Immunol. Cell Biol. (1990)68, 27-31
Antibodies to type II collagen in SLE: A role in the pathogenesis of deforming arthritis? Edward K. K. Choi,' Paul A. Gatenby,' John F. Bateman2 and William G. ^Clinical Immunology Centre, Royal Prince Alfred Hospital and University of Sydney, Sydney, NSIV 2006, Australia ^Department of Paediatrics, University of Melbourne, Royal Children's Hospital, Parkville, Vic, 3053, Australia (Submitted
25 August 1989, Accepted for publication 8 September
Summary The role of autoimmunity to type II collagen in the arthritis of systemic lupus erythematosus (SLE) has been assessed by ELISA and by Western blotting cyanogen bromide cleavage fragments separated on SDS-PAGE. The results show that antibodies to both native and heat-denatured collagen are quite common in SLE when measured by ELISA. Of particular interest is the demonstration of an association between antibodies to the CB 11 peptide and deforming arthritis in SLE. This is the arthritogenic peptide in murine models of collagen Il-induced autoimmune arthritis and the results presented here suggest a potential pathogenetic role in the deforming arthritis of SLE for this specific subset of antibodies to type II collagen.
INTRODUCTION Autoimmunity to type II collagen (Col II) may have a role in the pathogenesis of systemic lupus erythematosus (SLE) as antibodies to native Col II have been reported by us (1) and others (2-4). A more clear cut role of autoimmunity to native Col II has been reported in mouse models of type II collagen-induced arthritis. These models have the typical attributes of an autoimmune disease in which the B and T cell reactivity is under immunogenetic control (5-9). A further development in the mouse models has been the demonstration that CB 11 cyanogen bromide (CNBr) cleavage fragment of Col II contains the arthritogenic epitopes that induce the disease and react with the pathogenetic antibodies that are generated (10). We have previously detected Col II antibodies in SLE and proposed that they may have
Correspondence: Dr Paul A. Gatenby, Department of Clinical Immunology, Royal Prince Alfred Hospital, Missenden Rd, Camperdown, NSW 2050, Australia. Abbreviations used in this paper: CB, CNBr fragment; CNBr, cyanogen bromide; Col II, type II collagen; ELISA, enzyme linked immunosorbent assay; NCP, nitrocellulose paper; RA, rheumatoid arthritis; SDS-PAGE, polyacrylamide gel electrophoresis with sodium dodecyl sulphate; SLE, systemic lupus erythematosus.
resulted from a non-specific polyclonal response to the antigen. However, in view of the reports that specific CNBr peptides of Col II contain arthritogenic epitopes in mouse models of Col Il-induced arthritis, we examined sera from humans with SLE for antibodies directed against Col II CNBr peptides. We determined whether such antibodies were associated with deforming arthritis in SLE. MATERIALS AND METHODS Subjects The patient group comprised 109 individuals with SLE as defined by the revised American Rheumatism Association criteria (11). The patients attended the Clinical Immunology Department at Royal Prince Alfred Hospital and were assessed clinically by one of us (PG). This data was entered into a computerized data-base and was collected without knowledge of the anti-Col II status. The definitions of arthralgia, joint swelling, tendonitis and deforming arthritis were taken from the glossary for the American Rheumatism Association Medical Information Service (ARAMIS) data-base (D. McShane and J. H. Decker, unpubl. data) and were scored as present or absent. All individuals gave informed consent. Type II collagen antigen The Col II antigen was prepared as described previously from rib cartilage obtained at post-mortem of infants who died from sudden infant death syndrome (1).
E. K. K. CHOI ETAL.
Anti-collagen antibody assay
Antibodies to both native and heat-denatured Col II were measured by ELISA as described previously (1). Cyanogen bromide cleavage of type II collagen Cleavage of Col II was carried out with minor modifications as described by Scott and Veis (12). Cyanogen bromide (Sigma, St Louis, MO) was dissolved in formic acid (70% v/v) at a concentration of 50 mg/mL. Col II was added to this solution at a concentration of 1 05 mg/mL and digested for 4 h. The solution was then blown dry under a stream of air and the dried product was dissolved in 0-5 mol/L acetic acid and lyophilized. This dissolving and lyophilizing procedure was repeated to ensure the removal of volatile products. The final lyophilized materials were stored at 4°C. A sample of the cleavage products was analysed by electrophoresis in polyacrylamide gel containing sodium dodecyl sulphate (SDS-PAGE) to check on the production of previously identified CNBr fragments of collagen (13). Western blotting The CNBr cleavage fragments were separated by SDS-PAGE in 12 5% polyacrylamide gel, electrophoretically transferred on to nitrocellulose paper (Biorad, Richmond, CA) and immunoblotted with patients' sera (14). The CNBR-treated Col II was dissolved in sample buffer and 80 /JL aliquots were loaded on to the gel consisting of a 3-5% stacking gel on top of a 12-5% 2.0
polyacrylamide separating gel. Electrophoresis was carried out at 15°C with 20 mA current, increased to 40 mA afler the sample moved into the stacking gel at 500 volts for a total of 2 h. Electrophoretic transfer was carried out at 4°C, essentially as described by Towbin et al, (14) with current applied at 1 -2 A and 65 volts for 1 h and 55 volts for a subsequent hour. Statistical analysis The major comparison was between the patients with and without anti-CB 11 with regard to the development of arthritis and deforming arthritis. Fisher's Exact test was used. RESULTS Anti-Col II antibody detection in ELISA The incidence of antibodies and the relative optical densities of the positive samples to the positive control is shown in Fig. 1. Of the 30% of patients with antibodies to native Col II, only four had a relative OD of greater than 1-0. Only 17% had antibodies to denatured Col II, with just one demonstrating an OD greater than 1 0. Twenty individuals had anti-native Col II alone, 12 had anti-denatured Col II and only six had both types of antibody. These results are much the same as those reported by us for rheumatoid arthritis (RA), although the incidence of anti(b)
o 'c o o a,
I IT) O
-iNormal Systemic Lupus Normal Systemic Lupus Control Erythematosus Control Erythematosus (109) (109) (58) (58) Fig. 1. Serum IgG antibody to (a) native and (b) lieat-denatured Col II. The results are expressed as a ratio of sample OD to control. The mean and s.d. is shown with the positive sera indicated by (•). The numbers of individuals tested are shown in parentheses.
TYPE II COLLAGEN ANTIBODIES IN SLE
5 6 7 8 9 •
8 — 9.7
Fig. 2. Immunoblotting of CNBr peptides of human Col II by sera from individuals with SLE. The two positive controls are sera from patients with RA. Lanes 1-11 are sera from patients with SLE.
native Col II is a little higher in the lupus patients (1). Immunoblotting of CNBr peptides of Col II by SLE sera Immunoblotting of CNBr peptides of Col II by SLE sera is shown in Fig. 2. Some ofthe positive sera, for example lane 1, show as many reactivities with CNBr peptides as are seen in RA. Others, for example lane 5, show reactivity only to the CB 10-5 and CB 11 band. Lanes 6-11 showed reactivity to a slightly slower band, also evident in lane 1, not representing any of the previously well defined bands, and of uncertain significance. Of the 45 patients with positive ELISA results (to either native or denatured Col II) 42 of them had serum available for immunoblotting. Of these 42, six samples somewhat surprisingly showed no bands despite being repeated several times. All 64 who were ELISA negative had sera available for immunoblotting and 40 had antibodies against one or more ofthe CNBr peptides. Most positive samples in both groups had reactivity to CB 10-5, with the reactivity being stronger in the group who were
ELISA positive. The numbers in each group reactive with each ofthe four well defined bands are shown in Table 1. Apart from the overall higher rate of positivity in the ELISA positive patients no major differences between the two groups was evident. The finding of the negative immunoblots in some ofthe patients with positive ELISA may be due to the epitopes being present within small CNBr peptides that cannot be adequately detected in the gel system used here. Likewise epitopes may be exposed that allow reactivity in ELISA negative sera. Alternatively as we have quite a stringent cut-off for the ELISA (1) the Table 1. The number of patients in both the ELISA positive and ELISA negative gr9ups with positive immunoblotting against the major CNBr bands. CNBr band CB 10 5 CB 11 CB8 CB9 7
ELISA positive ( 42)
ELISA negative (/I = 65)
34 7 1 2
34 6 0 5
E. K. K. CHOI ETAL.
joints, potentially affected. The pathogenesis of joint disease in SLE is not particularly wellunderstood and in many instances articular manifestations have an underlying pathology that is largely periarticular. This would account for many cases of arthralgia and even joint swelling (15). In some relatively infrequent cases a proliferative synovitis can be associated with a deforming arthritis (16). Usually this form of arthritis can be distinguished from RA by the lack of significant erosive disease (16,17). That was the case with patients reported here, none of whom had erosions. Organ or tissue specific antibodies can be associated with particular clinical problems; for example, autoimmune haemolytic anaemia and immune thombocytopenia. It is possible that antibodies to Col II act in an analogous way and play a direct role in the pathogenesis of the deforming arthritis. Other workers have described antibodies to Col II in SLE, but no attempt was made to correlate these with the presence of joint disease (3). We believe this is the first study which has used immunoblotting to examine the epitope specificity of Col II antibodies in SLE and has allowed a more precise definition of a potentially pathogenic antibody. There are, however, individuals with anti-CB 11 without deforming arthritis (Table 3). It is possible with time that more of these patients will develop deforming arthritis. Likewise a small number of individuals have deforming arthritis without anti-CB 11 antibodies. Such cases may once have been positive, or alternatively there may be other mechanisms that lead to deforming arthritis. Both of these two problems will only be resolved with, respectively, long term follow-up of patients and repeated assays for anti-CB 11 over the whole course of disease in patients with deforming arthritis. In summary, these results have provided support for the concept that antibodies to Col II may have a role in the pathogenesis of deforming arthritis in SLE, although longitudinal studies of lupus patients will be needed to provide further evidence for these observations.
The data presented here provide preliminary support for the concept that antibodies to the arthritogenic peptide in Col Il-induced arthritis of mice, CB 11, play a role in the pathogenesis of deforming arthritis in SLE. The latter is a multisystem disorder with many organs, including the
Acknowledgements We thank the physicians who referred their patients and the Clive and Vera Ramaciotti Foundation for financial support. Edward Choi was supported by a Postgraduate Research Award from the Sydney LIniversity.
Table 2. The relationship between non-deforming arthritis and anti-CB II. CBll CBll Significance positive negative Past history of non-deforming arthritis Present 6 37 Absent 5 50 NS, 7'=019 Current evidence of non-deforming arthritis Present 4 23 Absent 7 64 NS,/'=0-22 Table 3. The relationship between deforming arthritis and anti-CB II. CBll positive
Past history of deforming arthritis Present 3 6 Absent 8 81 Current evidence of deforming arthritis Present 3 2 Absent 8 85
positive immunoblots may represent reactivity that we arbitrarily designated negative by ELISA. Detailed perusal of the OD obtained in the immunoblot positive, ELISA negative samples does not support this theory (unpubl. data). Relationship between the presence of antibodies to CB 11 and arthritis The patients were considered to have nondeforming arthritis if arthralgia and joint-swelling were scored as present. The results in Table 2 show that there was no relationship between the presence of non-deforming arthritis, either past or present, and antibodies to the CB 11 band. This is in contrast to the situation in deforming arthritis (Table 3) where there is a significant difference suggesting that deforming arthritis in SLE is associated with antibodies to the CB 11 band.
TYPE II COLLAGEN ANTIBODIES IN SLE
REFERENCES 1. Choi, E. K. K., Gatenby, P. A., McGill, N. W. et al. 1988. Autoantibodies to type ii collagen: Occurrence in rheumatoid arthritis, other arthritides, autoimmune connective tissue disease, and chronic inflammatory syndromes. Annal. Rheum. Dis. 47: 313-322. 2. Clague, R. B., Shaw, M. J. and Holt, P. J. L. 1980. Incidence of serum antibodies to native type I and type II collagens in patients with inflammatory arthritis. Ann. Rheum. Dis. 39: 201-206. 3. Gioud, M., Meghlaoui, A., Costa, O. and Monier, J. C. 1982. Antibodies to native type I and II collagens detected by an enzyme linked immunosorbent assay (ELISA) in rheumatoid arthritis and systemic lupus erythematosus. Collagen. Rel. Res. 2: 557-564. 4. Stuart, J. M., Huffstutter, E. H., Townes, A. S. and Kang, A. H. 1983. Incidence and specificity of antibodies to type I, II, III, IV and V collagen in rheumatoid arthritis and other rheumatic diseases as measured by '^'I-radioimmunoassay. Arthr. Rheum. 26: 832-840. 5. Clague, R. B., Morgan, K., Shaw, M. J. and Holt, P. J. L. 1980. Native type II collagen-induced arthritis in the rat: II. Relationship between the humoral immune response to native type II collagen and arthritis. / Rheumatol. 1: 775-782. 6. Stuart, J. M., Townes, A. S. and Kang, A. H. 1982. Nature and specificity of the immune response to collagen in type II collagen induced arthritis in mice. /. Clin. Invest. 69, 673-683. 7. Klareskog, L., Holmdahl, R., Larsson, E. and Wigzell, H. 1983. Role of T lymphocytes in collagen II induced arthritis. Clin. Exp. Immunol. 51: 117-125. 8. Griffiths, M. M. and DeWitt, C. W. 1984. Genetic control of collagen-induced arthritis in rats: the immune response to type II collagen among susceptible and resistant strains and evidence for multiple gene control. J. Immunol 132: 2830-2836.
9. Hom, J. T., Stuart, J. M., Tovey, X. and Chiller, J. M. 1986. Murine T cells reactive to type II collagen: II. Functional characterisation. J. Immunol. 136: 776-782. 10. Terato, K., Hasty, K. A., Cremer, M. A., Stuart, J. M., Townes, A. S. and Kang, A. H. 1985. Collageninduced arthritis in mice: Localization of an arthritogenic determinant to a fragment of the type II collagen molecule. J. Exp. Med. 162: 637-646. 11. Tan, E. M., Cohen, A. S., Fries, J. F. et al 1982. The 1982 revised criteria for the classification of systemikc lupus erythematosus. Arthr. Rheum. 25: 1271-1277. 12. Scott, P. G. and Veis, A. 1976. The cyanogen bromide peptides of bovine soluble and insoluble collagens: I. Characterization of peptides from soluble type I collagen by sodium dodecylsulphate polyacrylamide gel electrophoresis. Connect. Tissue. Res. 4: 107-116. 13. Bomstein, P. and Sage, H. 1980. Structurally distinct collagen types. Ann. Rev. Biochem. 49: 957-1003. 14. Towbin, H., Staehelin, T. and Gordon, J. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. NatlAcad. Sci. USA 76: 4350-4354. 15. Rothfield, N. 1985. Clinical features of systemic lupus erythematosus. In Textbook of Rheumatology, W. N. Kelley, E. D. Harris, S. Ruddy and C. B. Sledge (eds). W. B. Saunders, Philadelphia, PA, USA, pp. 1070-1097. 16. Russell, A. S., Percy, J. S., Rigal, W. M. and Wilson, G. L. 1974. Deforming arthropathy in systemic lupus erythematosus. Ann. Rheum. Dis. 33: 204. 17. Labowitz, R. and Schumacher, H. R. 1971. Articular manifestations of systemic lupus erythematosus. Ann. Intern. Med. 74: 911-921.