Complexes of types of MHC class II that bind to collagen type II peptides and their use on diagnosis and treatment

Abstract

Novel complexes of peptides from human collagen type II and types of MHC class II associated with rheumatoid arthritis are provided. There is also provided novel therapies and methods for diagnosis of rheumatoid arthritis.

Claims

1. A complex of a human recombinant MHC class II protein fragment of HLA-DRB1 *1001 and the synthesized or recombinantly produced peptide consisting of SEQ ID NO 46, wherein said fragment of said human recombinant MEW class II protein excludes at least the transmembrane domains of said MEW class II protein and is stabilized by at least one of: retaining parts of the alpha 2 and beta 2 inner domains, introducing a leucine zipper structure to the inner domains, introducing cysteine residues that form disulfide bridges linking the alpha and the beta chains, and introducing a peptide in the peptide binding groove that is covalently linked to an elongation of the beta domain.

2. A kit of parts comprising the complex of claim 1.

3. The kit according to claim 2 additionally comprising at least one cell culture vessel.

4. The kit according to claim 2 comprising means for detecting the expression of at least one protein selected from the group consisting of CD3, CD4, Foxp3, CD25, TNF alpha, interferon gamma, interleukin-17A, interleukin-17F, CD154, CD69, Ki 67, interleukin-2, interleukin-13 and interleukin-10.

5. The kit according to claim 4 where the means for detection is an antibody against said protein.

6. The kit according to claim 4 where the means for detection is a set of primers for RT-PCR.

7. A composition comprising the complex of claim 1 and an additive or excipient.

8. The complex of claim 1, wherein said complex is stable in solution.

9. The complex of claim 8, wherein said human recombinant WIC class II protein fragment excluding said transmembrane domains is stabilized by retaining parts of the alpha 2 and beta 2 inner domains.

10. The complex of claim 8, wherein the inner domains of said human recombinant WIC class II protein fragment excluding said transmembrane domains are stabilized by the introduction of a leucine zipper structure.

11. The complex of claim 8, wherein said human recombinant WIC class II protein fragment excluding said transmembrane domains is stabilized by introducing cysteine residues that form disulfide bridges linking the alpha and the beta chains.

12. The complex of claim 8, wherein said human recombinant WIC class II protein fragment excluding said transmembrane domains is stabilized by introducing a peptide in the peptide binding groove that is covalently linked to an elongation of the beta domain.

Description

FIGURES

(1) FIG. 1. (A) CD4+ T cell responses to alpha-enolase (ENO) peptides in a HLA-DR*0401-positive patient. (B) CD4+ T cell responses to collagen-II (CII) peptides in a HLA-DR*1001-positive patient. PBMC from patients were cultured during 5 days with medium (white bar) or with two peptides (grey bars) or two corresponding citrullinated (CIT) variants (black bars). FACS analysis of co-expression of CD154 together with IFN-gamma and IL-17 on CD4+ T cells was carried out after in vitro stimulation of PBMCs with the indicated peptides (peptide reference numbers are shown). Further details are provided in Examples 13 and 15.

(2) FIG. 2. PBMCs from a HLA-DRB1*0401-positive rheumatoid arthritis patient was stimulated with different enolase peptides for 5 days. Intracellular staining for cytokines was performed 6 hrs after restimulation and addition of Brefeldin A. The positive control (HA) primarily generate an IFNgamma response. The arginine versions of the a-enolase peptides does yield a response in this patient, while the citrullinated a-enolase peptides results in secretion of IL17A (all three peptides) and IFN gamma (one peptide). Further details are proved in Examples 4, 6 and 11.

(3) FIG. 3. An HLA-DRB1*0401 tetramer loaded with the citrinullated enolase peptide 313 (SEQ ID NO 23) and labeled with the fluorochrome PE is used to stain PBMC from an *0401-positive rheumatoid arthritis patient. The cells in the upper right quadrant are T-cells with a T-cell receptor recognizing this particular peptide-MHC complex. Further details are proved in Examples 16 and 17.

EXAMPLES

Example 1

(4) The inventors have identified peptides from human collagen type II, alpha-enolase and vimentin that specifically bind to the variants of MHC that are associated with a risk for developing rheumatoid arthritis and with a presence of antibodies to specific citrullinated proteins. These peptides bind to MHC with genotypes HLA DRB1 0101, 0401, 0404, 0405, 0408 and 1001 as indicated in tables 1, 2 and 3. Interestingly, none of the peptides bind to MHC with genotypes HLA-DR 03 and 0402, genotypes that are not associated with developing rheumatoid arthritis.

(5) Binding of the peptides to recombinant MHC class II protein of various alleles has been confirmed in an in vitro binding assay and the results are shown in Tables 1, 2 and 3. In table 1, 2 and 3, the listing of the allele in the column for the allele indicates binding. Thus 0401 in the Binds to 0401 column indicates binding of the peptide to MHC class II of the HLA DRB1 0401 genotype. no or no data indicates no binding to a protein of that genotype. Empty boxes indicate no binding or that binding has not been determined. In the tables it is also indicated if the peptide has a citrulline residue, where yes indicates the presence of a citrulline residue and no indicates the absence of a citrulline residue.

(6) In addition, the inventors have identified previously unknown peptides from alpha-enolase (SEQ ID NO 15 to 22 and 24), and one citrullinated peptide from collagen type II (SEQ ID NO 30) that bind specifically HLA-DRB1 0401, were the non-citrullinated peptide did not bind. Thus for these peptides no binding to HLA-DR0401 was observed when citrulline was exchanged with the original amino acid arginine (Table 4). No binding of the citrullinated peptides was seen to HLA-DR 03 or 02 variants.

(7) Thus peptides identified and presented in Table 4 bind to those variants of HLA-DR that associate with rheumatoid arthritis and anti-alpha-enolase and anti-collagen type II antibody responses (HLA-DRB1 0401 and/or 0404).

(8) TABLE-US-00001 TABLE 1 Binds Binds Binds Binds Binds Binds SEQ Peptide to to to to to to ID Protein ref no 0101 0401 0404 0405 0408 1001 Citrinullation NO a-enolase 1 *0101 no *0404 *0405 no *1001 no 1 a-enolase 6 *0101 no *0404 *0405 no no no 2 a-enolase 12 *0101 no no *0405 *0408 *1001 no 3 a-enolase 13 *0101 no *0404 no no no no 4 a-enolase 14 *0101 no *0404 no *0408 no no 5 a-enolase 15 *0101 no *0404 no *0408 no no 6 a-enolase 26 *0101 *0401 *0404 *0405 *0408 no no 7 a-enolase 34 *0101 *0401 *0404 *0405 no no no 8 a-enolase 38 *0101 *0401 no *0405 *0408 no no 9 a-enolase 57 *0101 *0401 *0404 *0405 *0408 *1001 no 10 a-enolase 66 *0101 *0401 *0404 *0405 *0408 no no 11 a-enolase 80 *0101 no no *0405 *0408 no no 12 a-enolase 81 *0101 no *0404 no no no no 13 a-enolase 287 *0101 *0401 *0404 *0405 no *1001 yes 14 a-enolase 289 no *0401 no *0405 no no yes 15 a-enolase 291 *0101 *0401 *0404 *0405 no *1001 yes 16 a-enolase 292 *0101 *0401 *0404 *0405 no no yes 17 a-enolase 297 no *0401 no *0405 no no yes 18 a-enolase 300 no *0401 no *0405 no no yes 19 a-enolase 301 no *0401 *0404 *0405 no no yes 20 a-enolase 302 no *0401 *0404 *0405 no no yes 21 a-enolase 305 no *0401 no *0405 no no yes 22 a-enolase 313 *0101 *0401 *0404 *0405 no no yes 23 a-enolase 324 no *0401 *0404 *0405 no no yes 24 a-enolase 36 *0101 no no *0405 no no no 32 a-enolase 45 *0101 *0401 *0404 *0405 *0408 no no 33 a-enolase 49 no *0401 *0404 *0405 *0408 no no 34 a-enolase 69 no no no *0405 no no no 35 a-enolase 299 no *0401 no *0405 *0408 no yes 36

(9) TABLE-US-00002 TABLE 2 Peptide Binds Binds Binds Binds Binds Binds SEQ ref to to to to to to ID Protein no 0101 0401 0404 0405 0408 1001 Citrinullation NO type II collagen 195 *0401 *0404 no 25 type II collagen 242 *0101 *0401 *0404 *0405 no 26 type II collagen 257 *0404 no 27 type II collagen 137 *0101 *0401 *0404 no 28 type II collagen 138 *0101 *0401 no 29 type II collagen 386 *0401 *0404 yes 30 type II collagen 411 *0401 *0404 *0405 yes 31 type II collagen 100 *0404 no 37 type II collagen 166 *0404 no 38 type II collagen 178 *0405 no 39 type II collagen 224 *0404 no 40 type II collagen 241 no *0404 *0405 no 41 type II collagen 263 *1001 no 42 type II collagen 268 *1001 no 43 type II collagen 272 *0405 no 44 type II collagen 332 *0401 *0404 *0405 yes 45 type II collagen 357 *1001 yes 46 type II collagen 362 *1001 yes 47 type II collagen 384 *0408 *1001 yes 48

(10) TABLE-US-00003 TABLE 3 Binds Binds Binds Binds Binds Binds SEQ Peptide to to to to to to ID Protein ref no 0101 0401 0404 0405 0408 1001 Citrinullation NO vimentin 1 *0101 *0401 *0404 no no 49 vimentin 3 *0101 *0401 *0404 *0405 no 50 vimentin 6 no 51 vimentin 13 *0101 no no 52 vimentin 14 *0404 no 53 vimentin 16 no *0401 no 54 vimentin 20 *0401 no 55 vimentin 27 *0101 no no 56 vimentin 32 *0101 *0404 no 57 vimentin 33 no *0401 *0404 no no 58 vimentin 34 no *0401 no no 59 vimentin 41 *0401 *0404 *0405 no 60 vimentin 42 *0101 *0401 no no 61 vimentin 44 *0404 no 62 vimentin 55 *0101 *0401 no 63 vimentin 59 *0401 no 64 vimentin 64 *0101 *0401 *0405 no 65 vimentin 71 *0401 no 66 vimentin 77 *0101 *0405 no 67 vimentin 79 *0101 no *0404 no 68 vimentin 81 no *0401 no no no 69 vimentin 92 *0404 no 70 vimentin 93 no *0401 *0404 yes 71 vimentin 95 *0101 *0401 yes 72 vimentin 98 *0401 *0405 yes 73 vimentin 99 *0101 *0401 yes 74 vimentin 104 *0101 *0401 *0404 no yes 75 vimentin 105 *0101 *0401 no no yes 76 vimentin 106 *0101 *0401 *0404 *0405 yes 77 vimentin 108 no *0401 no no yes 78 vimentin 111 *0101 *0401 No no yes 79 vimentin 117 *0101 No no yes 80 vimentin 132 *0101 yes 81 vimentin 133 no *0401 No no yes 82 vimentin 145 *0401 yes 83 vimentin 150 *0401 yes 84 vimentin 156 *0101 yes 85 vimentin 159 *0401 No yes 86 vimentin 166 no no *0404 no yes 87

(11) TABLE-US-00004 TABLE 4 Citrullination Corresponding non- necessary for citrullinated peptide binding to these Peptide Binds to HLA binds to HLA HLA SEQ Protein no. subtype Citrullination subtype: genotypes: ID NO -enolase 287 *0401 yes 0101, 0404 0401 14 -enolase 289 *0401 yes 0401 15 -enolase 291 *0401, *0404 yes 0404 0401 16 -enolase 292 *0401 yes 0401 17 -enolase 297 *0401 yes 0101 0401 18 -enolase 300 *0401 yes 0401 19 -enolase 301 *0401 yes 0404 0401 20 -enolase 302 *0401 yes 0401 21 -enolase 305 *0401 yes 0401 22 -enolase 313 *0401 yes 0101, 0401 23 -enolase 324 *0401 yes 0401 24 type II collagen 195 *0404 no na 25 type II collagen 242 *0401, *0404 no na 26 type II collagen 257 *0404 no na 27 type II collagen 137 *0401 no na 28 type II collagen 138 *0401 no na 29 type II collagen 386 *0401 yes 0404 0401 30 type II collagen 411 *0401, *0404 yes 0401, 0404 31

Example 2

(12) PBMCs were isolated from several rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 14 (peptide reference number 287) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place.

Example 3

(13) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 15 (peptide reference number 289) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. In addition, the non-citrinullated version of the peptide was used as a negative control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. The non-citrinullated peptide did not result in activation.

Example 4

(14) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 16 (peptide reference number 291) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. Data from one patient is shown in FIG. 2, which shows a strong increase in IL-17 compared to the controls. Data from one patient is shown in FIG. 2. The corresponding non-citrinullated peptide (peptide reference number 6, SEQ ID NO 2) did not result in a response, which could be expected because SEQ ID NO 2 does not bind MHC with genotype 0401 (Table 1).

Example 5

(15) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 17 (peptide reference number 292) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. In addition, the non-citrinullated version of the peptide was used as a negative control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. The non-citrinullated peptide did not result in activation.

Example 6

(16) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 18 (peptide reference number 297) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. Data from one patient is shown in FIG. 2 which shows a strong increase in IFN gamma and IL-17 compared to the controls. Data from one patient is shown in FIG. 2.

Example 7

(17) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 19 (peptide reference number 300) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. In addition, the non-citrinullated version of the peptide was used as a negative control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. The non-citrinullated peptide did not result in activation.

Example 8

(18) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 20 (peptide reference number 301) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place.

Example 9

(19) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 21 (peptide reference number 302) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. In addition, the non-citrinullated version of the peptide was used as a negative control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. The non-citrinullated peptide did not result in activation.

Example 10

(20) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The non-citrinullated enolase peptide with SEQ ID NO 34 (peptide reference number 49) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place.

Example 11

(21) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 22 (peptide reference number 305) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. Data from one patient is shown in FIG. 2.

Example 12

(22) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The non-citrinullated enolase peptide with SEQ ID NO 11 (peptide reference number 66) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place.

Example 13

(23) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 23 (peptide reference number 313) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. Data from one patient is shown in FIG. 1A.

Example 14

(24) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-0401. The citrinullated enolase peptide with SEQ ID NO 24 (peptide reference number 324) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. In addition, the non-citrinullated version of the peptide was used as a negative control. The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. The non-citrinullated peptide did not result in activation.

Example 15

(25) PBMCs were isolated from rheumatoid arthritis patients with HLA genotype HLA DRB1-1001. The citrinullated collagen type II-peptide with SEQ ID NO 46 (peptide reference number 357) was added to the cells. As a negative control, cells were incubated without peptide. An influenza peptide was used as a positive control. In addition, the non-citrinullated version of the peptide was used as a negative control (peptide reference number 148). The peptides were added at concentrations ranging from 5 to 50 ug/ml and after a 5-day incubation in a CO.sub.2 incubator, cells were fixed, permeabilized and intracellular cytokine staining were performed. Antibodies for CD3, CD4, viable cells, CD154 (an activation marker) and cytokines IFNgamma, IL-17A and TNF alpha were used for analysis in a multiparameter flow cytometry panel. Cells from all patients exposed to the citrinullated peptide double stained positive for both CD154 and at least one cytokine, which indicates that activation of T-cells has taken place. The non-citrinullated peptide did not result in activation. Data from one patient is shown in FIG. 1 B.

Examples 2-15

(26) Data is from examples 2-15 are summarized in table 5. Activation of T-cells means that all patients are positive for at least one cytokine. This is not necessarily indicated in FIGS. 1A and B, which show selected data from individual patients for selected cytokines.

(27) TABLE-US-00005 TABLE 5 Binds MHC SEQ Peptide HLA in Activation ID Protein ref no tested Citrinullation vitro of T-cells NO a-enolase 66 *0401 no yes yes 11 a-enolase 287 *0401 yes yes yes 14 a-enolase 289 *0401 yes yes yes 15 a-enolase 291 *0401 yes yes yes 16 a-enolase 292 *0401 yes yes yes 17 a-enolase 297 *0401 yes yes yes 18 a-enolase 300 *0401 yes yes yes 19 a-enolase 301 *0401 yes yes yes 20 a-enolase 302 *0401 yes yes yes 21 a-enolase 305 *0401 yes yes yes 22 a-enolase 313 *0401 yes yes yes 23 a-enolase 324 *0401 yes yes yes 24 a-enolase 49 *0401 no yes yes 34 type II 357 *1001 yes yes yes 46 collagen

Example 16

(28) Production of peptide-MHC complex: Recombinant HLA-DRB1 *0401 was produced as described in Novak et al, J Clin Invest 1999; 104: R63-7. Briefly soluble DR0401 was purified from insect cell culture supernatants and biotinylated at a sequence specific site using biotin ligase (Avidity) prior to dialysis into phosphate storage buffer. The biotinylated monomer was loaded with 0.2 mg/ml of the peptide with SEQ ID NO 23 (peptide reference number 313) by incubation at 37 C. for 72 hours in the presence of 2.5 mg/ml n-octyl-beta-D-glucopyranoside and 1 mM Prefabloc SC (Sigma-Aldrich).

Example 17

(29) Production and use of tetramers. The peptide-MHC complexes in Example 16 were conjugated to tetramers using R-phycoerythrin-streptavidin (Invitrogen) at a molar ratio of 8:1. PBMCs from one patient with rheumatoid arthritis, HLA type *0401, was stained using the tetramers and immediately analyzed by FACS. Several CD4 positive T-helper cells were identified. It could be estimated that about 1 out of 350 000 CD4 positive cells reacted with the peptide, which indicates as significant immune response. Data is presented in FIG. 3, where cells in the upper-right corner bind to the peptide-MHC complex.