HLA-DR/DQ Antigens and Reactivit Cell Alloantigen D8/17 in Indian Patients With Rheumatic Heart Disease

Transcription

1 335 HLA-DR/DQ Antigens and Reactivit Cell Alloantigen D8/17 in Indian Patients With Rheumatic Heart Disease to B Veena Taneja, MSc, Narinder K. Mehra, PhD, K. Srinath Reddy, MD, DM, Jagat Narula, MD, DM, Rajendra Tandon, MD, DM, Mahesh C. Vaidya, MS, PhD, and Madan L. Bhatia, MD, DM D8/17, a B cell alloantigen, and its occurrence with DR/DQ antigens was examined in 54 patients with rheumatic heart disease and matched North Indian controls. Thirty-four patients (62.9%) carried D8/17 as compared with a frequency of 12.5% in controls (X2 by Yates', 18.75; p<0.0001). A marginally increased frequency of human leukocyte antigen (HLA)-DR3 (44.2% vs. 28.8%) and a significantly reduced frequency of HLA-DR2 (21.1% vs. 46.6%; p<o.005) was observed in patients as compared with controls. However, the most significant positive association was observed with HLA-DQw2 (63.4% in patients and 31.5% in controls; V' by Yates', 9.85; p<0.005). The DQw2 association was significant only in the D8/17-positive patients. These observations suggest that the disease susceptibility gene in rheumatic heart disease may be nearer to the HLA-DQ locus than to the HLA-DR. (Circulation 1989;80: ) T here is increasing evidence to indicate that patients with rheumatic fever may be genetically programmed to respond abnormally to streptococcal infections. Many investigators have postulated an important role of inherited susceptibility in the causation of rheumatic heart disease. This observation is based on epidemiologic surveys indicating familial occurrence of the disease, presence or absence of certain blood groups and ABH secretions,3 and association with certain HLA antigens4 of the major histocompatibility complex of humans. Although a simple autosomal recessive pattern of inheritance has been suggested,5 attempts to identify genetic markers of susceptibility in rheumatic heart disease have been only partially successful. We recently reported a significant increase of HLA-DR3 and reduced frequency of HLA-DR2 in a large series of patients with rheumatic heart disease of Asian-Indian origin.6 Recent studies have rcvealed the existence of a genetically intricate system of serologically defined From the Cellular Immunology Laboratory, Department of Anatomy, and Department of Cardiology, All India Institute of Medical Sciences, New Delhi, India. Supported by the Indian Council of Medical Research (ICMR) under their project "Advanced Centre for Rheumatic Fever and Rheumatic Heart Disease." Address for correspondence: N.K. Mehra, PhD, Cellular Immunology Laboratory, Department of Anatomy, All India Institute of Medical Sciences, New Delhi, India, Received June 30, 1988; revision accepted April 4, alloantigens that are selectively expressed on B lymphocytes. With a serum containing B cell alloantibody called 883+, a positive reaction was obtained in approximately 72% of all patients with rheumatic fever from New York and Bogota, Columbia.7 Further, use of two B cell-specific monoclonal antibodies (83s19.23 and 256s10) by the same investigators identified more than 92% of all patients with rheumatic fever. The former antibody (comparable to B cell alloantigen 883+) was positive in at least 59.1% of Indian patients with rheumatic heart disease compared with its reactivity in 16.6% of control subjects.8 In an effort to further refine the monoclonal genetic markers, these investigators made yet another hybridoma clone, D8/17, the properties of which were different from those of 83s19.23 and 256s10. A preliminary study revealed a 100% positivity of D8/17 in rheumatic patients from New York as compared with 15% in control subjects.9 The present study was undertaken to investigate the prevalence of B cell alloantigen D8/17 and HLA-DR/DQ antigens in patients with rheumatic heart disease of Asian origin and to calculate the frequency of co-occurrence of HLA-DR3 and D8/17 in the patient group as compared with controls. Our results are consistent with the hypothesis that susceptibility to rheumatic heart disease is genetically determined.

2 336 Circulation Vol 80, No 2, August 1989 Methods Patients The present study included 54 patients with established rheumatic heart disease of both sexes admitted to cardiology wards of the All-India Institute of Medical Sciences Hospital, New Delhi. Ethnically, the patients represented North Indian Hindus belonging to the states of Punjab, Haryana, Uttar Pradesh, Bihar, and Delhi. The rheumatic etiology of the valvular lesions was diagnosed when one or more of the following criteria were fulfilled: history of rheumatic fever; active rheumatic fever at the time of evaluation, diagnosed by revised Jones' criteria10; or presence of mitral stenosis, which is almost invariably of rheumatic etiology. Controls Immunostaining for D8/17 alloantigen was conducted in 32 normal, healthy, unrelated subjects matched for age and sex and representing the same ethnic group as the patients. The HLA data of the patients were compared with that of 163 healthy subjects who resided in the same geographic area. HLA-DQ antigens were tested on 57 of these subjects. HLA Typing HLA typing for class 1 (HLA-ABC) and class II (HLA-DR/DQ) antigens was performed by the standard two-stage microlymphocytotoxicity technique."1 These antigens were defined by 180 antisera (120 for HLA class I and 60 for HLA class II) obtained by serum exchange with Dr. John Hansen (Fred Hutchinson Cancer Research Center, Seattle, Washington) and Dr. Jon van Rood (Eurotransplant Foundation, Leiden, The Netherlands). Many of the sera used for the DR/DQ antigen typing were requested from the third Asia-Oceania (Sapporo, Japan, June 26-July 1, 1986) and 10th International Histocompatibility Workshops. Some of the controls used in the study were retested on HLA trays provided by the workshop. Approximately ml heparinized blood (10 IU preservative-free heparin/ml blood) was obtained from each patient and control subject. TABLE 1. Frequency of B Cell Alloantigen D8/17 in North Indians With and Without Rheumatic Heart Disease Patient Category D8/17+ (%) X' by Yates' With rheumatic heart disease (n=54) 34 (62.9) (p<0.0001) Without rheumatic heart disease (n=32) 4 (12.5) Peripheral blood lymphocytes obtained after density gradient centrifugation were further incubated in nylon wool columns (Fenwal Laboratories, USA) in drinking straws for the separation of T and B cells. The former were used for HLA-A,B,C typing and the latter for detecting HLA class II antigens (DR and DQ alleles). The remainder of the B lymphocytes were used for making smears for immunostaining of D8/17. The HLA-DR typing tray included multiple antisera for specificities DR1 through DRw10, DRw52, and DRw53 as well as DQ antigens, DQwl through DQw3. Immunostaining for D8/17 Immunostaining for D8/17 alloantigen was performed according to the avidin-biotin peroxidase complex technique.12 Peripheral blood B lymphocytes enriched by nylon wool columns were washed three times with phosphate-buffered saline (ph 7.4) and adjusted to a concentration of 4 x 166 cells/ml in RPMI 1640 medium. Thin-layer smears were made on clean microglass slides and air dried for 2-3 hours. The fixed slide preparations were stored at -20 C in sealed plastic bags and further processed for immunostaining within 1-6 weeks. The primary antibody D8/17 was obtained through the kind courtesy of Dr. Zabriskie (USA). The rabbit anti-mouse immunoglobulin was used as the secondary antibody and avidin-biotinylated horseradish peroxidase complex (Vector Laboratories, TABLE 2. Percent Phenotype Frequencies of HLA.DR jantigens in Normal Control Subjects and Patients With Rheumatic Heart Disease With or Without D8/17 Antigen Antigen Control Subjects (n = 163) Patients (n=52) D8/17+ (n=33) D8/17- (n= 19) DR DR t 15.78t DR DR DR t DRw DR DRw DR DRw1O *p<0.005, tp<0.05.

3 Taneja et al HLA and B Cell Alloantigen D8/17 in RHD 337 TABLE 3. HLA Phenotypes of Patients With Rheumatic Heart Disease Positive for D8/17 HLA antigens S.No. Patient A B C DR DQw R.S. G.P. S.N. M.R. P.M. D.K. A.K. B.B. A.K. J.K. S.D. D.K. H.K. G.C. S.K. R.K. D.S. P.K. H.A. M.D. S.D. G.R. R.K. B.D. S.K. S.B. Y.D. S.B. V.M. R.K. S.D. S.K. S.T. B.K ,- 1,w33 29,- w , ,- 28,- 10 2,11 19,- 2,11 1,w33 2,29 2,11 13,w62 35,17 35,37 7,35 5,8 7,35 35,- 7,35 4 7,44 44,18 40,- 17,27 8,17 7,8 5,w60 w62,17 13,37 40,- 37,40 5,7 w50,- 5,w50 5,- 5,w62 40,44 44,40 5,35 5,44 17,35 15,- 27,44 35,44 5,22 3,7 3,w6 3,1 3,7 3,w10 3,6 2,4 4,5 7,- 4,5 w6,7 7,- 2,4 2,4 2,7 7,w8 7,w8 7,w10 2,7 NT 1,- 1,- 1,- NT Burlingame, California) constituted the avidin-biotin peroxidase complex reagent. 3,3'-Diaminobenzidine tetrahydrochloride was used as the chromogen. The slides were counterstained with Mayers' hematoxylin and mounted in glycerin jelly. Control slides were prepared by omitting incubation with the primary antibody. Blackish-brown staining was observed over the membrane of the positively stained cells. Cells with positive surface staining were counted by two independent observers using a microscope. Statistical Analysis The significance of difference in the expression of B cell alloantigen and various HLA antigens in patients and control subjects was calculated by the standard x2 test with Yates' correction. p values were corrected by multiplying them by 40 (i.e., the total number of HLA antigens studied). Relative risk for measuring strength of the association with each antigen was calculated by the method of Woolf (see Svejgaard et al13). The etiologic fraction or proportion of disease due to the HLA marker associated with the disease was calculated according to the method of Green.14 Results The percent frequency of B cell alloantigen marker D8/17 in patients with rheumatic heart disease and control subjects is presented in Table 1. Out of a total of 54 patients, 34 (62.9%) reacted with the monoclonal antibody D8/17 as compared with four of the 32 control subjects (12.5%) (X2 by Yates', 18.75;p<O.OOO1; relative risk, 11.9). All but two patients were tested for DR and DQ antigens. No significant deviation was observed in

4 338 Circulation Vol 80, No A August 1989 TABLE 4. HLA Phenotypes of Patients With Rheumatic Heart Disease Negative for D8/17 HLA antigens S.No. Patients A B C DR DQw 1 R.N. 1,- 35,- -- 3,7 2 S.N ,w62 3,9 3 R.R w3 4 4 R.B. 44,17 3,7 5 A.K. 5, M.D. 16 8, ,7 7 P.W. 18,40 3,7 8 M.U. 1,9 5,62 9 K.K. 3,9 7, ,7 10 L.S. 3,9 40,44 11 N.B. 28,- 18,40 2,w6 1,- 12 A.K. 9,- 40,w62 2,w6 1,- 13 B.P. 3 7,- w6,- 1,- 14 P.S. 16 8,4 -- 4,5 15 A.K , L.C. 2,11 5, A.J. 7, ,- 18 P.C. 11,19 35, M.K. 5,- -- 1,4 20 P.S. 16 8,35 NT NT any of the HLA-A and -B antigens in patients as compared with control subjects. In the DR locus, an important finding is the reduced frequency of HLA- DR2 in the patient group as compared with controls (46.62% vs %;p<0.005) (Table 2). This value is significant even after p correction. Similarly, frequency of HLA-DR5 was also decreased in patients as compared with control subjects, although this deviation was not significant after p correction was applied. HLA-DR3, on the other hand, was only marginally increased in the total patients without reaching statistical significance. The decrease in frequency of DR2 was observed in both D8/ 17-positive as well as -negative groups. Full HLA phenotypes of patients positive and negative for D8/17 are given in Tables 3 and 4, respectively. An interesting finding was the increasing occurrence of antigen DQw2 of the HLA-DQ locus in patients with rheumatic heart disease (63.4% vs. 31.5%; x2 by Yates', 9.85;p<0.005). This increase in DQw2 was significant only in the D8/17-positive group (Table 5). The corrected p value for HLA- DQw2 in D8/17+ group was statistically significant. Discussion An abnormal humoral and cellular immune response to various streptococcal antigens suggests its role in the pathogenesis of rheumatic fever and rheumatic heart disease. Although a strong familial association exists, the mode of inheritance of rheumatic heart disease is unclear. Also, the penetration of this disease is relatively low as evidenced by the lower concordance of rheumatic fever in identical twins15 compared with other infections like poliomyelitis and tuberculosis.16 While a streptococcal etiology of the disease has been proven by several independent lines of inquiry (epidemiologic, serologic, and prophylactic),2'8'9 the low conversion rates of streptococcal pharyngitis into rheumatic fever make the study of host factors, especially genetic susceptibility, relevant. Because HLA is the most highly polymorphic genetic system thus far known and because human immune response genes exist within this system,17 numerous investigators have sought an association of HLA antigens with rheumatic fever and rheumatic heart disease. An important aspect of the present study concerns the role of HLA class II (DR and DQ) gene products in governing susceptibility to rheumatic heart disease in humans. Numerous investigators have suggested the involvement of HLA-DR antigens in governing susceptibility to rheumatic heart disease, although the exact allele in all these studies is different This allelic discrepancy in the HLA-DR antigen association in these studies (including ours) can be related to the geographic and racial differences in the genetic background of different populations. In our previous study involving 110 patients, HLA-DR3 appeared significantly increased (p<0.0001) and DR2 significantly reduced (p<0.0001) in patients with rheumatic heart disease of Asian-Indian origin as compared with control subjects.6 In the present study, although DR2 was significantly reduced, HLA-DR3 was only marginally increased (X2 by Yates', 3.2), despite the fact that the controls in the two studies belong to the same ethnic background. This could be explained on the basis of the smaller number of

5 Taneja et al HLA and B Cell Alloantigen D8/17 in RHD 339 TABLE 5. Percent Frequencies of HLA DQ Antigens in Control Subjects and Patients With Rheumatic Heart Disease With or Without D8/17 Antigen Control subjects Patients with rheumatic heart disease Antigen (n=57) (n=52) D8/17+ D8/17- DQwl Frequency X2 by Yates' p NS NS NS RR EF DQw2 Frequency X2 by Yates' p <0.005* <0.005* NS RR EF DQw3 Frequency x2by Yates' p NS NS NS RR EF *P corrected significant. RR, relative risk; EF, etiologic fraction. patients who were available to us for HLA-DR testing. The significant positive association with HLA-DQw2 observed in the present study indicates that the "hypothetic" disease susceptibility gene in rheumatic heart disease may be nearer to the DQ locus than to HLA-DR because HLA-DQw2 is known to be in strong linkage disequilibrium with HLA-DR3. This may be another reason why DR3 did not show a positive association in the present study. This observation assumes significance in view of the recent studies on another autoimmune disease, namely, insulin-dependent diabetes mellitus. Analysis of DNA sequences from diabetics indicates that alleles of HLA-DQB determine both disease susceptibility and resistance and that the structure of the DQ molecule specifies the autoimmune response against the insulin-producing islet cells.2' Recently, Zabriskie9 proposed that a B cell alloantigen D8/17 could probably be a near-perfect marker for rheumatic fever as it was demonstrated in 100% of their patients with rheumatic heart disease of North American origin. However, this antigen occurred in only 62% of patients with rheumatic heart disease of North Indian origin as evidenced from the results of the present study. The ethnic variability in the expression of D8/17 in different populations does not seem to explain the discrepancy because the frequency of D8/17 in the control Indian population is remarkably similar to that observed in the US caucasians. Nevertheless, the study of D8/17 in patients belonging to other ethnic and racial groups would be important in assessing the value of this marker in rheumatic fever and rheumatic heart disease. We also tried to investigate whether cooccurrence of D8/17 and DR3/DQw2 produced an "additive effect" on the relative risk over and above that observed when considered separately. Although HLA-DR3 as well as DQw2 were increased in the D8/17-positive samples, the relative risk did not alter significantly when all these markers were considered together. It may be argued that both these antigen systems are independent of each other and that there is no direct relation between them. In this regard, our results support the observations of Zabriskie,8 who conducted screening of B cell alloantisera against a panel of known HLA-ABC and DR specificities and found no association with any of the known HLA phenotypes. The observed increased occurrence of HLA-DQw2 in the D8/17+ patient group as compared with the D8/17 negative group is indeed intriguing. It could be explained on the basis of some degree of structural homology between the two antigen systems. This could be tested by the recently defined molecular hybridization procedures using specific complementary DNA probes for HLA-DR and DQ gene products (and of the D8/17 antigen, if available). The restriction fragment length polymorphisms may prove extremely useful in identifying particular restriction fragments in patients with rheumatic heart disease with or without DQw2 or D8/17. Acknowledgments We express our gratitude to Prof. John Zabriskie, Rockefeller University, New York, providing the D8/17 monoclonal antibody. B. for References 1. Cheadle WB: Harveian lectures on the various manifestations as exemplified in childhood and early life. Lancet 1889; 1: Stollerman G: Epidemiology ofrheumatic Fever and Streptococcal Infections. New York, Grune & Stratton, 1975, p 92

6 340 Circulation Vol 80, No 2, August Glynn LE, Holborrow EJ: Relation between blood groups, secretor status and susceptibility to rheumatic fever. Arthritis Rheum 1961;4: Read SE, Zabriskie JB: Immunologic concepts in rheumatic fever, in: Clinical Immunology of the Heart. New York, John Wiley Medical Publication 1981, p Wilson MG, Schweitzer M: Pattern of hereditary susceptibility in rheumatic fever. Circulation 1954;10: Jhinghan B, Mehra NK, Reddy KS, Taneja V, Vaidya MC, Bhatia ML: HLA, blood groups and secretor status in patients with established rheumatic fever and rheumatic heart disease. Tissue Antigens 1986;27: Patarroyo ME, Winchester RJ, Vejerano A, Gibofsky FC, Zabriskie JB, Kunkel HG: Association of a B cell alloantigen with susceptibility to rheumatic fever. Nature 1979;278: Zabriskie JB: Rheumatic fever: The interplay between host, genetics, and microbe. Circulation 1985;71: Zabriskie JB: Rheumatic fever: A model for the pathological consequences of microbial-host mimicry. Clin Exp Rheumatol 1986;4: Ad-hoc Committee to Revise the Jones Criteria (Modified) of the Council of Rheumatic Fever and Congenital Heart Disease of the American Heart Association 1984: Jones criteria (revised) for guidance in the diagnosis of rheumatic fever. Circulation 1984;69:204A 11. Terasaki PI, McCelland JD: Microdroplet assay of human serum cytotoxins. Nature 1964;204: Paradis IL, Merrall EJ, Kasell JM, Dauper JH, Rogers RM, Robin BS: Lymphocyte enumeration: a comparison between a modified avidin-biotin-immuno peroxidase system and flow cytometry. J Histochem Cytochem 1984;32: Svejgaard A, Platz P, Ryder LP, Staub-Nielsen L, Thomson M: HLA and disease association A survey. Transplant Rev 1975;22:3 14. Green A: The epidemiological approach to studies of association between HLA and disease: II. Estimation of absolute risks, etiological and preventive fraction. Tissue Antigens 1982;19: Taranta AS, Toros day S, Matrakos WH, Jefler W, Uchida J: Rheumatic fever in monozygotic and dizygotic twins. Circulation 1959;20: Kallman FJ, Reisner D: Twin studies on the significance of genetic factors in tuberculosis. Am Rev Tuberculosis 1943; 47: Nepom GT, Hansen JA: Human immune response genes, in Gupta S, Talal N (eds): Immunology ofrheumatic Diseases. New York, Plenum Publishing Corp, 1985, p Ayoub EM, Barrett DJ, Maclaren NK, Krischer JP: Association of class II: Human histocompatibility leukocyte antigens with rheumatic fever. J Clin Invest 1986;77: Maharaj B, Hammand MG, Appadoo B, Leary WP, Pudifin DJ: HLA-A,B,DR and DQ antigens in black patients with severe chronic rheumatic heart disease. Circulation 1987; 76: Anatasiou-Nana MI, Anderson JL, Carlquist JP, and Nanas JN: HLA-DR typing and lymphocyte subset evaluation in rheumatic heart disease: A search for immune response factors. Am Heart J 1986; 112: Todd JA, Bell JI, McDevitt OH: HLA-DQB gene contributes to susceptibility and resistance to insulin dependent diabetes mellitus. Nature 1987;329:599 KEY WORDS * genetics * avidin-biotin peroxidase complex antibodies * antigens