International Immunology, Vol. 3, No. 10, pp.

965-972

© 1991 Oxford University Press 0953-8178/91 $3.00

Epitope specificity and MHC restriction of rheumatoid arthritis synovial T cell clones which recognize a mycobacterial 65 kDa heat shock protein J. S. Hill Gaston, Paul F. Life, Ruurd van der Zee 1 , Peter J. Jenner2, M. Joseph Colston2, Susan Tonks3, and Paul A. Bacon

Key words: heat shock protein, HLA-DP, rheumatoid arthritis, T cell clone

Abstract CD4+ T cell clones specific for the mycobacterial hsp 65 were obtained from synovial fluid of a DR4 homozygous rheumatoid arthritis (RA) patient. A stimulatory epitope was defined using both deletion mutants of the mycobacterial hsp 65 and synthetic peptides and proved to be in a highly conserved region of the molecule. Despite this, however, there was no recognition by these clones of either the recombinant human homologue of mycobacterial hsp 65, P60, nor of a synthetic peptide containing an amino acid sequence from P60 corresponding to the epitope defined in the mycobacterial hsp 65. When the pattern of HLA restriction shown by the hsp-65-specific T cell clones was investigated, all clones tested proved to be restricted by HLA-DP rather than the more usual HLA-DR. Inhibition experiments suggested that this restriction also applied to the polyclonal synovial T cell response to hsp 65, but not to other antigens. Exclusive restriction of T cell recognition of an antigen by HLA-DP has not been reported previously, and strongly suggests that in this case the T cell repertoire for recognizing hsp 65 in the context of DR4 is deficient. Such an association between DR4 and the inability to respond to an immunodominant bacterial antigen may have implications for the pathogenesis of RA.

Introduction There is a strong association between rheumatoid arthritis (RA) and the class II MHC antigen HLA-DR4 (1,2). Since MHC antigens govern T cell recognition (3) it is assumed that susceptibility to RA relates to the response mounted by DR4+ individuals to a critical antigen, as yet undefined. A possible candidate antigen has been suggested by studies in rodents which have clearly implicated the mycobacterial 65 kDa heat shock protein (hsp 65) in the pathogenesis of experimental arthritis (4 - 6). T cells specific for this antigen have also been isolated from human arthritic joints (7,8). The present study was designed to investigate the epitope(s) recognized by hsp-65specific T cell clones from an DR4 + patient with RA. In particular we wished to determine whether the epitope recognized in the

mycobacterial hsp 65 was also shared by the homologous human hsp. It has been suggested previously that the high degree of conservation in the amino acid sequences of hsp from both pro- and eukaryotic organisms could allow T cells specific for a bacterial hsp to cross-react with the homologous human hsp, and in this way give rise to an autoimmune response (9). Using synthetic peptides we have recently defined a DR3-restricted antigenic epitope recognized by synovial fluid (SF) T cells from a patient with a self-limiting reactive arthritis (10). In that case the epitope recognized proved to be in a non-conserved part of the mycobacterial hsp 65 sequence, and the T cell clones did not display autoreactivity. We now report the characterization of hsp-65-specific T cell clones from an RA patient homozygous

Correspondence to: J. S. H. Gaston Transmitting editor: S. H. E. Kaufman

Received 11 April 1991, accepted 13 June 1991

Downloaded from http://intimm.oxfordjournals.org/ at Georgetown University on May 22, 2015

Department of Rheumatology, University of Birmingham, Birmingham B15 2TJ, UK National Institute of Public Health and Environmental Protection, PO Box 1, 3720 BA Bilthoven, The Netherlands 2 National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, UK 3 Tissue Antigen Laboratory, ICRF, Lincoln's Inn Fields, London WC2A 3PX, UK

966

Rheumatoid arthritis synovial T cell clones

for HLA-DR4. Surprisingly, recognition of hsp 65 appeared to be exclusively restricted by HLA-DP rather than by HLA-DR. The critical residues determining this DP-restricted T cell response have also been defined. Although the epitope proved to reside in a highly conserved part of the hsp 65 sequence, T cells clearly distinguished between the self and mycobacterial antigens.

Methods Patient SF was obtained at therapeutic aspiration from the knee of a patient with classical (11) rheumatoid-factor-positive, erosive RA which had been present for 14 months.

The method used and the bacterial antigens tested have been described previously in detail (7). Briefly, MC were separated from heparinized SF and PB by centrifugation over Ficoll-Paque; prior to this the SF was incubated for 30 min at 37°C with hyaluronidase (10U/ml). The cells were washed twice and resuspended in RPMI 1640 culture medium containing 2 mM glutamine, 100 nglm\ streptomycin, 100 U/ml penicillin, and 10% heat-inactivated A+ human serum (complete medium). Cells were cultured with antigen for 6 days in 0.2 ml round-bottomed 96-well microtitre plates at 5 x 104 cells/well and [3H]thymidine incorporation over the last 18 h of culture measured using standard liquid scintillation techniques.

Blocking of proliferation with mAbs mAbs L243 (anti-DR), Leu 10 (anti-DQ) and B721.2 (anti-DP) (all at 25 uglrrft) were obtained from Becton Dickinson. All mAbs were tested at several dilutions and were present throughout the proliferation assay; results are shown for the dilution giving maximum specific inhibition. Control experiments established that these concentrations did not inhibit T cell responses to PHA and IL-2.

T cell clones

Results

All antigen-specific clones were isolated and expanded essentially as previously described (10); T cell blasts, stimulated with the relevant antigen for 6 days, were cloned by limiting dilution in 20 /il Terasaki wells using autologous irradiated MC (106/ml), 100 U/ml recombinant interleukin 2 (IL-2) (Cetus, Emeryville, CA) and antigen. Complete medium was supplemented with 1 % non-essential amino acids and 1 % sodium pyruvate. Resulting clones were expanded by restimulation at 10 - 1 4 day intervals with allogeneic MC and 1 /ig/ml phytohaemagglutinin (PHA) (HA16, Wellcome, Beckenham, UK). IL-2 was added every 3 - 4 days. The phenotype of the clones was determined by staining with mAbs and flow cytometry. Specificity of the clones was examined at least 8 days after stimulation with PHA by coculturing 1 - 2 x 104 T cells for 3 days with 105 irradiated (3000 rad) autologous MC as antigen-presenting cells (APC). In some experiments MC were replaced by irradiated (6000 rad) Epstein -Barr virus-transformed lymphoblastoid cells (LCL). [3H]Thymidine incorporation during the last 18 h of culture was measured. LCL derived for homozygous typing cells expressing subtypes of HLA-DR4 (12) were obtained from the European Collection of Animal Cell Culture, Porton Down; an LCL was also derived from the patient SFMC by infecting the cells with Epstein - Barr-virus-containing supernatants.of the B95-8 cell line followed by culture in the presence of cyclosporin A (1 jig/ml).

Properties of synovial hsp-65-specific T cell clones

Antigens and synthetic peptides used in epitope mapping Purified protein derivative of tuberculin (PPD) was obtained from Statens Seruminstitut, Copenhagen, Denmark. The recombinant

Since immune responses to the mycobacterial hsp 65 have been implicated in experimental arthritis and demonstrated in human reactive arthritis (4 - 8), we screened SF and PB T cells from RA patients for recognition of this antigen. SF, but not PB T cells from a DR4 homozygous patient CF with active, classical disease responded vigorously; in contrast, responses to other bacterial antigens tested were relatively similar in both compartments (Table 1). Hsp-65-specific T cell clones were derived by limiting dilution of SF T cell blasts stimulated for 6 days with PPD from Mycobacterium tuberculosis (which contains hsp 65) and cloned

Table 1 . Proliferative response to bacterial antigens Stimulus

[3H]Thymidine incorporation (c.p.m. ± SEM) SF

Medium 5723 ± 405 M. leprae hsp 65 20901 ± 2129

PB 1882 ± 5 8 3 3752 ± 174

Clone 519 96 ± 6 25823 ± 1 3 1 1

Salmonella

22562 ±3618 10219 ± 1458

68 ± 5

E. coli GroEL Yersinia PPD

ND ND 15266 ± 2927 13321 ± 1343 60452 ± 3438 37249 ± 1380

46 ± 7 ND ND

Proliferative responses of polyclonal MC from SF and PB, and of a SF-derived T cell clone (0.5 cells/well), to recombinant M.leprae hsp 65, M.tuberculosis PPD and E.coli GroEL (all at 10 ^g/ml); heat-killed Yersinia enterocolitica 0:3 and irradiated Salmonella agona were used at 2 x 106 and 2 x 108 organisms/ml respectively.

Downloaded from http://intimm.oxfordjournals.org/ at Georgetown University on May 22, 2015

SF and peripheral blood (PB) mononuclear cell (MC) proliferation assays

65 kDa antigen of Mycobacterium leprae was prepared as described (13). Suspensions of heat-killed Yersinia enterocolitica 0:3 and irradiated Salmonella agona were used as described (7). Escherichia coli GroEL was a kind gift from Dr Douglas Young, Hammersmith Hospital, London, UK. Deletion mutants of Mycobacterium bovis BCG hsp 65 were constructed as described (4); pEX2, a plasmid expressing beta-galactosidase only was used as a control. The human hsp 60 was expressed in E. coli as a beta-galactosidase fusion protein from plasmid pRH710. This pEX2-derived plasmid was constructed from a phage Iambda-gt11 clone containing the human gene (14) which was obtained through Dr R. A. Young (Boston). For assay the protein was dissolved at 1 mg/ml in PBS containing 0.45% SDS. Synthetic peptides were prepared and used in assays of T cell proliferation as described in detail (10); peptides were solubilized in TFA, applied to nitrocellulose, and the TFA removed by evaporation under vacuum. In later experiments peptides solubilized in PBS were used and tested directly.

Rheumatoid arthritis synovial T cell clones using recombinant M. leprae hsp 65. Seventy resulting clones were screened for the ability to respond to mycobacterial hsp 65; 40% of those seeded at 1 or 0.5 cells/well (n = 39) were positive. The specificity of a representative clone is demonstrated in Table 1; this clone, and all clones described subsequently in this paper, were shown by flow cytometry to have a CD4 + .CD8-, TCRde/te-phenotype. There was response to recombinant M. leprae hsp 65 but not to the homologous hsp 65 of E. coli (GroEL), or to Salmonella antigens. Further definition of the specificity of the clones was achieved using deletion mutants of M. bovis BCG hsp 65 (Table 2). All clones tested recognized deletion mutants containing amino acids 424-540 or 1 -481 to the same extent as the intact molecule, showing that the epitope recognized was contained within residues 424-481.

When 106 synthetic peptides spanning the complete M. leprae hsp 65 sequence were tested, only two overlapping peptides were found to be stimulatory, namely peptides 91 and 92 (amino acids 451 - 466 and 456 - 471 respectively) (Table 3). However, the clones differed in their ability to respond to these two peptides; only clone CF13 recognized peptide 91, whereas the other clones responded to peptide 92. Detailed analysis was carried out using clone CF13. As a reflection of the highly conserved nature of the hsp 65 family in bacteria and eurkayotes (15), peptide 91 was noted to have 9/16 and 7/16 amino acids in common with the equivalent regions in E. coli GroEL and the human homologue P60 respectively (Table 4), as well as other

Table 2. Response of hsp-65-specific T cell clone CF13 to deletion mutants of M. bovis BCG hsp 65 Stimulus

Amino acids

[3H]Thymidine incorporation (c.p.m. ± SEM)

Medium M. leprae hsp 65 pRIB1404 pRIB1421 PRIB1426 pRIB1444 pRIB1453 pRIB1302 pEX2

—

178 48,100 39,601 40,777 40,320 55,359 48,451 37,398 883

1 -541 2-540 62-540 171-540 303-540 424-540 1-481 -

± ± ± ± ± ± ± ±

21 2947 3860 3197 2459 3929 1830 455 135

Table 3. Response of hsp-65-specific T cell clone CF13 to synthetic peptides spanning amino acids 441 - 4 7 5 of M. leprae hsp 65 hsp 65 pep89 pep90 pep91 pep92 pep93

1-541 441-456 446-461 451 -466 456-471 460-475

29,985 112 262 8475 434 118

± ± ± =t ± ±

535 9 59 2167 10 27

Synthetic peptides were used at a final concentration of 20 pg/ml. Both peptides and hsp 65 were applied to nitrocellulose as previously described (9).

conservative changes. However, E.coli GroEL was not recognized by any clone (see Table 1), suggesting that residues critical to the epitope were not conserved between M. leprae and E. coli. Comparison of the mycobacterial, human and E. coli sequences in this region does reveal that at the C-terminal region of peptide 91 there are two residues (464 and 465) which are identical in the mycobacterial and human sequences, but have non-conservative changes in the E. coli sequence. Since the surrounding residues show conservative changes in the mycobacterial and human sequences, it was possible that the clone recognized an epitope common to the mycobacterial hsp 65 and the human homologue P60, but absent from E. coli GroEL. To test directly for this possibility further synthetic peptides were synthesized in order to define the epitope more precisely. The results are shown in Figure 1. The two conserved C-terminal amino acids of peptide 91, residues 464 and 465, were critical for T cell recognition since no peptide lacking these was recognized. In contrast, N-terminal residues could be removed as far as residue 455 without losing recognition, although the shortest of these peptides was only active at the highest concentration tested. Having shown that the minimal epitope for CF13 was indeed located at the C-terminal of peptide 91, in a region where the human and mycobacterial sequences are very similar, the ability of the clone to recognize either a synthetic peptide containing the human sequence homologous to peptide 91, or intact recombinant human hsp 60 fusion protein was tested. Neither of these proved stimulatory (Fig. 1, peptide 12 and hsp 60). Thus this clone did not recognize an epitope present in the self hsp. Likewise clones which recognized peptide 92 also failed to respond to human hsp 60 (data not shown). MHC restriction of hsp-65-specific clones The HLA restriction of these clones derived from a DR4 homozygous individual was investigated by using HLA-DR homozygous LCL expressing subtypes of DR4 to present hsp 65. Only LCL LS40 (DR4,Dw14 + ) was effective (Fig. 2A and B), although each of the LCL was shown to be capable to antigen presentation to T cell clones specific for bacterial antigens other than mycobacterial hsp 65 (data not shown). These data would have been consistent with restriction of the response to hsp 65 by Dw14. However, when testing the ability of HLA-specific mAbs to block proliferation, inhibition by an anti-DP (B721.2) antibody and not by anti-DR (L243) was clearly shown for 3/3 clones tested (data for CF519 and CF13 are shown in Fig. 3A). In control experiments using DR4,Dw4-restricted Vers/n/a-specific T cell clones (29), the anti-DR antibody effectively blocked the responses, while the anti-DP antibody had no effect (Fig. 3B). Using sequence-specific oligonucleotides LS40 was shown to

Table 4. Amino acid sequence of residues 4 5 1 - 4 6 6 for M. leprae (27), E. coli (28) and human (14) hsp 60 homologues 451 M. leprae E. coli Human

466

I A F N S G M E P G V V A E K V • V L C * E • • S • * • N T * G S L I V * • I * K * A * V

"Denotes a conserved residue. Numbering of residues based on the M. leprae sequence.

Downloaded from http://intimm.oxfordjournals.org/ at Georgetown University on May 22, 2015

Mapping of epitope specificity using synthetic peptides

967

968

Rheumatoid arthritis synovial T cell clones 3H-Thymidine uptake 100000E

10000

r«

1000

100

rS

J

m l n,« l~Ll 8

9

10

J 11

12

PPDhsp60CM

CZZI 0.2ug/ml

I 2 ug/ml

I 20 ug/ml

Fig. 1. Proliferative responses of clone CF13 to synthetic peptides of differing length spanning residues 451 - 466 of mycobacterial hsp65(1 -11) or the corresponding sequence from the human homologue P60 (peptide 12). Also tested were the human P60 fusion protein ('hsp 60', used at 5 /ig/ml), M. tuberculosis PPD (2 and 20 jtg/ml) and complete medium (CM) as a control. The sequences of peptides 1-12 were as follows: peptide 1, IAFNSGLEPGVVAEK; peptide 2, IAFNSGLEPGVVA; peptide 3, IAFNSGLEPGW; peptide 4, IAFNSGLEPGV; peptide 5, IAFNSGLEPG; peptide 7, AFNSGLEPGVVAEK; peptide 8, FNSGLEPGWAEK; peptide 9, NSGLEPGVVAEK; peptide 10, SGLEPGVVAEK; peptide 11, GLEPGVVAEK; peptide 12, IAKNAGVEGSLIVEKI. Note that residue 457 in peptides 1-11 is leucine corresponding to the M. tuberculosis sequence; the equivalent residue in peptide 91 (Table 2) is methionine corresponding to the M.leprae sequence.

express the DPB*0402 allele, while patient CF expresses DPB*0401 (16). The failure to detect DR4-restricted recognition of hsp 65 was shown not to be due to non-functional DR4 antigens, since an autologous Sa/mone//a-specific clone responded to antigen presented by LCL BSM (DR4,Dw4 + ) (Fig. 2C). These results also indicated that the patient was at least heterozygous for the Dw4 subtype of HLA-DR4, and subsequent typing using allele-specific oligonucleotides suggests that the patient is in fact homozygous for DR4,Dw4 (D. Pilling and M. Salmon, unpublished results). To exclude the unusual HLA restriction of the clones examined being due to an atypical founder clone, a further cloning was undertaken from the original SF T cells. Table 5 shows the effect of mAbs on the proliferative response of clone CFB39 derived from the second cloning. Inhibition by the anti-DP but not the anti-DR reagent is evident. This clone also showed the same pattern of response to M.hovis hsp 65 deletion mutants, recognized peptide 92 and failed to recognize the human hsp 60 fusion protein. Identical results were obtained for an additional hsp-65-specific clone obtained in this second cloning. Thus from two separate cloning experiments, 5/5 clones tested showed clear evidence of restriction by H LA-DP. Finally, since it was possible that the cloning procedure in some way selected for HLA-DP-restricted clones, the effect of the mAbs on the polyclonal SF MC responses to mycobacterial hsp 65 was tested. As a control the effect of the same antibodies on the response to Salmonella already noted in Table 1 was tested; results are shown in Fig. 3(C). The Salmonella response was effectively blocked by anti-DR and not by anti-DP; conversely,

anti-DR failed to block the response to hsp 65, whereas significant blocking was obtained with the anti-DP antibody. Together these results strongly suggest that in this patient the response to mycobacterial hsp 65 in the inflamed joint was dominated by HLA-DP-restricted T cells. Discussion This study was designed to characterize in detail mycobacterial hsp-65-specific T cell clones derived from RA SF. A previous study (10) of DR3-restricted synovial T cell clones from a patient with self-limiting reactive arthritis clearly showed that the epitope recognized resided in a non-conserved part of the hsp 65 sequence so that in that case there was no opportunity for the mycobacterial hsp-65-specific T cells to cross-recognize the homologous human hsp, P60. It was of interest to determine whether the same finding would apply to clones from a patient with persistent, potentially autoimmune arthritis such as rheumatoid disease, particularly since T cell clones which can respond to DR-restricted epitopes within the self hsp have been described (9,17). The results shown here demonstrate that clones from the RA patient recognized a highly conserved portion of the molecule. Clone CF13 recognized peptide 91 (amino acids 451 -466); interestingly, a synthetic peptide containing amino acids 451 - 4 6 5 has previously been reported to stimulate a mycobacteria-specific polyclonal T cell line from a DR1 + individual, but neither the MHC restriction of this response nor the HLA-DP type of the donor were stated (18). However, despite

Downloaded from http://intimm.oxfordjournals.org/ at Georgetown University on May 22, 2015

Peptide no.

Rheumatoid arthritis synovia! T cell clones 969

A. HLA Restriction CF13

B. HLA Restriction CF519 3H-Thymidine (dpm/1000)

3H-Thymidine (dpm/1000)

8Y

20-f

6-

15-

CF

AL10

I T*EB

SSTO

^M

BSM

LS4O

AL10

IT+EB

T*EB/PPD

SSTO

BSM

LS40

IT*EB/PPD

C. HLA Restriction CFB32 3H-Thymidine (dpm/1000)

5Y

4-

1-

AL10

!T*EB

SSTO

BSM

LS40

I T+EB/Salm

Fig. 2. Proliferative responses of hsp-65-specific (A and B), and Salmonella-specific (C) T cell clones from patient CF to antigen (10 /ig/ml hsp 65 or SDS-solubilized Salmonella) presented by autologous (CF) or allogeneic DR4-homozygous LCL expressing Dw10 (AL10), Dw13 (SSTO), Dw4 (BSM), and Dw14 (LS40). Antigens were used as described in Table 1 except that SDS-solubilized antigens (10 /ig/ml) were used instead of whole organisms.

Downloaded from http://intimm.oxfordjournals.org/ at Georgetown University on May 22, 2015

2

970

Rheumatoid arthritis synovial T cell clones

B.

A. 70

%lnhibition of proliferation

%lnhibition of proliferation

120

60 A

1001

50 H 80 A 40 H 60 H

30 A

20

10 H

o-H DP

DQ

DR

DP

DQ

CF519

DP

DR

CF13

DQ

DR

DP

DR

CM1.8

CM1.9 I

F~~l Dilution 1:100

DQ

I Dilution 1:160

%lnhibition of proliferation 80

60 H

40 1

20 1

gar -so J — ; DQ

;

;

DR

DP

1

;

i

DQ

DR

Salmonella • • 1:160

r DP

65kD CHI 1:80

Fig. 3. Effects of monoclonal antibodies on the proliferative responses of (A) hsp-65-specific clones to hsp 65, (B) control DR4-restricted Yersiniaspecific clones to Yersinia, (C) SFMC to Salmonella and hsp 65. Antigens were used as described in the legend to Figure 2. Comparisons for panel C were as follows. Responses to Salmonella: proliferation in the presence of anti-DR was significantly less than in the presence of anti-DP (P < 0.001) or anti-DQ (P < 0.001), when the antibodies were used at 1:80 dilution. Responses to hsp 65: proliferation in the presence of anti-DP was significantly less than in the presence of anti-DR (P < 0.001) or anti-DQ (P < 0.05) when the antibodies were used at 1:80 dilution.

Downloaded from http://intimm.oxfordjournals.org/ at Georgetown University on May 22, 2015

40

201

Rheumatoid arthritis synovial T cell clones 971 Table 5. Inhibition of proliferative responses of clone CFB39 by mAbs specific for class II MHC antigens Antibody

[3H]Thymidine uptake (c.p.m. ± SEM)

None Anti-DR Anti-DR Anti-DQ Anti-DQ Anti-DP Anti-DP

24959 (961) 23777(1890) 20840(120) 28253 (2268) 33084 (857) 11047(703) 5144 (39)

1:160 1:80 1:160 1:80 1:160 1:80

% inhibition

5 17 -13 -33 56 79

Clone CFB39 was assayed using autologous LCL cells and PPD (10/ig/ml). In the same experiment the anti-DP reagent did not inhibit the response of the clone to PHA and IL-2. Irradiated LCL cultured alone gave 1867 ± 106 c.p.m. while T cells co-cultured with LCL in the absence of antigen gave 1699 ± 8 c.p.m.

An unexpected feature of the synovial hsp-65-specific clones was their restriction by HLA-DP. All clones recognized antigen presented by APC expressing either DPB*0401 or DPB*0402. These alleles differ at three amino acid positions (16), but since they are not distinguished by primed lymphocyte typing it is possible that the amino acid differences do not influence antigen presentation to T cells. Although DP-restricted T cell clones have been described previously (19,20), they have been a minority population accompanying DR-restricted responses to the same antigen, and to our knowledge a DP-restricted antigenic epitope has not been defined previously in detail. The fact that the response to hsp 65 by unselected SF T cells was also inhibited to a significantly greater degree by the anti-DP mAb than by the anti-DR antibody argues against the possibility that the cloning procedure somehow selected for DP-restricted clones. This is the first example in which the response to a common recall

The consequences of recognition of an antigen in the context of DP rather than DR are not known; no particular function has been ascribed to DP-restricted T cells and there is no information on their control. The tissue expression of DP is more circumscribed than that of DR but is prominent in the inflammatory synovium, probably due to strong expression on dendritic cells (24), so that presentation of the antigen to DP-restricted clones would be optimal at this site. This is suggested by the enhanced response to hsp 65 in SF (Table 1) and by our previous observations on the efficacy of antigen presentation by cells within SF (25). There is a report correlating diminished DP expression in the synovium with improvement in arthritis (26). Although well characterized in the mouse, clear examples of the action of MHC antigens acting as immune response genes are rare in man, and their undoubted association with disease susceptibility has not been linked previously to the control of T cell recognition of particular antigens. The demonstration in this study of an apparent influence by the RA-associated MHC antigen DR4 on the immune response to an 'arthritogenic' antigen may represent an example of such a linkage.

Acknowledgements We thank EuroCetus for providing recombinant IL-2, Dr A. R. Poole, McGill University, Montreal, for providing purified human proteoglycan, Dr D. Young for the gift of E. coli GroEL, and Lynne Bailey for excellent technical assistance. This work was supported by grants from the Wellcome Trust and the Arthritis and Rheumatism Council. J.S.H.G. holds a Wellcome Senior Research Fellowship.

Abbreviations APC hsp IL-2 LCL MC PB PHA PPD RA SF

antigen-presenting cells heat shock protein interleukin 2 lymphoblastoid cell line mononuclear cells peripheral blood phytohaemagglutinin purified protein derivative of tuberculin rheumatoid arthritis synovial fluid

Downloaded from http://intimm.oxfordjournals.org/ at Georgetown University on May 22, 2015

the fact that the minimal 10 amino acid epitope defined for clone CF13 (residues 456-465) contains (when compared with the homologous region in human P60) four identical amino acids and four conservative substitutions, the clone recognized neither a peptide corresponding to the human sequence nor recombinant human P60. The epitope recognized by other isolated clones was located in the same part of the molecule as that defined for CF13, but unlike CF13 they failed to respond to residues 451 -466 but only the overlapping peptide (residues 456-471). Since peptide 92 contains the minimal epitope defined for CF13 (and CF13 responds to peptide 92, albeit weakly) it seems probable that all clones recognize the same 'core' epitope, although the responses by individual clones are also influenced by adjacent residues in the synthetic peptides. Despite these minor differences in peptide recognition, however, there was no evidence of recognition of the self hsp by any of the synovial clones. In this respect they resemble the clones isolated from the patient with reactive arthritis (10), and also the rat arthritogenic hsp-65-specific clone (4). The latter recognizes amino acids 180-188 which are not in a conserved region of the molecule. However, the arthritogenic rat T cell clone is apparently autoreactive in that it responds to cartilage proteoglycan. In contrast none of the clones from the RA patient responded to purified proteoglycan (J. S. H. Gaston, unpublished results). Thus, in this patient there was no evidence of autoreactivity associated with the recognition of the mycobacterial hsp 65.

antigen was completely dominated by a DP-restricted population. In this respect it is interesting that antigenic epitopes within the hsp 65 recognized by T cells [including those from the joint (10)] have been defined for most DR alleles, but not for DR4 (21). The lack of such epitopes could occur, either if hsp 65 could not be processed to produce a T cell epitope-containing peptide capable of binding to DR4 (22), or alternatively, if the T cell repertoire formed under the influence of DR4 was incapable of recognizing hsp 65 peptides bound to DR4 (23). In either case the failure of a DR4-restricted recognition would account for the emergence of a dominant DP-restricted response. It will be important to determine whether a DP-restricted response to hsp 66 is a common feature in DR4+ RA patients; preliminary experiments in a small group of RA patients have not shown significant responses to peptide 92, but only one of these patients was shown to have a DP allele capable of presenting this peptide to CF clones (J. S. H. Gaston, unpublished observations).

972

Rheumatoid arthritis synovial T cell clones

References 15 16

17

18

19

20

21

22 23 24 25

26

27 28

29

Downloaded from http://intimm.oxfordjournals.org/ at Georgetown University on May 22, 2015

1 Stastny, P. 1978. Association of the B-cell alloantigen DRw4 with rheumatoid arthritis. N. Engl. J. Med. 298:869. 2 Gregersen, P. K., Silver, J., and Winchester, R. J. 1987. The shared epitope hypothesis: an approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis. Arthritis Rheum. 30:1205. 3 Schwartz, R. H. 1985. T lymphocyte recognition of antigen in association with gene products of the major histocompatibility complex. Annu. Rev. Immunol. 3:239. 4 van Eden, W., Thole, J. E. R., van der Zee, R., Noordzij, A., van Embden, J. D. A., Hensen, E. J., and Cohen, I. R. 1988. Cloning of the mycobacterial epitope recognized by T lymphocytes in adjuvant arthritis. Nature 331:171. 5 van der Broek, M. F., van Bruggen, M. C. J., Hogervorst, E. J., van Eden, W., van der Zee, R, and van den Berg, W. B. 1989. Protection against streptococcal cell-wall induced arthritis by pre-treatment with the 65kD mycobacterial heat shock protein. J. Exp. Med. 170:449. 6 Thompson, S. J., Rook, G. A. W., Brealey, R. J., van der Zee, R., and Elson, C. J. 1990. Autoimmune reactions to heat shock proteins in pristane-induced arthritis. Eur. J. Immunol. 20:2479. 7 Gaston, J. S. H., Life, P. F., Bailey, L. C , and Bacon, P. A. 1989. In vitro response to a 65 kilodalton mycobacterial protein by synovial T cells from inflammatory arthritis patients. J. Immunol. 143:2494. 8 Holoshitz, J., Koning, F., Coligan, J. E., de Bruyn, J., and Strober, S. 1989. Isolation of CD4-, CD8- mycobacterial-reactive T lymphocyte clones from rheumatoid arthritis synovial fluid. Nature 339:226. 9 Lamb J. R., Bal, V., Mendez-Samperio, P., Mehlert, A., Ivanyi, J., So, A., Rothbard, J., Jindal, S., Young, R. A., and Young, D. B. 1989. Stress proteins may provide a link between the immune response to infection and autoimmunity. Int. Immunol. 1:191. 10 Gaston, J. S. H., Life, P. F., Jenner, P. J., Colston, M. J., and Bacon, P. A. 1990. Recognition of a mycobacteria-specific epitope in the 65kD heat-shock protein by synovial fluid-derived T cell clones. J. Exp. Med. 171:831. 11 Arnett, F. C , Edworthy, S. M., Bloch, D. A., McShane, D. J., Fries, J. F., Cooper, N. S., Healey, L. A., Kaplan, S. R., Liang, M. H., Luthra, H. S., Medsger, T. A., Mitchell, D. M., Neustadt, D. H., Pinals, R. S., Schaller, J. G., Sharp, J. T., Wilder, R. L, and Hunder, G. G. 1988. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 31:315. 12 Gregersen, P. K., Shen, M., Song, Q., Merryman, P., Degar, S., Seki, T., Maccari, J., Goldberg, D., Murphy, H., Schwenzer, J., Wang, C. Y., Winchester, R. J., Nepom, G. T., and Silver, J. 1986. Molecular diversity of HLA-DR4 haplotypes. Proc. NatlAcad. Sci. USA 83:2642. 13 Lamb, F. I., Kingston, A. E., Estrada-G, I., and Colston, M. J. 1988. Heterogeneous expression of the 65-kilodalton antigen of mycobacterium leprae and murine T-cell responses to the gene product. Infect. Immunol. 56:1237. 14 Jindal, S., Dudani, A. K., Singh, B., Harley, C. B., and Gupta, R. S. 1989. Primary structure of a human mitochondrial protein homologous

to the bacterial and plant chaperonins and to the 65-kilodalton mycobacterial antigen. Mol. Cell Biol. 9:2279. Young, R. A. 1990. Stress proteins and immunology. Annu. Rev. Immunol. 8:401. Begovich, A. B., Bugawan, T. L., Nepom, B. S., Klitz, W., Nepom, G. T., and Erhlich, H. A. 1989. A specific HLA-DPB allele is associated with pauciarticular juvenile rheumatoid arthritis but not adult rheumatoid arthritis. Proc. Natl Acad. Sci. USA 86:9489. Munk, M. E., Schoel, B., Modrow, S., Karr, R. W., Young, R. A., and Kaufmann, S. H. E. 1989. T lymphocytes from healthy individuals with specificity to self epitopes shared by the mycobacterial and human 65-kilodalton heat shock protein. J. Immunol. 143:2844. Oftung, F., Mustafa, A. S., Shinninck, T. M., Houghten, R. A., Kvalheim, G., Degre, M., Lundin, K. E. A., and Godal, T. J. 1988. Epitopes of the mycobacterium tuberculosis 65-kilodalton protein antigen as recognized by human T cells. J. Immunol. 141:2749. Eckels, D. D., Lake, P., Lamb, J. R., Johnson, A. H., Shaw, S., Woody, J. N., and Hartzman, R. J. 1983. SB-restricted presentation of influenza and herpes simplex virus antigen to human T-lymphocyte clones. Nature 301:716. de Koster H. S., Anderson, D. C, and Termijtelen, A. J. 1989. T cells sensitized to synthetic HLA-DR3 peptide give evidence of a continuous presentation of denatured HLA-DR3 molecules by HLA-DP. J. Exp. Med. 169:1191. van Schooten, W. C. A., Elferink, D. G., van Embden, J. D. A., Anderson, D. C , and de Vries, R. R. P. 1989. DR3-restricted T cells from different HLA-DR3-positive individuals recognize the same peptide (amino acids 2 -12) of the mycobacterial Hsp 65a heat shock protein. Eur. J. Immunol. 19:2075. Sette, A., Buus, S., Colon, S., Smith, J. A., Miles, C , and Grey, H. M. 1987. Structural characteristics of an antigen required for its interaction with la and recognition by T cells. Nature 328:395. Marrack, P. and Kappler, J. W. 1988. The T cell repertoire for antigen and MHC. Immunol. Today 9:308. Knight, S. C, Fryer, P., Griffiths, S., Harding, B., Dixey, J., and Mansell, B. 1989. Class II antigens on dendritic cells from the synovial fluids of patients with inflammatory arthritis. Clin. Exp. Immunol. 78:19. Life, P. F., Viner, N. J., Gaston, J. S. H., and Bacon, P. A. 1990. Synovial fluid antigen-presenting cells unmask peripheral blood T cell responses to bacterial antigens in inflammatory arthritis. Clin. Exp. Immunol. 79:189. Walters, M. T., Smith, J. L, Moore, K., Evans, P. R., and Cawley, M. I. D. 1987. An investigation of the action of disease modifying anti-rheumatic drugs on the rheumatoid synovial membrane: reduction in T lymphocyte subpopulations and HLA-DP and DQ antigen expression after gold or penicillamine therapy. Ann. Rheum. Dis., 46:7. Mehra, V., Sweetser, D., and Young, R. A. 1986. Efficient mapping of protein antigenic determinants. Proc. NatlAcad. Sci. USA 83:7013. Hemmingsen, S. M., Woolford, C , van der Vies, S. M., Tilly, K., Dennis, D. T., Georgopoulos, C. P., Hendrix, R. W., and Ellis, R. J. 1988. Homologous plant and bacterial proteins chaperone oligomeric protein assembly. Nature 333:30. Viner, N. J., Bailey, L. C, Life, P. F., Bacon, P. A., and Gaston, J. S. H. 1991. Isolation of Yersina-specific T cell clones from synovial membrane and synovial fluid of a patient with reactive arthritis. Arthritis Rheum, in press.