, experimental studies have suggested that CD31-signalling might play a protective role in rheumatoid arthritis (Wong et al. 2005. J Clin Immunol 25:19-28), multiple sclerosis (Graesser et al. 2002. J Clin Invest 109:383-392) and non-alchoolic fatty liver disease (Goel et al. 2007. Am J Physiol Gastrointest Liver Physiol 293:G1205-1214).

Example 10: Evaluation of Ninety Six Peptides According to the Invention

(70) The ninety six peptides consisting of a fragment having a sequence selected from the group consisting of: amino acids 2 to 23, 3 to 23, 4 to 23, 5 to 23, 6 to 23, 7 to 23, 8 to 23, 9 to 23, 10 to 23, 11 to 23, 12 to 23, 13 to 23, 14 to 23, 15 to 23, 16 to 23, 17 to 23, 18 to 23, 19 to 23, 20 to 23 and 21 to 23 of SEQ ID NO: 5; and amino acids 2 to 23, 3 to 23, 4 to 23, 5 to 23, 6 to 23, 7 to 23, 8 to 23, 9 to 23, 10 to 23, 11 to 23, 12 to 23, 13 to 23, 14 to 23, 15 to 23, 16 to 23 and 17 to 23 of SEQ ID NO: 6; and amino acids 1 to 22, 1 to 21, 1 to 20, 1 to 19, 1 to 18, 1 to 17, 1 to 16, 1 to 15, 1 to 14, 1 to 13, 1 to 12, 1 to 11, 1 to 10, 1 to 9, 1 to 8, 1 to 7, 1 to 6, 1 to 5, 1 to 4 and 1 to 3 of SEQ ID NO: 5; and amino acids 1 to 22, 1 to 21, 1 to 20, 1 to 19, 1 to 18, 1 to 17, 1 to 16, 1 to 15, 1 to 14, 1 to 13, 1 to 12, 1 to 11, 1 to 10, 1 to 9, 1 to 8, 1 to 7, 1 to 6, 1 to 5, 1 to 4 and 1 to 3 of SEQ ID NO: 6; and amino acids 2 to 22, 3 to 21, 4 to 20, 5 to 19, 6 to 18, 7 to 17, 8 to 16, 9 to 15, 10 to 14 and 11 to 13 of SEQ ID NO: 5; and amino acids 2 to 22, 3 to 21, 4 to 20, 5 to 19, 6 to 18, 7 to 17, 8 to 16, 9 to 15, 10 to 14 and 11 to 13 of SEQ ID NO: 6;
are tested for confirming that these peptides induce CD31-ITIM phosphorylation. CD31-ITIM phosphorylation is assessed by studying the effect of increasing doses of the peptides according to the invention on CD31-ITIM phosphorylation of cultured Jurkat cells stimulated with anti-CD3 and/or anti-CD28 antibodies, as described in Example 3.

(71) It is confirmed that the above peptides inhibit T-cell proliferation using the protocol described by Caligiuri et al. (2005 Arterioscler Thromb Vasc Biol 25:1659-1664).

(72) It is confirmed that the peptides can achieve effective immunosuppression using the protocol described in the Current Protocols in Immunology (2001, 4.0.1-4.0.2 Unit 4.2).

(73) The efficiency of the peptides for treating a thrombotic disorder is confirmed in the model of angiotensin II infusion into aged apolipoprotein E/ mice, which closely mimicks atherothrombosis in humans.

(74) The efficiency of the peptides for treating an autoimmune disorder is confirmed in the Experimental Autoimmune Encephalomyelitis (EAE) model, which is an animal model of Multiple Sclerosis, and in a model for rheumatoid arthritis.

Example 11: Evaluation of the Peptides According to the Invention in an Animal Model of the Rheumatoid Arthritis (RA)

(75) It is confirmed that the peptides according to the invention (e.g. PepReg and/or the peptides described in Example 10) are capable of arresting disease development and/or reducing disease extension in an animal model of Rheumatoid arthritis (RA).

(76) Rheumatoid arthritis (RA) is a chronic and systemic inflammatory autoimmune disorder that causes the immune system to attack the joints. The disease is characterized by aggressive synovial hyperplasia (pannus formation) and inflammation (synovitis), which, if left untreated, lead to progressive destruction of joint cartilage and bone. The destructive lesions result from both immune responses and non-antigen-specific innate inflammatory processes. Several studies have shown that CD31 lays a critical role in this disease since the disease onset and progression is accelerated in its absence.

(77) DBA/1 mice are used in this experiment. Induction of RA is initiated on 12 week-old mice. On day 0, mice are immunized intradermally at the base of the tail with 150 g of bovine type II collagen (CII) emulsified with an equal volume of Freund's complete adjuvant containing 200 g of H37RA Mycobacterium tuberculosis. On day 21, mice are given a booster (intradermal injection of 150 g of bovine CII in Freund's incomplete adjuvant). Simultaneously, mice receive an intravenous injection of 50 g of LPS. Mice are followed up for two months. Following immunization, the animals develop an autoimmune polyarthritis that is characterized by severe cartilage and bone erosions. Mouse collagen-induced RA shares several clinical, histopathological and immunological features with human RA.

(78) The peptides according to the invention are administered following one of the below treatment schemes. Scheme 1: Preventive administration. Two doses (50 and 100 mg/kg) of the peptide is administered to the mice one day before the CII immunization and than either daily, or twice a week, or weekly, for the study period (2 months). A peptide with a scrambled sequence (i.e. a peptide comprising the same amino acids as the peptide according to the invention, but not the same sequence) is administered to a control group of mice. Treatment is pursued for one month. Scheme 2: Curative administration. The peptide (50 and 100 mg/kg) and the scramble peptide are administered to mice after the beginning of the symptoms, either daily, or twice a week or weekly until the end of the study. Equivalent groups of mice are kept in conventional housing facilities and are bled weekly from a tail vein to monitor bleeding time and specific pathogen antibody raise in sera (CDTA, Orleans).

(79) Arthritis development is monitored by physical examination 3 times per week. Inflammation in each of the 4 paws is graded on a scale of 0 to 3, and the scores for the 4 paws will be cumulated (yielding a maximum score of 12 per mouse). The arthritis index is calculated by dividing the total score in the experimental mice by the number of arthritic mice.

(80) Lesions in the joints are also followed. Ankle joints of mice are excised 6 weeks after immunization and fixed in 10% buffered formalin, decalcified in 10% EDTA, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. The intensity of synovial hyperplasia, cellular inflammation, and pannus formation is evaluated, and arthritis is graded in a blinded manner on a scale of 0 to 4.

(81) Immunohistochemistry is further used to track and to phenotype inflammatory cells infiltrated in the joints.

(82) The immunoregulation status is evaluated by measuring levels of serum IgG1 and IgG2a to CII. The measurement is performed by enzyme-linked immunosorbent assay (ELISA). The proliferation of T cells isolated from draining lymph nodes and the spleen is tested by the incorporation of .sup.3H thymidine in response to CII-loaded dendritic cells.

(83) Cell populations in the lymphoid organs and in the synovia is analyzed by flow cytometry.

(84) This experiment allows confirming that continuous administration of the peptides according to the invention prevents onset of RA. The curative phase allows evaluating the therapeutic potential of the peptides for treating RA in patients that do not respond to the current biologicals (i.e. 40% of the patients). Dose and frequency of the administrations able to drive a regression of the inflammatory cells in the joints, and regression of the clinical score, are also determined.