- Email: [email protected]
Immunoglobulin allotype markers and HLA DR genes in type I diabetes mellitus
Immunoglobulin
Allotype Markers and HLA DR Genes in Type I Diabetes Mellitus G. Schernthaner and W. R. Mayr
Although it is now well established that insulin-dependent (type I) diabetes mellitus is closely associated with genes of the HLA-DR locus, the genetics of the disease remains an area of controversy. It is generally believed that the HLA-linked diabetes genes provide the majority of disease susceptibility to type I diabetes. however. there is some evidence for the existence of other non-HLA-linked genetic loci predisposing to this disorder. Therefore, allotypes of the Gm (immunoglobulin heavy chain) locus on chromosome 14 and of Km (immunoglobulin light chain) locus on chromosome 2 were studied in 180 caucasoid type I diabetic patients. No association between immunoglobulin allotype markers and the whole group of type I diabetes could be observed. However, a particular immunoglobulin allotype, Glm(aj, and a particular Gm immunoglobulin phenotype (a-x-f + b + ) showed a significant heterogeneity within the diabetics subdivided by HLA-DR type. The data of this study support the concept that (1) the genes in the HLA region provide the majority of, but not the only, genetic susceptibility to type I diabetes mellitus and (2) Gm-associated genes could interact with these susceptibility genes at least in the DR3/4 heterozygote type I diabetics. Further studies should be undertaken in order to elucidate a possible role of these factors in the humoral immune response of type I diabetics and their families.
YPE-I (insulin-dependent) diabetes is a multifactorial disease in which immunogenetic and autoimmune as well as environmental factors play important roles.lm5Recently, cumulative evidence has shown that genes controlling immune responses are linked to the MHC (major histocompatibility complexJ6-* and/or to the immunoglobulin allotype genes. 9~‘oIt is not surprising to find that genes linked to HLA or to immunoglobulin (Ig) genes are associated with immune responses in general and autoimmune disease in particular,” because: (1) HLA-DR antigens (or closely related gene products) are necessary for antigen handling and presentation; markers in this region are important for the interaction of T cells during a response and (2) Ig genes are involved in T cell recognition and control. Furthermore, HLA-A, -B and -C antigens are important at the effector arm of the cellular response, whereas immunoglobulin and the MHC linked C2, C4, and Bf proteins are involved at the effector arm of the humoral response. It is now generally accepted that the major genetic susceptibility to autoimmune endocrine diseases such as type I diabetes, autoimmune thyroid disease, and Addison’s disease is determined by genes of the HLAD/DR region of the MHC complex.“~‘2-‘4 More than 90% of type I diabetics possess either HLA-DR3 or-DR4 or both,15 whereas clinical disease develops in less than 1% of individuals carrying these antigensI This observation led to the proposal that additional genetic loci are involved in the highly complex pathogenesis of type I diabetes.‘2,‘5,‘6 This study was designed to analyze immunoglobulin allotype markers in patients with type I diabetes because limited information has existed in caucasoids up to now.
T
Metabolism, Vol 33, No 9 (September).
1984
MATERIALS AND METHODS Patients Altogether were studied. ranging from ketosis-prone the criteria of
180 type I diabetic caucasoids of the Viennese area The mean age at onset of disease was 14.3 years, 1 to 29 years. All patients suffered from the classical insulin-deficient diabetes type, patients not fullfilling type I diabetes were excluded from the study.
tmmunoglobulin
Allotyping and HLA-typing
Serum samples were typed for IgG (Gm) and k light-chain (Km) allotypes by means of a hemagglutination-inhibition procedure.” We used the notation for human allotype markers recommended by the World Health Organisation. ‘r Typing was carried out for the following allotype markers: Glm(a), Cilm(x). Glm(f), G3m(b), and Km(l). For HLA-typing, 14 antigens of the HLA-A-locus (Al, A2, A3, Al 1. A25, A26, A28, A29, Aw23, Aw24, Aw30, Aw31, Aw32, and Aw3), 19 antigensof the HLA-B locus (B5,B7,B8,B12,Bl3,Bl4, 915, 917, 918, 927, 937, 940, 9~16. 9~22, 9~35, 9~41, 9~49, and 9~50) and 5 antigens of the HLA-C locus (Cwl, Cw2, Cw3, Cw4, and Cw5) were tested using the NIH standard-lymphocytotoxicity test. Six antigens of the HLA-DR locus (DRl, DR2, DR3, DR4, DR5, and DR7) were studied using the antisera of the 8th International Histocompatibility Workshop. Ig allotypes were determined in serum samples of 1602 (Gm) respectively 1334 (Km) healthy Viennese control subjects. HLA-A, -9, and -C antigens were typed in 3000 and HLA-DR antigens in 160 nondiabetic healthy controls.
From the Department of Medicine II and Institute of Blood Group Serology (National Blood Group Reference Laboratory World Health Organisation) National Tissue Typing Reference Laboratory (Council of Europe), University of Vienna Ludwig Boitzmann-Institute for Clinical Endocrinology, Vienna, Austria. Address reprint requests to Univ. Doz. Dr G. Schernthaner, Dept of Medicine II, University of Vienna, Garnisongasse 13, A-1090 Vienna, Austria. 01984 by Grune & Stratton, Inc.
833
SCHERNTHANER AND MAYR
834
STATISTICAL ANALYSIS
DISCUSSION
The evaluation of the statistical significance was done by using the x2 test, the RR was calculated according to Woolf.”
Considerable interest has been focused on the genetic polymorphism, which control the immunoglobulin allotypic determinants of the heavy chain and their possible associations with certain diseases which are believed to have an autoimmune basis. In recent studies, possible associations between Gm allotypes and several autoimmune diseases*%**and with immune responsiveness to specific antigens26 have been reported. Because most of the studies were carried out in Japan,1021322 information for the majority of autoimmune diseases in caucasoids is lacking. Significantly increased frequencies of a particular Gm haplotype (Gm1.2.2’= Gmaxg) were found in patients with myasthenia gravis,22 Hashimoto thyroiditis,20 and Graves’ disease,20 all of them are disorders associated with HLA-DR3 in caucasoid populations. Furthermore, a recent investigation*’ suggests that genes linked to the immunoglobulin heavy-chain allotype locus on chromosome 14 may influence, in addition to MHC genes on chromosome 6, susceptibility to gluten-sensitive enteropathy: 5 out of 30 white patients with coeliac disease lacking HLA-B8 and HLA-DR327 showed the IgG(Gm) immunoglobulin heavy-chain phenotype Gm(f, n, b). The present study revealed no disturbance of Gm allotype and phenotype frequencies in caucasoid patients with insulin-dependent type I diabetes, findings that are in accordance with results recently obtained in Japanese type I diabetic patients.” However, type I diabetics with different HLA-DR phenotypes showed a significant heterogeneity of a particular Gm allotype and phenotype suggesting that Gm-associated genes could play some additional role in the susceptibility of type I diabetes. Gm-associated
RESULTS Ig allotypes Glm (a, x, f), G3m (b), Km (l), and HLA-DR phenotypes of type I diabetics and controls are shown in Table 1. There was no significant association between type I diabetes and any particular immunoglobulin allotype in these patients, whereas the well-known association with HLA-DR3 and/or DR4 was confirmed. HLA-DR3 or DR4 were found in 68 (48%) or 100 (71%) respectively of the 141 type I diabetics, but in only 32 (20%) or 42 (26%) respectively of the 160 controls. The highest relative risk (RR) was found for DR3/DR4 heterozygotes (RR 11.6; Table 1). Immunoglobulin Gm and Km allotypes and Gm phenotypes according to HLA-DR phenotypes of the type I diabetics are shown in Tables 2 and 3. A significantly increased frequency (65.9%) of the Glm(a) was found in DR3/4 heterozygous type I diabetics compared to patients with other HLA-DR phenotypes (Table 2). The DR3/4 heterozygous patients demonstrated a significantly reduced frequency (36.6%) of the Gm phenotype (a-x - f + b +) compared to type I diabetics with other HLA-DR phenotypes (Table 3). The 17 classical type I diabetic patients lacking HLA-DR3 and DR4 demonstrated similar distribution of Ig allotypes as those who showed the typical HLA associations (DR3/x; DR4/x). The increase in Glm(a) seemed to be mostly confined to haplotypes a+x-f+b+ and a+x+f+b+ although this did not reach statistical significance.
Table 1. lmmunoglobulin
Gl m (a, x, f 1, G3m (b), Km(l)
Type
Allotypes and HLA-DR Phenotypes
I Dmbetics
N=
N = 1602(1334)
180
Positive
in Type I Diabetics and Controls.
Controls
Percent
Percent
Positive
X2
P
RR
Glm(a)
89
49.4
726
45.3 I
1.11
0.29
1.18
G 1m(x)
24
13.3
229
14.3
0.12
0.73
0.92
Glm(f)
166
92.2
1498
93.5
0.43
0.5 1
0.82
GBm(bl
170
94.4
1513
94.4
0.00
1.oo
1.00
Km(l)
27
15.0
213
16.0
0.11
0.74
0.93
DR3/4
44
31.2
6
3.7
40.79
<1o-6
11.64
DR3/x
24
17.0
26
16.2
0.03
0.86
1.06
DR4/x
56
39.7
36
22.5
10.47
1.21 x 1o-3
2.27
Neither DRB/nor DR4
17
12.1
92
57.5
67.01
<1o-6
0.10
N = 141
N = 1602
N = 1334
N = 160
HLA-DRx means a HLA-DR specificity different from DR3 and DR4.
Gm MARKERS AND TYPE
I
835
DIABETES MELLITUS
Table 2. Analysis of the lmmunoglobulin
Gl m Allotypes According to HLA-DR Phenotypes
(DR3/4:
DR3/x:
DR4/x:
DR3 - /DR4-)
in 141
Type I Diabetics. HLA-DR
Phenotype Neither DR3
lmmunoglobulin
All Patients
DR3/4
DR3/x
DR4Ix
nor DR4
Allotypes
(N = 141)
(N = 44)
(N = 24)
(N = 56)
IN = 17)
X2
P
24 (42.9%)
6 (35.3%)
9.41
0.02 0.20
8 (33.3%)
Glm(a)
67 (47.5%)
29 (65.9%)
GlmW
20 114.2%)
8 (18.2%)
2 (8.3%)
10(17.9%)
0 (0.0%)
4.68
Glm(f)
130 (92.2%)
38 (86.4%)
24 (100%)
52 (92.9%)
16 (94.1%)
4.23
0.24
G3m(b)
134 (95.O%i
41 (93.2%)
24 (100%)
53 (94.6%)
15 (94.1%)
1.62
0.65
8 (14.3%)
3 (17.6%)
0.26
0.97
Km(l)
21 (14.9%)
HLA-DRx means a HLA-DR specificity
3 112.5%)
7 (15.9%) different from DR3 and DR4
Table 3. Analysis of 5 lg Gm Phenotypes
According to HLA-DR Phenotypes
(DR3/4;
DR3/x;
DR4lx;
DR3-
/DR4-)
in 137 type I
diabetics. HLA-DR Phenotvpe Neither DR3 lmmunoglobulln
All Patlents
DR3/4
DR3/x
DR4/x
nor DR4
Gm Phenotype
(N = 137)
IN = 41)
(N = 24)
(N = 551
(N = 17)
a+x-f-b-
5 (3.6%)
2 (4.9%)
0 (0.0%)
2 (3.6%)
1 (5.9%)
1.33
a+x+f-b-
2 (1.5%)
1 (2.4%)
0 (0.0%)
1(1.8%)
0 (0.0%)
0.93
0.82
a+x-f+b+
39 (28.5%)
17 (41.5%)
6 (25%)
5 (29.4%)
5.49
0.14
0 (0.0%)
3.75
0.29
11 (64.7%)
7.73
0.05
a+x+f+b+
17 (12.4%)
a-x-f+b+
74 (54%)
HLA-DRx means a HLA-DR specificity
11 (20%)
6 (14.6%)
2 (8.3%)
15 (36.6%)
16 (66.7%)
9 (16.4%) 32 (58.2%)
X1
P
0.72
different from DR3 and DR4.
genes could interact with the HLA genes at least in the DR3/4 heterozyte type I diabetics. An increase of the Km( 1) allotype marker was found neither in the whole group of type I diabetics nor in the subgroups divided by the HLA-DR type. Recently, Dunsworth et al** found no evidence for tight linkage between type I diabetes and the Kidd (Jk) red blood cell marker locus, which contrasts with the positive earlier report by Hodge et a1.29The lack of linkage between Kidd blood group and type I diabetes seems to be relevant to our data because the Kidd locus is linked to Km on chromosome 2. Although we found no evidence for association between Gm and Km allotype markers and the whole group of type I diabetes, this negative finding does not completely exclude the involvement of such factors, especially since a particular Gm allotype, Glm(a), was found with significantly increased frequency and a particular Gm phenotype (a-x-f+b+) with significantly decreased frequency in the DR3/4 heterozygote patients. Nakao et ali0 have suggested that the immune response to insulin may be Gm-associated in Japanese type I diabetics, findings that have to be confirmed in caucasoid type I diabetics. Furthermore, Uno et a124 have provided evidence using the sibpair analysis that both HLA and Gm play a role in the susceptibility to Graves’ disease, although there was no demonstrable association between Gm markers and the disease in unrelated patients.
Although there seems little doubt that the major genetic susceptibility to type I diabetes is determined by genes in the HLA complex, there are several facts of the immune genetic basis which remain unresolved. Interestingly, the distribution of organ specific autoantibodies-islet-cell antibody, thyroid microsomal antibody, and gastric parietal cell antibody-in the nonaffected siblings of type I diabetic probands is independent of the potential HLA genetic susceptibility, as judged by the HLA haplotype concordance with the proband. Therefore, further studies dealing with the investigation of IgG heavy-chain (Gm) allotypes in families with type I diabetes and other autoimmune endocrinopathies may provide further useful information concerning the development of autoantibodies in diseases with a suspected autoimmune etiology. ADDENDUM
After submission of this paper, Bertrams and coworkers reported similar data showing no association between Gm markers and the whole group of type-1 diabetics (J Immunogenet 10:305-3 10, 1983). However, Gm markers were not analyzed within the diabetics subdivided by HLA DR type. REFERENCES 1. Cahill GF, McDevitt HO: Insulin-dependent diabetes mellitus: the initial lesion. N Engl J Med 304:1454 1981 2. Nerup J, Christy M, Green A, et al: HLA and insulindependent diabetes-population studies, in Kiibberling J, Tattersall R
836
(eds): The Genetics of Diabetes Mellitus. Academic Press, New York, 1982, pp 3542 3. Cudworth AG, Wolf E: The genetic susceptibility to type-I (insulin-dependent) diabetes mellitus. Clin Endocrinol Metab 11:389-408,1982 4. Schernthaner G: The relationship between clinical, immunological and genetic factors in insulin-dependent diabetes. In Kiibberling J, Tattersall R (eds): The Genetics of Diabetes Mellitus. Academic Press, New York, 1982, pp 99-l 14 5. Yoon J-W: Viruses in the pathogenesis of type-1 diabetes. In Kolb H, Schernthaner G, Gries FA (eds): Diabetes and Immunology: Pathogenesis and Immunotherapy. Hans Huber Pub], BernStuttgart-Vienna, 1983, pp 1 l-37 6. Stastny P, Ball EJ, Dry PJ, et al: The human immune response region (HLA-D) and disease susceptibility. Immunological Rev 70:113-153, 1983 7. Schernthaner G, Ludwig H, Mayr WR: Immunoglobulin G-insulin antibodies and immune region-associated alloantigens in insulin-dependent diabetes mellitus. J Clin Endocrinol Metab 48:403-407,1979 8. Ludwig H, Schernthaner G, Mayr WR: la-type alloantigens and humoral autoimmune responsiveness in insulin-dependent diabetes mellitus. Metabolism 28:797-800, 1979 9. Smith SM. Ness DB, Talcott JA, et al: Genetic control of IgM responses to (TG)-A-L, H-2 and Ig-1 linkage. Immunogenetics 4~221-232, 1977 10. Nakao Y, Matsumoto H, Miyazaki T, et al: IgG heavy-chain (Gm) allotypes and immune response to insulin in insulin-requiring diabetes mellitus. N Engl J Med 304:407409, 198 1 11. Welsh KJ: HLA-and Gm-linked genes and antigen nonspecific immunoregulation. Immunol Today 3:129-l 30, 1982 12. Rotter JI, Anderson CE, Rubin R, et al: HLA genotypic study of insulin-dependent diabetes. The excess of DR3/DR4 heterozygotes allows rejection of the recessive hypothesis. Diabetes 32:169-174, 1983 13. Farid NR, Bear JC: The human major histocompatibility complex and endocrine disease. Endocrinol Rev 2:50-86, I98 1 14. Schleusener H. Schernthaner G, Mayr WR, et al: HLA-DR3 and HLA-DR5 associated thyrotoxicosis-two different types of toxic diffuse goiter. J Clin Endocrinol Metab 56:781-785, 1983 15. Svejgaard A, Platz P, Ryder LP: Insulin-dependent diabetes mellitus, in Terasaki PI, (ed): Histocompatibility testing. UCLA Tissue Typing Laboratory, Los Angeles, CA 1980 pp 638-656 16. Cudworth AG, Wolf E: The genetics of type I (insulin-
SCHERNTHANER
AND MAYR
dependent) diabetes, in Kolb H, Schernthaner G, Gries FA (eds): Diabetes and Immunology: Pathogenesis and Immunotherapy, Current Problems in Clinical Biochemistry. Hans Huber Publ, BernStuttgart-Vienna, 1983,45-55 17. Schanfield MS: Genetic markers of human immunoglobulins, in Fudenberg HH, Stites DP, Caldwell JL, et al (eds): Basic and Clinical Immunology, 2nd ed. Lange Medical Publishers, Los Altos 1978~~59-65 18. Review of the notation for the allotypic and related markers of human immunoglobulins. Eur J Immunol6:599-601, 1976 19. Woolf B: On estimating the relation between blood group and disease. Ann Hum Genet 19:25 l-253, 1955 20. Farid NR, Newton RM, Elke PN, et al: Gm phenotypes in autoimmune thyroid disease. J Immunogenet 4:429432, 1977 21. Nakao Y, Matsumoto H, Miyazaki T, et al: IgG heavy chain allotypes (Cm) in autoimmune diseases. Clin Exp Immunol 42:2026.1980 22. Nakao Y, Matsumoto H, Miyazaki T, et al: Gm allotypes in myasthenia gravis. Lancet 1:677-680, 1980 23. Whittingham S, Mathews JD, Schantield MS, et al: HLA and Gm genes in systemic lupus erythematosus. Tissue Antigens 21:5&57, 1983 24. Uno H, Sasazuki T, Tamai H, et al: Two major genes, linked to HLA and Gm, control susceptibility to Graves’ disease. Nature 292:768-770, 1981 25. Pollack MS, Safai B, Myskowski PL, et al: Frequencies of HLA and Gm immunogenetic markers in Kaposi’s sarcoma. Tissue Antigens 21:1-8, 1983 26. Whittingham S, Mathews JD, Schanfield MS, et al: Interactive effect of Cm allotypes and HLA-B locus antigens on the human antibody response to a bacterial antigen. Clin Exp Immunol 40:815, 1980 27. Kagnoff MF, Weiss JB, Brown RJ, et al: Immunoglobulin allotype markers in gluten-sensitive enteropathy. Lancet 1:952-953, 1983 28. Dunsworth TS, Rich SS, Swanson J, et al: No evidence for linkage between diabetes and the Kidd marker. Diabetes 31:991993, 1982 29. Hodge SE, Anderson CE, Neiswanger K, et al: Close genetic linkage between diabetes mellitus and Kidd blood group. Lancet 2:893-895, 198 1 30. Gorsuch AN, Dean B, Bottazzo GF, et al: Evidence that type I diabetes and thyreogastric autoimmunity have different genetic determinants. Br Med J I : l45-, 1980
Allotype Markers and HLA DR Genes in Type I Diabetes Mellitus G. Schernthaner and W. R. Mayr
Although it is now well established that insulin-dependent (type I) diabetes mellitus is closely associated with genes of the HLA-DR locus, the genetics of the disease remains an area of controversy. It is generally believed that the HLA-linked diabetes genes provide the majority of disease susceptibility to type I diabetes. however. there is some evidence for the existence of other non-HLA-linked genetic loci predisposing to this disorder. Therefore, allotypes of the Gm (immunoglobulin heavy chain) locus on chromosome 14 and of Km (immunoglobulin light chain) locus on chromosome 2 were studied in 180 caucasoid type I diabetic patients. No association between immunoglobulin allotype markers and the whole group of type I diabetes could be observed. However, a particular immunoglobulin allotype, Glm(aj, and a particular Gm immunoglobulin phenotype (a-x-f + b + ) showed a significant heterogeneity within the diabetics subdivided by HLA-DR type. The data of this study support the concept that (1) the genes in the HLA region provide the majority of, but not the only, genetic susceptibility to type I diabetes mellitus and (2) Gm-associated genes could interact with these susceptibility genes at least in the DR3/4 heterozygote type I diabetics. Further studies should be undertaken in order to elucidate a possible role of these factors in the humoral immune response of type I diabetics and their families.
YPE-I (insulin-dependent) diabetes is a multifactorial disease in which immunogenetic and autoimmune as well as environmental factors play important roles.lm5Recently, cumulative evidence has shown that genes controlling immune responses are linked to the MHC (major histocompatibility complexJ6-* and/or to the immunoglobulin allotype genes. 9~‘oIt is not surprising to find that genes linked to HLA or to immunoglobulin (Ig) genes are associated with immune responses in general and autoimmune disease in particular,” because: (1) HLA-DR antigens (or closely related gene products) are necessary for antigen handling and presentation; markers in this region are important for the interaction of T cells during a response and (2) Ig genes are involved in T cell recognition and control. Furthermore, HLA-A, -B and -C antigens are important at the effector arm of the cellular response, whereas immunoglobulin and the MHC linked C2, C4, and Bf proteins are involved at the effector arm of the humoral response. It is now generally accepted that the major genetic susceptibility to autoimmune endocrine diseases such as type I diabetes, autoimmune thyroid disease, and Addison’s disease is determined by genes of the HLAD/DR region of the MHC complex.“~‘2-‘4 More than 90% of type I diabetics possess either HLA-DR3 or-DR4 or both,15 whereas clinical disease develops in less than 1% of individuals carrying these antigensI This observation led to the proposal that additional genetic loci are involved in the highly complex pathogenesis of type I diabetes.‘2,‘5,‘6 This study was designed to analyze immunoglobulin allotype markers in patients with type I diabetes because limited information has existed in caucasoids up to now.
T
Metabolism, Vol 33, No 9 (September).
1984
MATERIALS AND METHODS Patients Altogether were studied. ranging from ketosis-prone the criteria of
180 type I diabetic caucasoids of the Viennese area The mean age at onset of disease was 14.3 years, 1 to 29 years. All patients suffered from the classical insulin-deficient diabetes type, patients not fullfilling type I diabetes were excluded from the study.
tmmunoglobulin
Allotyping and HLA-typing
Serum samples were typed for IgG (Gm) and k light-chain (Km) allotypes by means of a hemagglutination-inhibition procedure.” We used the notation for human allotype markers recommended by the World Health Organisation. ‘r Typing was carried out for the following allotype markers: Glm(a), Cilm(x). Glm(f), G3m(b), and Km(l). For HLA-typing, 14 antigens of the HLA-A-locus (Al, A2, A3, Al 1. A25, A26, A28, A29, Aw23, Aw24, Aw30, Aw31, Aw32, and Aw3), 19 antigensof the HLA-B locus (B5,B7,B8,B12,Bl3,Bl4, 915, 917, 918, 927, 937, 940, 9~16. 9~22, 9~35, 9~41, 9~49, and 9~50) and 5 antigens of the HLA-C locus (Cwl, Cw2, Cw3, Cw4, and Cw5) were tested using the NIH standard-lymphocytotoxicity test. Six antigens of the HLA-DR locus (DRl, DR2, DR3, DR4, DR5, and DR7) were studied using the antisera of the 8th International Histocompatibility Workshop. Ig allotypes were determined in serum samples of 1602 (Gm) respectively 1334 (Km) healthy Viennese control subjects. HLA-A, -9, and -C antigens were typed in 3000 and HLA-DR antigens in 160 nondiabetic healthy controls.
From the Department of Medicine II and Institute of Blood Group Serology (National Blood Group Reference Laboratory World Health Organisation) National Tissue Typing Reference Laboratory (Council of Europe), University of Vienna Ludwig Boitzmann-Institute for Clinical Endocrinology, Vienna, Austria. Address reprint requests to Univ. Doz. Dr G. Schernthaner, Dept of Medicine II, University of Vienna, Garnisongasse 13, A-1090 Vienna, Austria. 01984 by Grune & Stratton, Inc.
833
SCHERNTHANER AND MAYR
834
STATISTICAL ANALYSIS
DISCUSSION
The evaluation of the statistical significance was done by using the x2 test, the RR was calculated according to Woolf.”
Considerable interest has been focused on the genetic polymorphism, which control the immunoglobulin allotypic determinants of the heavy chain and their possible associations with certain diseases which are believed to have an autoimmune basis. In recent studies, possible associations between Gm allotypes and several autoimmune diseases*%**and with immune responsiveness to specific antigens26 have been reported. Because most of the studies were carried out in Japan,1021322 information for the majority of autoimmune diseases in caucasoids is lacking. Significantly increased frequencies of a particular Gm haplotype (Gm1.2.2’= Gmaxg) were found in patients with myasthenia gravis,22 Hashimoto thyroiditis,20 and Graves’ disease,20 all of them are disorders associated with HLA-DR3 in caucasoid populations. Furthermore, a recent investigation*’ suggests that genes linked to the immunoglobulin heavy-chain allotype locus on chromosome 14 may influence, in addition to MHC genes on chromosome 6, susceptibility to gluten-sensitive enteropathy: 5 out of 30 white patients with coeliac disease lacking HLA-B8 and HLA-DR327 showed the IgG(Gm) immunoglobulin heavy-chain phenotype Gm(f, n, b). The present study revealed no disturbance of Gm allotype and phenotype frequencies in caucasoid patients with insulin-dependent type I diabetes, findings that are in accordance with results recently obtained in Japanese type I diabetic patients.” However, type I diabetics with different HLA-DR phenotypes showed a significant heterogeneity of a particular Gm allotype and phenotype suggesting that Gm-associated genes could play some additional role in the susceptibility of type I diabetes. Gm-associated
RESULTS Ig allotypes Glm (a, x, f), G3m (b), Km (l), and HLA-DR phenotypes of type I diabetics and controls are shown in Table 1. There was no significant association between type I diabetes and any particular immunoglobulin allotype in these patients, whereas the well-known association with HLA-DR3 and/or DR4 was confirmed. HLA-DR3 or DR4 were found in 68 (48%) or 100 (71%) respectively of the 141 type I diabetics, but in only 32 (20%) or 42 (26%) respectively of the 160 controls. The highest relative risk (RR) was found for DR3/DR4 heterozygotes (RR 11.6; Table 1). Immunoglobulin Gm and Km allotypes and Gm phenotypes according to HLA-DR phenotypes of the type I diabetics are shown in Tables 2 and 3. A significantly increased frequency (65.9%) of the Glm(a) was found in DR3/4 heterozygous type I diabetics compared to patients with other HLA-DR phenotypes (Table 2). The DR3/4 heterozygous patients demonstrated a significantly reduced frequency (36.6%) of the Gm phenotype (a-x - f + b +) compared to type I diabetics with other HLA-DR phenotypes (Table 3). The 17 classical type I diabetic patients lacking HLA-DR3 and DR4 demonstrated similar distribution of Ig allotypes as those who showed the typical HLA associations (DR3/x; DR4/x). The increase in Glm(a) seemed to be mostly confined to haplotypes a+x-f+b+ and a+x+f+b+ although this did not reach statistical significance.
Table 1. lmmunoglobulin
Gl m (a, x, f 1, G3m (b), Km(l)
Type
Allotypes and HLA-DR Phenotypes
I Dmbetics
N=
N = 1602(1334)
180
Positive
in Type I Diabetics and Controls.
Controls
Percent
Percent
Positive
X2
P
RR
Glm(a)
89
49.4
726
45.3 I
1.11
0.29
1.18
G 1m(x)
24
13.3
229
14.3
0.12
0.73
0.92
Glm(f)
166
92.2
1498
93.5
0.43
0.5 1
0.82
GBm(bl
170
94.4
1513
94.4
0.00
1.oo
1.00
Km(l)
27
15.0
213
16.0
0.11
0.74
0.93
DR3/4
44
31.2
6
3.7
40.79
<1o-6
11.64
DR3/x
24
17.0
26
16.2
0.03
0.86
1.06
DR4/x
56
39.7
36
22.5
10.47
1.21 x 1o-3
2.27
Neither DRB/nor DR4
17
12.1
92
57.5
67.01
<1o-6
0.10
N = 141
N = 1602
N = 1334
N = 160
HLA-DRx means a HLA-DR specificity different from DR3 and DR4.
Gm MARKERS AND TYPE
I
835
DIABETES MELLITUS
Table 2. Analysis of the lmmunoglobulin
Gl m Allotypes According to HLA-DR Phenotypes
(DR3/4:
DR3/x:
DR4/x:
DR3 - /DR4-)
in 141
Type I Diabetics. HLA-DR
Phenotype Neither DR3
lmmunoglobulin
All Patients
DR3/4
DR3/x
DR4Ix
nor DR4
Allotypes
(N = 141)
(N = 44)
(N = 24)
(N = 56)
IN = 17)
X2
P
24 (42.9%)
6 (35.3%)
9.41
0.02 0.20
8 (33.3%)
Glm(a)
67 (47.5%)
29 (65.9%)
GlmW
20 114.2%)
8 (18.2%)
2 (8.3%)
10(17.9%)
0 (0.0%)
4.68
Glm(f)
130 (92.2%)
38 (86.4%)
24 (100%)
52 (92.9%)
16 (94.1%)
4.23
0.24
G3m(b)
134 (95.O%i
41 (93.2%)
24 (100%)
53 (94.6%)
15 (94.1%)
1.62
0.65
8 (14.3%)
3 (17.6%)
0.26
0.97
Km(l)
21 (14.9%)
HLA-DRx means a HLA-DR specificity
3 112.5%)
7 (15.9%) different from DR3 and DR4
Table 3. Analysis of 5 lg Gm Phenotypes
According to HLA-DR Phenotypes
(DR3/4;
DR3/x;
DR4lx;
DR3-
/DR4-)
in 137 type I
diabetics. HLA-DR Phenotvpe Neither DR3 lmmunoglobulln
All Patlents
DR3/4
DR3/x
DR4/x
nor DR4
Gm Phenotype
(N = 137)
IN = 41)
(N = 24)
(N = 551
(N = 17)
a+x-f-b-
5 (3.6%)
2 (4.9%)
0 (0.0%)
2 (3.6%)
1 (5.9%)
1.33
a+x+f-b-
2 (1.5%)
1 (2.4%)
0 (0.0%)
1(1.8%)
0 (0.0%)
0.93
0.82
a+x-f+b+
39 (28.5%)
17 (41.5%)
6 (25%)
5 (29.4%)
5.49
0.14
0 (0.0%)
3.75
0.29
11 (64.7%)
7.73
0.05
a+x+f+b+
17 (12.4%)
a-x-f+b+
74 (54%)
HLA-DRx means a HLA-DR specificity
11 (20%)
6 (14.6%)
2 (8.3%)
15 (36.6%)
16 (66.7%)
9 (16.4%) 32 (58.2%)
X1
P
0.72
different from DR3 and DR4.
genes could interact with the HLA genes at least in the DR3/4 heterozyte type I diabetics. An increase of the Km( 1) allotype marker was found neither in the whole group of type I diabetics nor in the subgroups divided by the HLA-DR type. Recently, Dunsworth et al** found no evidence for tight linkage between type I diabetes and the Kidd (Jk) red blood cell marker locus, which contrasts with the positive earlier report by Hodge et a1.29The lack of linkage between Kidd blood group and type I diabetes seems to be relevant to our data because the Kidd locus is linked to Km on chromosome 2. Although we found no evidence for association between Gm and Km allotype markers and the whole group of type I diabetes, this negative finding does not completely exclude the involvement of such factors, especially since a particular Gm allotype, Glm(a), was found with significantly increased frequency and a particular Gm phenotype (a-x-f+b+) with significantly decreased frequency in the DR3/4 heterozygote patients. Nakao et ali0 have suggested that the immune response to insulin may be Gm-associated in Japanese type I diabetics, findings that have to be confirmed in caucasoid type I diabetics. Furthermore, Uno et a124 have provided evidence using the sibpair analysis that both HLA and Gm play a role in the susceptibility to Graves’ disease, although there was no demonstrable association between Gm markers and the disease in unrelated patients.
Although there seems little doubt that the major genetic susceptibility to type I diabetes is determined by genes in the HLA complex, there are several facts of the immune genetic basis which remain unresolved. Interestingly, the distribution of organ specific autoantibodies-islet-cell antibody, thyroid microsomal antibody, and gastric parietal cell antibody-in the nonaffected siblings of type I diabetic probands is independent of the potential HLA genetic susceptibility, as judged by the HLA haplotype concordance with the proband. Therefore, further studies dealing with the investigation of IgG heavy-chain (Gm) allotypes in families with type I diabetes and other autoimmune endocrinopathies may provide further useful information concerning the development of autoantibodies in diseases with a suspected autoimmune etiology. ADDENDUM
After submission of this paper, Bertrams and coworkers reported similar data showing no association between Gm markers and the whole group of type-1 diabetics (J Immunogenet 10:305-3 10, 1983). However, Gm markers were not analyzed within the diabetics subdivided by HLA DR type. REFERENCES 1. Cahill GF, McDevitt HO: Insulin-dependent diabetes mellitus: the initial lesion. N Engl J Med 304:1454 1981 2. Nerup J, Christy M, Green A, et al: HLA and insulindependent diabetes-population studies, in Kiibberling J, Tattersall R
836
(eds): The Genetics of Diabetes Mellitus. Academic Press, New York, 1982, pp 3542 3. Cudworth AG, Wolf E: The genetic susceptibility to type-I (insulin-dependent) diabetes mellitus. Clin Endocrinol Metab 11:389-408,1982 4. Schernthaner G: The relationship between clinical, immunological and genetic factors in insulin-dependent diabetes. In Kiibberling J, Tattersall R (eds): The Genetics of Diabetes Mellitus. Academic Press, New York, 1982, pp 99-l 14 5. Yoon J-W: Viruses in the pathogenesis of type-1 diabetes. In Kolb H, Schernthaner G, Gries FA (eds): Diabetes and Immunology: Pathogenesis and Immunotherapy. Hans Huber Pub], BernStuttgart-Vienna, 1983, pp 1 l-37 6. Stastny P, Ball EJ, Dry PJ, et al: The human immune response region (HLA-D) and disease susceptibility. Immunological Rev 70:113-153, 1983 7. Schernthaner G, Ludwig H, Mayr WR: Immunoglobulin G-insulin antibodies and immune region-associated alloantigens in insulin-dependent diabetes mellitus. J Clin Endocrinol Metab 48:403-407,1979 8. Ludwig H, Schernthaner G, Mayr WR: la-type alloantigens and humoral autoimmune responsiveness in insulin-dependent diabetes mellitus. Metabolism 28:797-800, 1979 9. Smith SM. Ness DB, Talcott JA, et al: Genetic control of IgM responses to (TG)-A-L, H-2 and Ig-1 linkage. Immunogenetics 4~221-232, 1977 10. Nakao Y, Matsumoto H, Miyazaki T, et al: IgG heavy-chain (Gm) allotypes and immune response to insulin in insulin-requiring diabetes mellitus. N Engl J Med 304:407409, 198 1 11. Welsh KJ: HLA-and Gm-linked genes and antigen nonspecific immunoregulation. Immunol Today 3:129-l 30, 1982 12. Rotter JI, Anderson CE, Rubin R, et al: HLA genotypic study of insulin-dependent diabetes. The excess of DR3/DR4 heterozygotes allows rejection of the recessive hypothesis. Diabetes 32:169-174, 1983 13. Farid NR, Bear JC: The human major histocompatibility complex and endocrine disease. Endocrinol Rev 2:50-86, I98 1 14. Schleusener H. Schernthaner G, Mayr WR, et al: HLA-DR3 and HLA-DR5 associated thyrotoxicosis-two different types of toxic diffuse goiter. J Clin Endocrinol Metab 56:781-785, 1983 15. Svejgaard A, Platz P, Ryder LP: Insulin-dependent diabetes mellitus, in Terasaki PI, (ed): Histocompatibility testing. UCLA Tissue Typing Laboratory, Los Angeles, CA 1980 pp 638-656 16. Cudworth AG, Wolf E: The genetics of type I (insulin-
SCHERNTHANER
AND MAYR
dependent) diabetes, in Kolb H, Schernthaner G, Gries FA (eds): Diabetes and Immunology: Pathogenesis and Immunotherapy, Current Problems in Clinical Biochemistry. Hans Huber Publ, BernStuttgart-Vienna, 1983,45-55 17. Schanfield MS: Genetic markers of human immunoglobulins, in Fudenberg HH, Stites DP, Caldwell JL, et al (eds): Basic and Clinical Immunology, 2nd ed. Lange Medical Publishers, Los Altos 1978~~59-65 18. Review of the notation for the allotypic and related markers of human immunoglobulins. Eur J Immunol6:599-601, 1976 19. Woolf B: On estimating the relation between blood group and disease. Ann Hum Genet 19:25 l-253, 1955 20. Farid NR, Newton RM, Elke PN, et al: Gm phenotypes in autoimmune thyroid disease. J Immunogenet 4:429432, 1977 21. Nakao Y, Matsumoto H, Miyazaki T, et al: IgG heavy chain allotypes (Cm) in autoimmune diseases. Clin Exp Immunol 42:2026.1980 22. Nakao Y, Matsumoto H, Miyazaki T, et al: Gm allotypes in myasthenia gravis. Lancet 1:677-680, 1980 23. Whittingham S, Mathews JD, Schantield MS, et al: HLA and Gm genes in systemic lupus erythematosus. Tissue Antigens 21:5&57, 1983 24. Uno H, Sasazuki T, Tamai H, et al: Two major genes, linked to HLA and Gm, control susceptibility to Graves’ disease. Nature 292:768-770, 1981 25. Pollack MS, Safai B, Myskowski PL, et al: Frequencies of HLA and Gm immunogenetic markers in Kaposi’s sarcoma. Tissue Antigens 21:1-8, 1983 26. Whittingham S, Mathews JD, Schanfield MS, et al: Interactive effect of Cm allotypes and HLA-B locus antigens on the human antibody response to a bacterial antigen. Clin Exp Immunol 40:815, 1980 27. Kagnoff MF, Weiss JB, Brown RJ, et al: Immunoglobulin allotype markers in gluten-sensitive enteropathy. Lancet 1:952-953, 1983 28. Dunsworth TS, Rich SS, Swanson J, et al: No evidence for linkage between diabetes and the Kidd marker. Diabetes 31:991993, 1982 29. Hodge SE, Anderson CE, Neiswanger K, et al: Close genetic linkage between diabetes mellitus and Kidd blood group. Lancet 2:893-895, 198 1 30. Gorsuch AN, Dean B, Bottazzo GF, et al: Evidence that type I diabetes and thyreogastric autoimmunity have different genetic determinants. Br Med J I : l45-, 1980