Cloned primed lymphocyte test cells recognize the fourth, fifth, and sixth hypervariable regions at amino acid positions 65-87 of the DPB1 molecule.

Genetic polymorphisms of the HLA-DPB1 gene in Japanese and Caucasian panel cells defined by PLT were analyzed by the PCR-based genotyping technique PCR-RFLP, and suballeles of DPw3 (DPB1*03) and DP"Cp63" (DPB1*09) could be detected. PLT-defined DPw3 cells were typed by PCR-RFLP as either DPB1*0301 or DPB1*1401. On the other hand, PLT-defined DPCp63-typed cells were typed as DPB1*0901 or DPB1*1001. These results indicate that both DPw3 and DPCp63 are split into two subantigens. DPw2 and DPw4 are DPB1*0201 and 0202 and DPB1*0401 and 0402, respectively. Comparative analysis of the amino acid sequences of the DPw2-, DPw4-, DPw3-, and DPCp63-associated alleles revealed that the fourth (C), fifth (D), and sixth (E) hypervariable regions at amino acid positions 65-87 were shared within the same PLT-defined DP antigen groups, suggesting that these three hypervariable regions are recognized by cloned T cells in PLT, thus determining DP antigen specificity. On the basis of this model, 44 DPB1 alleles can be classified into 18 antigen groups, each of which may possibly represent a PLT-defined single DP specificity.

HLA DPB1*0201 gene confers disease susceptibility in japanese with childhood onset type I diabetes, independent of HLA-DR and DQ genotypes.

HLA is an important etiologic genetic factor in Type I diabetes and specific HLA-class II genes are closely related to the onset of the disease. Many differences in the patterns of susceptible and resistant DRB1, DQA1, and DQB1 genes have been observed among various ethnic groups. We have previously shown that DRB1*0405, DRB1*0901 and DQA1*0301-DQB1*0302 were the major susceptible alleles or haplotype to Type I diabetes while DR-DQ haplotype studies suggested the important role of DR and DQ alleles in susceptibility and resistance in Japanese patients. Based on the analysis of 90 Japanese patients with childhood onset Type I diabetes and 136 unrelated healthy Japanese controls by polymerase chain reaction-restriction fragment polymorphism method (PCR-RFLP), we report here the association of Type I diabetes with DPB1*0201 (relative risk = 2.29; Pc = 0.027) in this population. Comparison of linkage disequilibrium patterns between patients and controls showed that the significantly high prevalence of DPB1*0201 among patients cannot be attributed simply to linkage disequilibrium with susceptible DRB1 alleles and DQA1-DQB1 haplotypes. Our results suggest that in addition to alleles at the DRB1, DQA1, DQB1 loci, polymorphism at DPB1 locus also influences the risk of Type I diabetes.

[Rheumatoid factor in diabetes mellitus (author's transl)].

This study was undertaken to try to determine the incidence of rheumatoid factor in patients with diabetes mellitus by the Hyland RA slide test and the RAHA test (the sensitized sheep red cell agglutination test). Sera from 507 patients with diabetes mellitus and 2073 controls were investigated for the presence of rheumatiod factors. RA test was positive in 8.7 per cent and RAHA test in 6.1 per cent in sera from patients with diabetes mellitus compared to 4.1 percent and 3.6 per cent respectively in healthy subjects.

[Thyroglobulin and microsomal antibodies in diabetes mellitus in childhood (author's transl)].

The purpose of our study was to determine the incidence of thyroid antibodies in diabetes mellitus in childhood and to discuss the correlation between thyroid antibodies and insulin antibodies. The sera of 50 children with diabetes mellitus and 437 children as disease controls were tested by thyroglobulin and microsome-coated tanned red cell hemagglutination test (Fuji Zoki Co. Tokyo). One of the 50 children with diabetes mellitus (2%) was positive with antithyroglobulin antibodies and eleven (22%) were positive with antimicrosomal antibodies compared to 0.4% and 1.1% respectively in 437 disease controls. To clarify the association of insulin antibodies and thyroid antibodies in diabetes mellitus in childhood, insulin antibodies were demonstrated by using a modified method described by Wright. One of the 33 children with negative insulin antibody (2.8%) was positive with antithyroglobulin antibodies and eight (24%) were positive antimicrosomal antibodies. No evident correlation was observed between antithyroid antibodies and insulin antibodies.

[Leucocyte migration inhibition test in diabetes mellitus (author's transl)].

The leucocyte migration inhibition test (LMT) using the agarose plate method introduced by Clausen is simple and highly reproducible. Using thyroglobulin, mitochondria and thyroid microsomal fractions as antigens, LMT was performed on ten patients with insulin dependent diabetics, twenty patients with insulin independent diabetics, eleven patients with Graves' diseases, ten patients with chronic thyroiditis and ten healthy controls. The migration index was expressed as a percentage of migration and calculated from the following formula; (formula: see text). Using a thyroid microsomal fraction, the mean migration index for the insulin dependent diabetics (93.2 +/- 11.6) was significantly lower than in the normal subjects (104.3 +/- 10.9) (0.025 less than p less than 0.05). Using thyroglobulin as an antigen for the insulin dependent diabetics and insulin independent diabetics, the mean migration indices (+/-SD) were 93.7+/-10.8 and 102.8+/-10.6 respectively. The corresponding values for the control group were 100.6+/-7.3. Using a mitochondria, MI values for insulin dependent and independent diabetics were 97.7+/-5.5 and 99.6+/-12.6 respectively, while MI value for the control group was 103.9+/-13.6. The mean migration indices obtained with the mitochondria and thyroglobulin were not significantly depressed when the diabetic groups were compared with the normal subjects.

Thyroglobulin and microsomal antibodies in patients with insulin dependent diabetes mellitus and their relatives.

The sera for 88 parents and 9 siblings of 73 patients with insulin dependent diabetes mellitus in childhood and 437 controls matched in age and sex, were tested by the thyroglobulin and microsome-coated tanned red cell hemagglutination test (Fuji-Zoki Co. Tokyo). None of 73 children with diabetes mellitus had antithyroglobulin antibodies, whereas twelve (16.4%) had antimicrosomal antibodies compared with the incidence of 0.4% and 1.1%, respectively, in 437 controls. In the parents and siblings of these probands, thyroid antibodies were also found in increased incidence. The incidence of antimicrosomal antibodies in the 68 mothers was significantly higher than in controls matched for age and sex, but the incidence of the positive thyroid antibodies in the 20 fathers and 9 siblings was not significantly different from that in control populations. The incidence of thyroid antibodies tended to be higher, though not significant, in parents and siblings of diabetic children with positive thyroid antibodies than in those of diabetics with negative ones. These findings suggest that immunogenetic factors may be responsible for the pathogenesis of some cases of diabetes mellitus in childhood.

HLA in hypertrophic cardiomyopathy and rheumatic heart disease.

1. HLA antigens were determined in 30 Japanese patients with hypertrophic cardiomyopathy. Several antigens were more common in patients compared with controls, but statistically significant differences were not evidenced. In families one and two, six of seven kindred who inherited HLA-A9 and B7 had the disease. None of five kindred lacking HLA-B7 showed evidence of the disease. In families three and four, affecting family members had HLA-A2 and BW-35. Our finding suggest that the HLA system may play some role in the pathogenesis of hypertrophic cardiomyopathy with familial occurrence. 2. Twenty patients with rheumatic valvular heart disease were also studied. There was no significant difference in frequencies of HLA antigens between patients and controls.

Islet-cell antibodies and HLA type in Japanese insulin-dependent diabetics.

Pancreatic islet-cell antibodies (I.C.Ab.) were detected in the sera of 6 of 123 patients with insulin-dependent diabetes mellitus of recent onset, compared with only 2 positive results among the 434 control sera. The prevalence of humoral I.C.Ab. was strongly dependent on the duration of the diabetes, being 4 of 6 I.C.Ab. positive patients during the first year from diagnosis and falling to 2 of 6 I.C.Ab. positive patients at one to three years. Over four years from the time of diagnosis, there was no I.C.Ab. in the sera of the diabetics. Six I.C.Ab. positive in our group of 123 diabetic patients showed no association with any particular B locus antigen or DYT.

The leucocyte migration inhibition test and subpopulations of peripheral lymphocytes in insulin-dependent diabetics.

The leucocyte migration inhibition test (LMT) was performed by the agarose plate method with thyroid and pancreatic antigens in patients with insulin-dependent or independent diabetes mellitus. The mean migration indices with thyroglobulin, thyroid mitochondria and beef insulin were not significantly different in insulin-dependent diabetics from those in insulin-independent diabetics or normal controls. However, significant inhibition of leucocyte migration was observed in insulin-dependent diabetics when thyroid microsome or pancreatic extract was used as antigen. Although no significant difference was found in the percentages of T and B lymphocytes between insulin-dependent diabetics and insulin-independent diabetics or normal controls, the results of LMT strongly suggest the presence of cellular immunity against the thyroid and pancreas in insulin-dependent juvenile-onset diabetes.

HL-A and hypertrophic cardiomyopathy.

HL-A antigens were determined in 26 unrelated Japanese patients with hypertrophic cardiomyopathy. Several antigens were more common in patients compared with controls, but statistically significant differences were not evidenced. We also studied two families in which many had a hypertrophic cardiomyopathy. All the affected individuals revealed HL-A-A9 and B7, while none among the unaffected family members had HL-A-B7. Our findings suggest that the HLA-A system may play some role in the pathogenesis of hypertrophic cardiomyopathy with familial occurrence.

Islet cell antibodies in the Japanese population and subjects with type 1 (insulin-dependent) diabetes.

Islet cell antibodies were studied in 1,112 non-diabetic adults, 473 normal school children and 162 Type 1 (insulin-dependent) diabetic patients in a Japanese population. The prevalence of islet cell antibodies was 0.5%, 0.4% and 32%, respectively. Most islet cell antibodies positive subjects with Type 1 diabetes had short duration of the disease. No patients who had over 10 years from the onset had islet cell antibodies. Six non-diabetic adults with islet cell antibodies were followed for 4 years. Only one with Hashimoto's thyroiditis showed a diabetic pattern in her oral glucose tolerance test. However, none developed overt insulin-dependent diabetes until 1984. Two out of these six subjects continued to be positive for both islet cell antibodies and antithyroid antibodies or antinuclear antibodies. Islet cell antibodies in the remaining four patients disappeared during the second year. It is difficult to predict the onset of Type 1 diabetes by islet cell antibodies in non-diabetic individuals because they may be transient.

[In vitro stimulation of peripheral blood lymphocytes with allogeneic lymphocytes or phytomitogens in patients with insulin-dependent diabetes mellitus (author's transl)].

The lymphocyte transformation response to the allogeneic lymphocytes (mixed lymphocyte culture, MLC) was determined in nineteen well-controlled insulin-dependent diabetics (IDD) and nineteen matched normal subjects. All possible combinations between lymphocytes from the patients and controls were mixed in both one-way and two-way MLC. From the results of one-way MLC, the stimulatory capacity (SC) and responding capacity (RC) of IDD lymphocytes were compared with those of normal lymphocytes as follows: (1) Nm leads to N: 10,538 +/- 3,937 N; normal lymphocytes (2) Nm leads to D: 8;466 +/- 5,387 D; IDD lymphocytes (3) Dm leads to N: 7,562 +/- 3,088 m; mitomycin-treated stimulating lymphocytes (4) Dm leads to D: 7,102 +/- 4,873 (leads to; stimulatory direction, results; M +/- SD cpm) IDD lymphocytes showed a marked depressive function as stimulators (SC, (1) -- (3)), but the RC of IDD lymphocytes was unchanged ((1) -- (2)). Phytomitogen-response was studied simultaneously for the same responding lymphocytes (N, D) of MLC. IDD lymphocytes exhibited significantly decreased responses to phytohemagglutinin P, pokeweed mitogen and concanavalin A.

Antimicrosomal antibodies, gastric parietal cell antibodies and antinuclear factors in insulin dependent diabetes mellitus.

Thyroid antimicrosomal antibodies, gastric parietal cell antibodies (PCA) and antinuclear factors were studied in 208 insulin dependent diabetic (IDD) according to the duration of diabetes and patient's age at the time of testing. Antimicrosomal antibodies were found in 11 out of 47 (23.4%) IDD with the duration of less than one year, however this value declined to 13.1% at 1 to 3 years, 15.3% at 4 to 5 years, 10.8% at 6 to 10 years and 5.8% at more than 10 years. Of the 47 IDD, 7 (14.8%) were positive for gastric parietal cell antibodies. The prevalence of PCA declined with increasing duration of diabetes. However, this decrease in the prevalence of antimicrosomal antibodies and PCA was not so extreme as that of pancreatic islet cell antibodies. Antinuclear factors did not reveal a significant correlation with the duration of diabetes. In normal controls, the prevalence of antimicrosomal antibodies, PCA and the antinuclear factors increased progressively with age. In IDD, the prevalence of the antinuclear factors was also progressively greater with age. However, the prevalence of antimicrosomal antibodies in IDD decreased with age and those of PCA showed the lowest percent in the 40-69 year-age group.

[Subpopulations of peripheral lymphocytes in insulin-dependent diabetics (author's transl)].

There is an increasing evidence that autoimmune mechanisms may have a role in the pathogenesis in insulin-dependent diabetics. The numerical and functional study of peripheral blood lymphocytes in diabetes mellitus might indirectly contribute to the understanding of its pathogenesis. In this study, detection of peripheral blood T lymphocytes was measured by rosettes with sheep red blood cells (SRBC), and B lymphocytes were measured by immunofluorescence with specific antiserum to immunoglobulins. The mean (+/- SD) percentage of SRBC was 67.6 +/- 7.2 in 21 normal subjects, 71.5 +/- 7.0 in 15 insulin-dependent diabetics, and 68.6 +/- 6.7 in 30 insulin-independent diabetics. There was no difference in the absolute T-lymphocyte number per mm3 in these three groups. Insulin-dependent diabetics showed a normal percentage and absolute number of B lymphocytes when compared with normal subjects.

[The clinical and immunological association of insulin antibodies and thyroid antibodies in diabetics (author's transl)].

We have already reported a high rate of occurrence of antimicrosomal antibodies in diabetes mellitus. Thirteen of 507 diabetics (2.5%) were positive with antithyroglobulin antibodies and thirty-one (6.1%) were positive with antimicrosomal antibodies compared to 2.3% and 2.5% respectively in normal controls. Two of 34 insulin dependent diabetics (5.9%) were positive with antithyroglobulin antibodies and ten (29.4%) were positive with antimicrosomal antibodies compared to 2.3% and 4.4% respectively in 473 insulin independent diabetics. To clarify the association of insulin antibodies and thyroid antibodies in diabetics, antithyroid antibodies in 507 diabetics were tested by tanned red cell hemagglutination test and insulin antibodies were demonstrated by using a method descrived by Wright in a modified form. Twelve of 482 diabetics negative-insulin antibody (2.5%) were positive with antithyroglobulin antibodies and thirty(6.2%)were positive with antimicrosomal antibodies. Only one of 25 positive-insulin antibody (4%) was positive with antithyroid antibodies respectively. No evident correlation was observed between antithyroid antibodies and insulin antibodies.

[Leucocyte migration inhibition test in diabetes mellitus (author's transl)].

The leucocyte migration inhibition test (LMT) using the agarose plate method introduced by Clausen is simple and highly reproducible. Using human pancreas extract and beef insulin as antigen, LMT was performed on ten patients with insulin dependent diabetes, twenty patients with insulin independent diabetes, and twelve healthy controls. The migration index was expressed as a percentage of migration calculated from the following formula. Migration index (MI) = average areas of migration in test suspension/average areas of migration in control suspension. Using human pancreas extract, the mean migration index for the insulin dependent diabetics (87.6 +/- 11.1) was significantly lower than in the normal subjects (99.3 +/- 6.3) (p less than 0.05). Using beef insulin as antigen for the insulin dependent diabetics and insulin independent diabetics, the mean migration indices (+/- SD) were 95.8 +/- 14.9 and 98.7 +/- 12.3 respectively. The corresponding values for the control group were 98.9 +/- 7.8. Cellular hypersensitivity to human pancreas extract was shown in the leucocyte migration inhibition test with insulin dependent diabetics, but a negative result was obtained with beef insulin.

A population survey of pancreatic islet cell antibodies and antithyroid antibodies.

In a survey of one Japanese population, we detected pancreatic islet cell antibodies (ICA), antithyroid antibodies (ATA) and antinuclear antibodies (ANA). The prevalence of ICA was 6 out of 1125 cases, or 0.5%. ATA and ANA were detected in 8.9% and 1%, respectively. There were no cases of either type I or type II diabetes in subjects with ICA. But there was one case who had ATA and another with ANA. Serum samples from this population had been obtained once a year from 1979 and one case with neither ATA nor ANA was found positive for ICA in 1980. Identical tests for ICA were performed on 80 childhood diabetics as were carried out on type I diabetics. Pancreatic isles cell surface antibodies (ICsA), ATA and ANA were studied simultaneously. The prevalence of ICA in 80 cases was 36.3% and that of ICsA was 13.8%. 4 cases had both ICA and ICsA. The prevalence of ATA was 11.2% and that of ANA was 16.3%.

Antinuclear antibodies in childhood diabetics.

Antinuclear antibodies (ANA) were detected both in type 1 diabetic children and in control subjects. The incidence of ANA in eighty of these diabetics was 16.3%, as determined using two different substrates, human pancreas and human peripheral leucocytes. The incidence and the patterns in the detection of ANA were the same. Four hundred and seventy-three children and one thousand one hundred and twenty-five adults served as the controls. The incidence of ANA in non-diabetic children was 0.8% and that in one adult population was 1.1%. Therefore, the incidence of ANA in childhood diabetics was significantly higher. We studied autoantibodies in childhood diabetics, in normal children and in one adult population. Pancreatic islet cell antibodies (ICA) were detected in 29 out of 80 type 1 diabetics (36.3%) in two out of 473 normal children (0.4%) and in six cases in one population (0.5%). thyroid microsomal antibodies (MCHA) were detected in 9 out of 80 childhood diabetics and the incidence of MCHA in type 1 diabetics was significantly higher than in the controls.

Circulating immune complexes in the serum of diabetes mellitus in childhood by a modified 125I-C1q binding test.

The 125I-C1q binding test for the detection of soluble immune complexes in native unheated human serum was applied to the study of sera from 52 patients with diabetes mellitus in childhood. This radiolabeled C1q binding test is more sensitive and reproducible among the various methods proposed for the detection of immune complexes. The 125I-C1q binding activity in 52 sera from diabetes mellitus in childhood was 9.47 +/- 0.36% compared to 6.94 +/- 0.74% in normal controls. 125I-C1q binding values in diabetes mellitus in childhood were significantly higher than normal controls. Slight high values were seen in 3 patients with positive anti-DNA-antibodies in diabetes mellitus in childhood. 125I-C1q binding was not significantly increased in patients with positive antithyroid antibodies and insulin antibodies. There was no significant correlation between the duration of diabetes and 125I-C1q binding activity.