To clarify heterogeneity in Japanese adult-onset type 1 diabetes, we analyzed the HLA-DR and -DQ haplotypes, depending on the clinical phenotype, and compared them with those in childhood-onset type 1 diabetes (CO). The patients in a previously reported Ehime Study were divided into subgroups by the mode of onset of diabetes: 68 acute-onset type 1 diabetic patients (AO) and 28 slowly progressive type 1 diabetic patients (SO). HLA haplotypes were compared with those of 80 CO patients and 190 control subjects. Two major susceptible HLA haplotypes in the Japanese, DRB1*0405-DQB1*0401 (DR4) and DRB1*0901-DQB1*0303 (DR9), were significantly increased in the AO and CO groups, but only DR9 was increased in the SO group. AO subjects had a higher frequency of DR9 than CO subjects. Accordingly, the DR9:DR4 frequency increased with increasing age of onset. Another susceptible haplotype, DRB1*0802-DQB1*0302 (DR8), was involved only in the CO group. Analysis of haplotype combinations revealed that DR4 and DR9 had significant dosage effects on the AO and CO groups (P < 0.0001), but only DR9 had such an effect in the SO group (P < 0.03). These results suggest differences in the contribution of HLA class II haplotypes to susceptibility of type 1 diabetes depending on the clinical phenotype and also indicate that HLA class II haplotypes may be associated with the onset age of type 1 diabetes.

Type 1 diabetes is the result of a destruction of pancreatic β-cells and leads to complete insulin deficiency (1,2). Although type 1 diabetes is frequently considered to be a childhood disease and is characterized by a rapid progression to insulin dependency, the clinical onset of type 1 diabetes is not confined to children. Epidemiological data suggest that 30–50% of type 1 diabetic patients may develop clinical signs of diabetes after age 20 years (3,4). Moreover, there is increasing evidence to indicate that type 1 diabetes, especially when developed in adulthood, is clinically and immunologically heterogeneous. Insulin secretion abruptly ceases in fulminant type 1 diabetes (5), whereas pancreatic β-cell function can be preserved for over a decade in some patients, a condition that is referred to as slowly progressive insulin-dependent diabetes (6) or latent autoimmune diabetes in adults (LADA) (7). Compared with our extensive knowledge of the epidemiological, clinical, and genetic characteristics of childhood-onset type 1 diabetes, our understanding of adult-onset type 1 diabetes is incomplete.

Susceptibility to type 1 diabetes is determined by both environmental and genetic factors. Although multiple genes have been implicated, HLA class II genes, especially the HLA-DR and -DQ genes, are most important and are estimated to account for ∼50% of the susceptibility to the disease (8). These HLA associations vary depending on geographic and ethnic origin (9). It has been shown that Japanese type 1 diabetic patients have different HLA associations than Caucasian patients. In Caucasian populations, the DRB1*0301-DQB1*0201 and/or DRB1*04-DQB1*0302 haplotypes indicate a strong predisposition to type 1 diabetes, whereas the DRB1*1501-DQB1*0602 haplotype confers strong protection (2,10,11). In the Japanese population, the DRB1*03-DQB1*0201 haplotype is nearly absent and the DRB1*04-DQB1*0302 is rare (12). Instead, three haplotypes—DRB1*0405-DQB1*0401, DRB1*0802-DQB1*0302, and DRB1*0901-DQB1*0303—that are rare in Caucasian populations confer susceptibility to type 1 diabetes (6,13–15). Moreover, the DRB1*1502-DQB1*0601 haplotype, which is rare in Caucasian populations, is a major protective haplotype, in addition to DRB1*1501-DQB1*0602.

For adult-onset type 1 diabetes, which follows a diverse clinical course after onset, little information is available concerning the immunogenetic background with reference to the clinical phenotype. In Caucasian populations, adult-onset type 1 diabetes is characterized by a longer symptomatic period before diagnosis and a better preservation of β-cell function at the age of diagnosis compared with the childhood-onset variety (16). The DR3/DR4 genotype, which is the highest indicator of risk in Caucasian type 1 diabetes, occurs at a higher frequency in cases of younger age of onset (16–20), whereas the heterogeneity of DRB1 is increased in subjects who are older at the time of diagnosis (18). A lower frequency of DQB1*0201/DQB1*0302 has been reported in LADA patients compared with typical type 1 diabetic patients (21). In the Japanese population, patients with slowly progressive type 1 diabetes (6) and fulminant type 1 diabetes (5,22,23) show differences in HLA-DR or -DQ haplotype involvement compared with patients with typical acute-onset type 1 diabetes. It is not clear whether these clinical phenotypes, namely LADA, slowly progressive type 1 diabetes, and fulminant type 1 diabetes, represent a late or rapid manifestation of type 1 diabetes or distinct clinical entities. An analysis of immunogenetic background in relation to clinical phenotype would provide useful information on the nature of these subtypes.

The intent of this study was to clarify the immunogenetic characteristics, including the contribution of the genotypic combination of HLA class II haplotypes, in relation to clinical heterogeneity in Japanese adult-onset type 1 diabetes in comparison with those of childhood-onset type 1 diabetes.

Type 1 diabetes is defined by insulin deficiency resulting from the destruction of pancreatic islet β-cells and includes those cases attributable to an autoimmune process and those for which the etiology is unknown (24,25).

In this study, insulin secretion was examined in a total of 4,980 adult-onset (age >20 years) diabetic patients who were registered in the hospital-based Ehime study (26). Insulin-deficient type 1 diabetic patients were divided into four groups according to the mode of onset of diabetes as reported by Kobayashi et al. (6) and Imagawa et al. (5). In the case of the acute-onset type 1 diabetes (AO) group, insulin treatment was initiated within 3 months, and in the intermediate-onset group, insulin treatment was initiated between 3 and 12 months. In the slow-onset group, insulin treatment was initiated >12 months after the diagnosis of diabetes or the development of hyperglycemic symptoms. Also in this group, patients with the GAD antibody were diagnosed as having slowly progressive type 1 diabetes (SO), as reported by Kobayashi et al. (6). Fulminant type 1 diabetes was diagnosed as reported by Imagawa A et al. (5) and excluded from the AO group. Of the 4,980 patients, 113 (2.3%) were diagnosed as having type 1 diabetes. Of these 113 patients, 104 (92%) could be classified according to the mode of onset: 52 (50%) were acute onset, 18 (17.3%) were intermediate onset, 25 (24%) were slow onset (20 of whom had the GAD antibody and were diagnosed as slowly progressive type 1 diabetic), and 9 (8.7%) had fulminant type 1 diabetes. Furthermore, 19 additional patients (11 in the AO and 8 in the SO groups) whose characteristics were confirmed by the above criteria were recruited. Therefore, a total of 63 AO patients and 28 SO patients were recruited in the analysis of the HLA class II genotyping.

The sex ratio (male/female) was 26/37 in the AO group and 14/14 in the SO group. The age of onset was 39.3 ± 15.1 (range 20–82) years in the AO group versus 46.9 ± 14.0 (range 25–87) years in the SO group (P < 0.05 vs. AO group). The duration of diabetes before insulin treatment was 1.2 ± 0.9 months in the AO group versus 60.8 ± 57.7 months in the SO group (P < 0.05). The duration of diabetes at the examination was 10.7 ± 7.6 years in the AO group and 14.6 ± 10.3 years in the SO group. BMI at the examination was 21.3 ± 2.3 kg/m2 years in the AO group and 20.8 ± 3.3 kg/m2 in the SO group. Daily insulin dosage was 0.60 ± 0.24 units/kg in the AO group and 0.58 ± 0.32 units/kg in the SO group. The prevalence of a history of diabetic ketoacidosis was higher in the AO than in the SO group (50.8 vs. 8.3%; P < 0.05). GAD-Ab positivity at the time of examination in the AO group was 60.3%. In each group, all patients who were positive for the insulinoma-associated protein 2 autoantibodies were also GAD-Ab positive. We also confirmed the clinical characteristics of fulminant type 1 diabetes, as reported previously (5,23). To clarify the difference between each group in adult-onset type 1 diabetes, a comparison was made between the HLA class II frequency for the AO and SO groups and the childhood-onset type 1 diabetes (CO) group.

A total of 80 patients (age <18 years) in the CO group were recruited from the Ehime prefecture, some of whose clinical and immunogenetic characteristics have been already reported by Kida and colleagues (27,28). All these CO patients were considered to be acute-onset types based on the above-mentioned criteria. The age of onset was 8.0 ± 4.3 (range 1.0–16.6) years and the sex ratio (male/female) was 39/41.

Serum C-peptide levels were measured in diabetic subjects 2 h after a standard breakfast (400–600 calories of mixed meal) with the usual morning insulin injection or 6 min after a 1-mg intravenous glucagon load after an overnight fast (13,29,30). The criterion for the insulin-deficient state was <0.33 nmol/l, as previously described (13,29), which was confirmed repeatedly before or at the time of the first registration and during follow-up over time.

The control group was comprised of subjects (128 men and 62 women) with a normal glucose tolerance, as assessed by a 75-g oral glucose tolerance test, with no history of diabetes among first-degree relatives. All patients and control subjects were informed of the purpose of the study and their consent obtained. The study was approved by the ethics committees of the Ehime University Hospital, Matsuyama Shimin Hospital, Ehime Prefectural Central Hospital, and Matsuyama Red Cross Hospital.

Autoantibody assay.

GAD-Ab was determined at the time of sampling by means of a commercially available radioimmunoassay kit (Cosmic, Tokyo, Japan), as previously described (26). Insulinoma-associated protein 2 autoantibodies were determined by radioligand binding assays, as previously described (31).

HLA genotyping and C-peptide concentration.

HLA-DRB1 and HLA-DQB1 were typed and serum C-peptide levels were measured, as previously described (26).

Statistical analysis.

Results are given as means ± SD, unless otherwise indicated. The χ2 test or Fisher’s exact probability test was used to determine the statistical significance of differences between group frequencies. Group comparisons of the levels of clinical parameters were analyzed by the Mann-Whitney U test. The prevalence of the DRB1 and DQB1 genotype in the four clinical phenotype groups was assessed relative to their prevalence in the 190 control subjects. The phase or pair of DRB1-DQB1 haplotypes in each subject was inferred manually based on the linkage disequilibrium between the two loci. In the analysis of haplotype combination, increases in prevalence were assessed relative to the prevalence of the neutral genotype combination (X/X) (Table 3), assuming that X/X represents a stable neutral denominator that would be expected to have a similar prevalence in diabetic patients and control subjects (32). The haplotype and genotype frequencies were compared by a χ2 test based on a 2 × 2 table. To confirm the association between the DRB1-DQB1 haplotype detected by the χ2 test and acute onset, slow onset, or childhood onset, the relative predispositional effects (RPE) method (33) was also applied. P < 0.05 was considered to be statistically significant.

Frequency of HLA DRB1-DQB1 haplotypes.

Among the type 1 diabetes−susceptible DRB1-DQB1 haplotypes reported in the Japanese population, DRB1*0405-DQB1*0401 (DR4) and DRB1*0901-DQB1*0303 (DR9) were significantly more frequent in the AO group than in control subjects (Table 1). In the SO group, the DR9 haplotype was more frequent compared with in control subjects, but the difference was not significant. In the CO group, in addition to the DR4 and DR9 haplotypes, the frequency of another type 1 diabetes−susceptible haplotype, DRB1*0802-DQB1*0302, was significantly increased compared with that in control subjects. When the AO and CO groups were compared, the DR9 haplotype was significantly more frequent in the AO group (P < 0.002). Furthermore, DRB1*1302-DQB1*0604 in the AO group, DRB1*1101-DQB1*0301 in the SO group, and DRB1*1101-DQB1*0302 in the CO group were significantly increased compared with in control subjects. All these results were confirmed by the RPE method (data not shown). The frequency of DRB1*1502-DQB1*0601 and DRB1*1501-DQB1*0602, protective haplotypes to Japanese type 1 diabetes, was rare in all groups.

Effect of DRB1-DQB1 haplotype combination.

In the AO group, both the DR4 and DR9 haplotypes conferred a high susceptibility to type 1 diabetes in the heterozygous state (odds ratio [OR] for DR4/X and DR9/X was 15.6 and 17.7, respectively) and a much stronger susceptibility was observed when it was present in the homozygote (OR for DR4/DR4 and DR9/DR9 was 73.8 and 95.9, respectively) (Table 2). In the CO group, a similar, but slightly weaker dosage effect of DR4 and DR9 was observed (the OR for DR4/X and DR9/X was 27.8 and 10.2, respectively, and for DR4/DR4 and DR9/DR9 was 49.2 and 29.5, respectively). For the DR8 haplotype combination, the heterozygote (DR8/X) was significantly more frequent in the CO group compared with in control subjects and showed a high OR. It is noteworthy that, when combined with the DR4 or DR9 haplotypes, the DR8 haplotype combination, especially the DR4/DR8 in the CO group, showed a higher OR than each heterozygote (DR4/X, DR9/X, or DR8/X) and was even higher than those of the homozygote (DR4/DR4 or DR9/DR9). Similar, but weaker effects of the DR8 combination, DR4/DR8 or DR9/DR8, were observed in the AO group.

In the SO group, the heterozygote of the DR9 haplotype (DR9/X) showed a slightly elevated OR (2.0) and the homozygote of the DR9 haplotype (DR9/DR9) showed a significantly higher OR (6.3, 95% confidence interval [CI] 1.17–34.3) than that of the heterozygote (DR9/X). In the DR8 haplotype, the DR4/DR8 combination showed a high OR (8.43, CI 1.02–69.58) compared with X/X. Of note is the fact that 25% of patients in the SO group possessed a neutral haplotype combination (X/X), a significantly higher percentage than that found in the AO group (3.2%; P < 0.005).

Because the OR differed between the heterozygote and homozygote of DR4 and DR9 haplotype, the dosage effect of these haplotypes (i.e., the number of haplotypes in each subject) was assessed by Armitage’s trend test. The DR4 and DR9 haplotype showed a significant dosage effect in the AO and CO groups (P < 0.0001), and only DR9 showed a significant dosage effect in the SO group (P < 0.03). On the other hand, the DR8 haplotype showed a dosage effect only in the CO group (P < 0.0001).

Frequency of DRB1-DQB1 haplotype with respect to onset age in acute-onset type 1 diabetes.

Although the clinical phenotypes for the AO and CO groups had a similar acute onset, the participation of the DR4 and DR9 haplotypes was different between these groups (P < 0.05) (Table 3). Therefore, the frequency of the DR4 and DR9 haplotypes with respect to onset age was examined. AO and CO subjects were combined as acute-onset type 1 diabetes and divided by onset age into three groups (ages 1–18, 20–49, and ≥50 years); the haplotype frequency was then compared between each group. The frequencies of DR9 and DR4 were significantly different between these groups (P < 0.05). The ratio of the DR9/DR4 haplotype frequency increased with onset age. Therefore, DR9 was a major contributory haplotype in subjects in the late-onset AO group. The DR8 haplotype was mainly found in the CO group and was not observed in patients who developed acute-onset type 1 diabetes after age 36 years.

This hospital-based study demonstrated 1) the frequency of a clinically heterogeneous group in adult-onset type 1 diabetes, and 2) the different contributions of type 1 diabetes−susceptible HLA class II haplotypes DR4 (DRB1*0405-DQB1*0401), DR9 (DRB1*0901-DQB1*0303), and DR8 (DRB1*0802-DQB1*0302) to adult-onset and childhood-onset type 1 diabetes.

Frequency of the clinically heterogeneous group of adult-onset type 1 diabetes.

In the previous hospital-based Ehime study, we observed that 2.3% of adult-onset diabetes was insulin-deficient type 1 diabetes (26). The results of the present study indicate that the AO group (classical type 1 diabetes) accounted for ∼50% of the adult-onset type 1 diabetic patients, the fulminant type 1 diabetes for ∼10%, and the SO group for ∼25%. These data are not consistent with previously reported characteristics of adult-onset type 1 diabetes (16), as >50% of the cases of adult-onset type 1 diabetes rapidly progressed to insulin deficiency. In a nationwide survey in Japan, the frequency of fulminant type 1 diabetes was reported to be ∼20% of the cases of acute-onset type 1 diabetes (34), most of which were adult-onset diabetes. These data are consistent with the present data because the frequency of fulminant type 1 diabetes was 14.1% among the AO plus fulminant diabetic patients, whereas no case was found among the CO group. Our findings in the SO group are also consistent with a previous report in which ∼20% of type 1 diabetes was reported as being slow onset (6).

Different contribution of HLA DRB1-DQB1 haplotype to age of onset.

HLA DR4, DR9, and DR8 haplotypes have been reported to confer susceptibility to type 1 diabetes in the Japanese population (6,13–15). The acute-onset type of both adult-onset and childhood-onset type 1 diabetes is considered to be classical type 1 diabetes. However, we observed differences in the contribution of the HLA-DR haplotype between these two groups (Tables 1–3), as the contribution of the DR9 haplotype increased with higher onset age and the involvement of the DR8 haplotype was mainly observed in the CO group.

No reports have appeared to date regarding the relation between the genetic background of Japanese type 1 diabetes and the age of onset. Some disagreements on Japanese type 1 diabetes−susceptible HLA class II haplotypes can be found in previous reports. The average age of onset of the patients examined was low in the reports of Kobayashi et al. (age 21.6 years) (6) and Kawabata et al. (age 16.1 years) (15), who found a significant susceptibility to DR4, DR9, and DR8. These findings are in general agreement with those for the CO group in this study. Concerning the DR9 haplotype, Awata et al. (13) showed a neutral or weak preposition effect, as they analyzed patients with a younger age of onset (age 26.4 years). The average age of onset of the patients was relatively high in the report of Tanaka et al. (34.2 years old) (22), and as a result, that group did not report a susceptibility to DR8. Taken together, the variation in onset age in the patient population examined might explain the differences in Japanese type 1 diabetes susceptible haplotypes reported to date.

Onset age−dependent HLA class II heterogeneity has been reported for Caucasian populations. The percentage of the DQB1*0201/DQB1*0302 (DR3/DR4) heterozygote, the most susceptible haplotype combination in Caucasian populations, decreases with age of onset (16,17). Conversely, the proportion of non-DR3/non-DR4 patients increases in older-onset type 1 diabetes. Caillat-Zucman et al. (17) reported that 26% of typical type 1 diabetes occurring after age 30 years is non-DR3/non-DR4; that is, it lacks any strong susceptible HLA class II haplotypes. However, the present study showed that only 6.4% of the AO group did not have type 1 diabetes−susceptible HLA class II haplotypes. It is noteworthy that in the SO group, 28.6% of the patients did not have a susceptible HLA class II haplotype. These discrepancies may be due to the fact that the previous reports (16,17) might not have strictly categorized the clinical phenotype of adult-onset type 1 diabetes, or that ethnic differences that could influence the HLA class II haplotype in adult-onset type 1 diabetes might be present.

Different contribution of genetic combination of HLA DRB1-DQB1 haplotype to adult- and childhood-onset type 1 diabetes.

The DR4 and DR9 haplotypes in the AO group showed dosage-dependent effects. A similar but weaker phenomenon, the higher OR in the homozygote of susceptible haplotypes than in the heterozygote haplotypes, was observed in the CO group. In Caucasian populations, such dosage effects of each susceptible haplotype have not been observed to date, because DR3/DR4 heterozygous effects are so strongly linked to susceptibility to type 1 diabetes (35–37). Kawabata et al. (15) reported differences in the effects of DR4 and DR9 haplotypes on the development of type 1 diabetes, with a dominant-like effect for the DR4 haplotype and a recessive-like effect for the DR9 haplotype. They did not find susceptibility to type 1 diabetes in DR9 heterozygotes. These discrepancies may be due to differences in onset age because those investigators analyzed mainly childhood-onset patients but also included some adult-onset patients (mean age 16.1 ± 12.4 years, range 2–59 years), as described above.

The DR4/DR9 genotype showed a higher susceptibility (OR 29.5) compared with the heterozygote of DR4 or DR9 in the AO group, but not as strong as the homozygote of these haplotypes. Similar results were obtained for the CO group. These results are inconsistent with those for Caucasian populations in which the DR3/DR4 heterozygotes have a strong susceptibility effect. The synergistic effect of Caucasian DR3/DR4 can be attributed to the transcomplementation of the DQ-αβ heterodimer, which arises from the fact that the DQA1 alleles of Caucasian DR3 and DR4 haplotypes, DQA1*0301 and DQA1*0501, are different (36). Because the DQA1 allele of the Japanese DR4 and DR9 haplotypes has the same amino acid sequence, DQA1*0303 and DQA1*0302, respectively (15), a transcomplementation effect of the DQ gene is not possible in the Japanese DR4/DR9 heterozygote. On the other hand, the DR4/DR8 genotype showed a higher susceptibility than DR4/DR9 in the AO group, especially in that of homozygotes (DR4/DR4 or DR9/DR9). An increased risk of the Japanese DR4/DR8 genotype was also reported by Awata et al. (13). The DQA1*0301 allele of the Japanese DR8 haplotype has the same amino acid sequence as DR4 and DR9 (12), and as a result, a transcomplementation effect would not be possible in the DR4/DR8 heterozygote. However, the DQB1*0302 allele in the DR8 haplotype is a strong type 1 diabetes−susceptible non-Asp allele in Caucasian populations. Therefore, it seems likely that this allele might generate a heterozygous effect in the Japanese DR4/DR8 genotype, similar to the Caucasian DR3/DR4.

In the SO group, the DR9 haplotype combination (DR9/DR9, DR9/X) was related to type 1 diabetes susceptibility. Moreover, patients with a neutral haplotype combination (X/X) made up 25% of the SO group, which was more frequent than that of the AO group (3.2%). Kobayashi et al. (6) reported an involvement of the DR4 haplotype in the slow-onset type. It is possible that differences in patient selection might have affected these discrepancies. In Caucasian populations, type 1 diabetic patients who possess DR3 without DR4 are reported to be associated with a phenotype with slower onset (38). Tuomi et al. (21) reported a lower frequency of DQB1*0201/DQB1*0302 in LADA patients compared with typical type 1 diabetic patients. We also reported a lower frequency of type 1 diabetes−susceptible haplotypes in GAD-Ab–positive non–insulin-deficient diabetic patients (26). Although the SO group was not the same as their LADA subjects because all the patients were in an insulin-deficient state in this study, differences in the participation of the type 1 diabetes−susceptible HLA haplotype, namely less involvement of the strong susceptible genotype, might exist in both Caucasian and Japanese slow-onset phenotypes of type 1 diabetes.

Possible effect of HLA class II antigen on onset of type 1 diabetes.

The most plausible explanation for the association of the specific HLA class II antigen with autoimmune diseases involves the direct role of the HLA molecule in specific peptide antigen binding and presentation to CD4+ T-cells. Different HLA-DR/DQ molecules might have different binding affinities to disease-associated peptides (39). The presentation of these peptides by an antigen-presenting cell (APC) occurs when many class II molecules may be occupied by autologous peptides. T-cells are activated by APCs that express as few as 210–340 specific peptide/class II complexes (40). The class II molecules on APCs may be more highly expressed in homozygotes than in heterozygotes, and they could bind more disease-associated peptides, resulting in a higher risk for development of type 1 diabetes, as was observed in the present study. In acute-onset type 1 diabetes, the contribution of the DR9 haplotype increases with increasing age at onset. Furthermore, only DR9 is associated with slowly progressive type 1 diabetes. These results indicate that DR9 might be related to the slow progression of β-cell destruction, resulting in late-onset type 1 diabetes or slowly progressive type 1 diabetes. On the other hand, DR4 might be related to the early onset of type 1 diabetes. The age-related association of HLA class I molecules has been observed in Japanese young-onset type 1 diabetic patients (mean age 17.4 ± 12.4 years) (41), whereas the association of HLA class II molecules with β-cell destruction has been reported in childhood-onset type 1 diabetes (42). However, the issue of whether or not the HLA class I or II molecule is associated with the onset age or β-cell destruction in adult-onset type 1 diabetes has not been addressed.

In summary, the present study revealed the clinical and immunogenetic heterogeneity of adult-onset type 1 diabetes, more than half of which cases were acute onset. DR9 confers susceptibility to late-onset type 1 diabetes or slowly progressive type 1 diabetes, whereas DR4 shows involvement in early-onset type 1 diabetes. These results suggest that the HLA class II haplotype is an important factor in determining the age of onset of type 1 diabetes.

TABLE 1

Haplotype frequency of HLA DRB1–DQB1 in Japanese patients with type 1 diabetes and control subjects

Adult onset
COControlP
OR (95% CI) vs. control
AOSOAO vs. controlSO vs. controlCO vs. controlAO vs. childhood onsetAOSOCO
n 126 56 160 380 — — — — — — — 
DRB1*0101–DQB1*0501 1 (0.8) 5 (8.9) 11 (6.9) 24 (6.3) 0.0088 — — — 0.12 (0.02–0.89) — — 
DRB1*0401–DQB1*0301 — 1 (1.8) 2 (1.3) 3 (0.8) — — — — — — — 
DRB1*0403–DQB1*0302 2 (1.6) 2 (3.6) 1 (0.6) 14 (3.7) — — 0.0486 — — — 0.16 (0.02–1.26) 
DRB1*0405–DQB1*0401 34 (27) 11 (19.6) 53 (33) 43 (11.3) <0.0001 — <0.0001 — 2.9 (1.75–4.80) 1.92 (0.92–3.98) 3.88 (2.46–6.13) 
DRB1*0406–DQB1*0302 1 (0.8) — — 10 (2.6) — — — — — — — 
DRB1*0407–DQB1*0302 — 1 (1.8) 2 (1.3) 2 (0.5) — — — — — — — 
DRB1*0410–DQB1*0402 — — 2 (1.3) 3 (0.8) — — — — — — — 
DRB1*0802–DQB1*0302 5 (4) 3 (5.4) 16 (10) 11 (2.9) — — 0.0017 — 1.39 (0.47–4.07) 1.9 (0.51–7.03) 4.88 (2.20–10.82) 
DRB1*0802–DQB1*0402 — 1 (1.8) 1 (0.6) 7 (1.8) — — — — — — — 
DRB1*0803–DQB1*0601 7 (5.6) 6 (10.7) 4 (2.5) 32 (8.4) — — 0.0128 — — — 0.36 (0.12–1.03) 
DRB1*0901–DQB1*0303 54 (42.9) 14 (25) 39 (24.4) 56 (14.7) <0.0001 0.0508 0.0092 0.0014 4.34 (2.78–6.82) 1.93 (0.99–3.76) 2.59 (1.62–4.16) 
DRB1*1101–DQB1*0301 — 3 (5.4) 2 (1.3) 4 (1.1) — 0.0483 — — — 5.32 (1.16–24.43) — 
DRB1*1101–DQB1*0302 — 1 (1.8) 4 (2.5) 1 (0.3) — — 0.0288 — — — 9.72 (1.08–87.64) 
DRB1*1201–DQB1*0301 3 (2.4) — 4 (2.5) 10 (2.6) — — — — — — — 
DRB1*1302–DQB1*0604 10 (7.9) 2 (3.6) 8 (5) 7 (1.8) 0.00254 — — — 4.59 (1.71–12.34) — — 
DRB1*1401–DQB1*0502 1 (0.8) — — 8 (2.1) — — — — — — — 
DRB1*1401–DQB1*0503 — — — 7 (1.8) — — — — — — — 
DRB1*1405–DQB1*0503 1 (0.8) — — 7 (1.8) — — — — — — — 
DRB1*1501–DQB1*0602 — 1 (1.8) 1 (0.6) 34 (8.9) <0.0001 — <0.0001 — 0.00 0.19 (0.02–1.38) 0.064 (0.001–0.47) 
DRB1*1502–DQB1*0601 2 (1.6) — 2 (1.3) 34 (8.9) 0.0042 0.0138 <0.0001 — 0.16 (0.04–0.69) 0.00 0.13 (0.03–0.54) 
DRB1*1602–DQB1*0502 2 (1.6) — 1 (0.6) 3 (0.8) — — — — — — — 
Others 3 (2.4) 5 (8.9) 7 (4.4) 60 (15.8) — — — — — — — 
Adult onset
COControlP
OR (95% CI) vs. control
AOSOAO vs. controlSO vs. controlCO vs. controlAO vs. childhood onsetAOSOCO
n 126 56 160 380 — — — — — — — 
DRB1*0101–DQB1*0501 1 (0.8) 5 (8.9) 11 (6.9) 24 (6.3) 0.0088 — — — 0.12 (0.02–0.89) — — 
DRB1*0401–DQB1*0301 — 1 (1.8) 2 (1.3) 3 (0.8) — — — — — — — 
DRB1*0403–DQB1*0302 2 (1.6) 2 (3.6) 1 (0.6) 14 (3.7) — — 0.0486 — — — 0.16 (0.02–1.26) 
DRB1*0405–DQB1*0401 34 (27) 11 (19.6) 53 (33) 43 (11.3) <0.0001 — <0.0001 — 2.9 (1.75–4.80) 1.92 (0.92–3.98) 3.88 (2.46–6.13) 
DRB1*0406–DQB1*0302 1 (0.8) — — 10 (2.6) — — — — — — — 
DRB1*0407–DQB1*0302 — 1 (1.8) 2 (1.3) 2 (0.5) — — — — — — — 
DRB1*0410–DQB1*0402 — — 2 (1.3) 3 (0.8) — — — — — — — 
DRB1*0802–DQB1*0302 5 (4) 3 (5.4) 16 (10) 11 (2.9) — — 0.0017 — 1.39 (0.47–4.07) 1.9 (0.51–7.03) 4.88 (2.20–10.82) 
DRB1*0802–DQB1*0402 — 1 (1.8) 1 (0.6) 7 (1.8) — — — — — — — 
DRB1*0803–DQB1*0601 7 (5.6) 6 (10.7) 4 (2.5) 32 (8.4) — — 0.0128 — — — 0.36 (0.12–1.03) 
DRB1*0901–DQB1*0303 54 (42.9) 14 (25) 39 (24.4) 56 (14.7) <0.0001 0.0508 0.0092 0.0014 4.34 (2.78–6.82) 1.93 (0.99–3.76) 2.59 (1.62–4.16) 
DRB1*1101–DQB1*0301 — 3 (5.4) 2 (1.3) 4 (1.1) — 0.0483 — — — 5.32 (1.16–24.43) — 
DRB1*1101–DQB1*0302 — 1 (1.8) 4 (2.5) 1 (0.3) — — 0.0288 — — — 9.72 (1.08–87.64) 
DRB1*1201–DQB1*0301 3 (2.4) — 4 (2.5) 10 (2.6) — — — — — — — 
DRB1*1302–DQB1*0604 10 (7.9) 2 (3.6) 8 (5) 7 (1.8) 0.00254 — — — 4.59 (1.71–12.34) — — 
DRB1*1401–DQB1*0502 1 (0.8) — — 8 (2.1) — — — — — — — 
DRB1*1401–DQB1*0503 — — — 7 (1.8) — — — — — — — 
DRB1*1405–DQB1*0503 1 (0.8) — — 7 (1.8) — — — — — — — 
DRB1*1501–DQB1*0602 — 1 (1.8) 1 (0.6) 34 (8.9) <0.0001 — <0.0001 — 0.00 0.19 (0.02–1.38) 0.064 (0.001–0.47) 
DRB1*1502–DQB1*0601 2 (1.6) — 2 (1.3) 34 (8.9) 0.0042 0.0138 <0.0001 — 0.16 (0.04–0.69) 0.00 0.13 (0.03–0.54) 
DRB1*1602–DQB1*0502 2 (1.6) — 1 (0.6) 3 (0.8) — — — — — — — 
Others 3 (2.4) 5 (8.9) 7 (4.4) 60 (15.8) — — — — — — — 

Data are n (%) unless otherwise indicated. “Others” refers to rare haplotypes whose frequency in both patients and control subjects is < 5. AO, acute onset; CO, childhood onset; SO, slowly progressive type 1 diabetes.

View Large
TABLE 2

Genetic combinations of the HLA DRB1-DQB1 haplotype in Japanese patients with type 1 diabetes and control subjects

Adult onset
COControlP
OR (95% CI) vs. control
AO (n = 63)SO (n = 28)AO vs. controlSO vs. controlCO vs. controlAOSOCO
n 63 28 80 190 — — — — — — 
DR4/DR4 5 (7.9) 1 (3.6) 5 (6.3) 2 (1.1) <0.0001 NS 0.0001 73.8 (8.49–640.89) 4.2 (0.34–52.63) 49.2 (6.60–366.45) 
DR4/X 9 (14.3) 5 (17.9) 24 (30.0) 17 (8.9) 0.0002 NS <0.0001 15.6 (3.08–79.27) 2.5 (0.70–8.81) 27.8 (7.45–103.51) 
DR9/DR9 13 (20.6) 3 (10.7) 6 (7.5) 4 (2.1) <0.0001 0.0493 0.0001 95.9 (15.84–580.30) 6.3 (1.17–34.3) 29.5 (5.30–164.17) 
DR9/X 15 (23.8) 6 (21.4) 13 (16.3) 25 (13.2) <0.0001 NS 0.0002 17.7 (3.77–83.21) 2.0 (0.62–6.63) 10.2 (2.68–39.04) 
DR4/DR9 12 (19.0) 2 (7.1) 11 (13.8) 12 (6.3) <0.0001 NS <0.0001 29.5 (5.83–149.15) 1.4 (0.26–7.61) 18.0 (4.36–74.55) 
DR8/DR8 0 (0.0) 0 (0.0) 1 (1.3) 0 (0.0) — — NS — — — 
DR9/DR8 1 (1.6) 0 (0.0) 2 (2.54) 1 (0.5) NS NS 0.0140 29.5 (1.32–661.10) — 39.3 (2.73–565.76) 
DR4/DR8 3 (4.8) 2 (7.1) 7 (8.8) 2 (1.1) 0.0021 NS <0.0001 44.3 (4.53–431.56) 8.43 (1.02–69.58) 68.8 (9.76–485.39) 
DR8/X 1 (1.6) 1 (3.6) 5 (6.3) 5 (2.6) NS NS 0.0008 5.90 (0.45–76.95) 1.69 (0.17–16.57) 19.7 (3.60–107.417) 
R/Z 2 (3.2) 1 (3.6) 3 (3.8) 63 (33.2) NS NS NS 0.94 (0.13–6.86) 0.13 (0.02–1.12) 0.94 (0.18–4.82) 
X/X 2 (3.2) 7 (25.0) 3 (3.8) 59 (31.1) — — — — — — 
Adult onset
COControlP
OR (95% CI) vs. control
AO (n = 63)SO (n = 28)AO vs. controlSO vs. controlCO vs. controlAOSOCO
n 63 28 80 190 — — — — — — 
DR4/DR4 5 (7.9) 1 (3.6) 5 (6.3) 2 (1.1) <0.0001 NS 0.0001 73.8 (8.49–640.89) 4.2 (0.34–52.63) 49.2 (6.60–366.45) 
DR4/X 9 (14.3) 5 (17.9) 24 (30.0) 17 (8.9) 0.0002 NS <0.0001 15.6 (3.08–79.27) 2.5 (0.70–8.81) 27.8 (7.45–103.51) 
DR9/DR9 13 (20.6) 3 (10.7) 6 (7.5) 4 (2.1) <0.0001 0.0493 0.0001 95.9 (15.84–580.30) 6.3 (1.17–34.3) 29.5 (5.30–164.17) 
DR9/X 15 (23.8) 6 (21.4) 13 (16.3) 25 (13.2) <0.0001 NS 0.0002 17.7 (3.77–83.21) 2.0 (0.62–6.63) 10.2 (2.68–39.04) 
DR4/DR9 12 (19.0) 2 (7.1) 11 (13.8) 12 (6.3) <0.0001 NS <0.0001 29.5 (5.83–149.15) 1.4 (0.26–7.61) 18.0 (4.36–74.55) 
DR8/DR8 0 (0.0) 0 (0.0) 1 (1.3) 0 (0.0) — — NS — — — 
DR9/DR8 1 (1.6) 0 (0.0) 2 (2.54) 1 (0.5) NS NS 0.0140 29.5 (1.32–661.10) — 39.3 (2.73–565.76) 
DR4/DR8 3 (4.8) 2 (7.1) 7 (8.8) 2 (1.1) 0.0021 NS <0.0001 44.3 (4.53–431.56) 8.43 (1.02–69.58) 68.8 (9.76–485.39) 
DR8/X 1 (1.6) 1 (3.6) 5 (6.3) 5 (2.6) NS NS 0.0008 5.90 (0.45–76.95) 1.69 (0.17–16.57) 19.7 (3.60–107.417) 
R/Z 2 (3.2) 1 (3.6) 3 (3.8) 63 (33.2) NS NS NS 0.94 (0.13–6.86) 0.13 (0.02–1.12) 0.94 (0.18–4.82) 
X/X 2 (3.2) 7 (25.0) 3 (3.8) 59 (31.1) — — — — — — 

Data are n (%) unless otherwise indicated. DR4, DRB1*0405-DQB1*0401; DR8, DRB1*0802-DQB1*0302; DR9, DRB1*0901-DQB1*0303; R, DRB1*1501-DQB1*0602 or DRB1*1502-DQB1*0601; X, other than DR4, DR8, DR9, or R; Z, any haplotype; AO, acute onset; CO, childhood onset; SO, slow onset.

View Large
TABLE 3

HLA DRB1-DQB1 haplotype frequency with respect to onset age in acute-onset type 1 diabetes

Age of onset (years)
1.0–1820–49≥50
n 160 94 32 
DRB1*0901-DQB1*0303 39 (24.4) 36 (38.3) 18 (56.3) 
DRB1*0405-DQB1*0401 53 (33.1) 27 (28.7) 7 (21.9) 
DR9 frequency/DR4 frequency 0.73 1.33 2.57 
DRB1*0802-DQB1*0302 16 (10.0) 5 (5.3) 0 (0.0) 
Others 52 (32.5) 26 (27.6) 7 (21.9) 
Age of onset (years)
1.0–1820–49≥50
n 160 94 32 
DRB1*0901-DQB1*0303 39 (24.4) 36 (38.3) 18 (56.3) 
DRB1*0405-DQB1*0401 53 (33.1) 27 (28.7) 7 (21.9) 
DR9 frequency/DR4 frequency 0.73 1.33 2.57 
DRB1*0802-DQB1*0302 16 (10.0) 5 (5.3) 0 (0.0) 
Others 52 (32.5) 26 (27.6) 7 (21.9) 

Data are n (%). Acute onset type 1 diabetes includes both adult- and childhood-onset patients who were combined, then divided into three groups by onset age. DR4, DRB1*0405-DQB1*0401; DR9, DRB1*0901-DQB1*0303.

View Large

This work was supported by a Grant-in-Aid for Scientific Research on Priority Areas (C) Medical Genome Science (No. 12204007) and a Grant-in-Aid for Scientific Research (C) (Nos. 14571097, 14571095, and 14571096) from the Ministry of Education, Culture, Science, Sports, and Technology of Japan.

We thank J. Kanatani and Y. Watanabe for collecting clinical data and samples.

1.
Notkins AL, Lernmark Å: Autoimmune type 1 diabetes: resolved and unresolved issues.
J Clin Invest
 
108
:
1247
–1252,
2001
2.
Atkinson MA, Eisenbarth GS: Type 1 diabetes: new perspectives on disease pathogenesis and treatment.
Lancet
 
358
:
221
–229,
2001
3.
Laakso M, Pyörälä K: Age of onset and type of diabetes.
Diabetes Care
 
8
:
114
–117,
1985
4.
Mølbak AG, Christau B, Marner B, Borch-Johnsen K, Nerup J: Incidence of insulin-dependent diabetes mellitus in age groups over 30 years in Denmark.
Diabet Med
 
11
:
650
–655,
1994
5.
Imagawa A, Hanafusa T, Miyagawa J, Matsuzawa Y: A novel subtype of type 1 diabetes mellitus characterized by a rapid onset and an absence of diabetes-related antibodies: Osaka IDDM Study Group.
N Engl J Med
 
342
:
301
–307,
2000
6.
Kobayashi T, Tamemoto K, Nakanishi K, Kato N, Okubo M, Kajio H, Sugimoto T, Murase T, Kosaka K: Immunogenetic and clinical characterization of slowly progressive IDDM.
Diabetes Care
 
16
:
780
–788,
1993
7.
Zimmet PZ, Tuomi T, Mackay IR, Rowley MJ, Knowles W, Cohen M, Lang DA: Latent autoimmune diabetes mellitus in adults (LADA): the role of antibodies to glutamic acid decarboxylase in diagnosis and prediction of insulin dependency.
Diabet Med
 
11
:
299
–303,
1994
8.
Davies JL, Kawaguchi Y, Bennett ST, Copeman JB, Cordell HJ, Pritchard LE, Reed PW, Gough SCL, Jenkins SC, Palmer SM, Balfour KM, Rowe BR, Farrall M, Barnett AH, Bain SC, Todd JA: A genome-wide search for human type 1 diabetes susceptibility genes.
Nature
 
371
:
130
–136,
1994
9.
Tokunaga K, Ohashi J, Bannai M, Juji T: Genetic link between Asians and native Americans: evidence from HLA genes and haplotypes.
Hum Immunol
 
62
:
1001
–1008,
2001
10.
Noble JA, Valdes AM, Cook M, Klitz W, Thomson G, Erlich HA: The role of HLA class II genes in insulin-dependent diabetes mellitus: molecular analysis of 180 Caucasian, multiplex families.
Am J Hum Genet
 
59
:
1134
–1148,
1996
11.
Cucca F, Lampis R, Congia M, Angius E, Nutland S, Bain SC, Barnett AH, Todd JA: A correlation between the relative predisposition of MHC class II alleles to type 1 diabetes and the structure of their proteins.
Hum Mol Genet
 
10
:
2025
–2037,
2001
12.
Hashimoto M, Kinoshita T, Yamasaki M, Tanaka H, Imanishi T, Ihara H, Ichikawa Y, Fukunishi T: Gene frequencies and haplotypic associations within the HLA region in 916 unrelated Japanese individuals.
Tissue Antigens
 
44
:
166
–173,
1994
13.
Awata T, Kuzuya T, Matsuda A, Iwamoto Y, Kanazawa Y: Genetic analysis of HLA class II alleles and susceptibility to type 1 (insulin-dependent) diabetes mellitus in Japanese subjects.
Diabetologia
 
35
:
419
–424,
1992
14.
Yasunaga S, Kimura A, Hamaguchi K, Rønningen KS, Sasazuki T: Different contribution of HLA-DR and -DQ genes in susceptibility and resistance to insulin-dependent diabetes mellitus (IDDM).
Tissue Antigens
 
47
:
37
–48,
1996
15.
Kawabata Y, Ikegami H, Kawaguchi Y, Fujisawa T, Shintani M, Ono M, Nishino M, Uchigata Y, Lee I, Ogihara T: Asian-specific HLA haplotypes reveal heterogeneity of the contribution of HLA-DR and -DQ haplotypes to susceptibility to type 1 diabetes.
Diabetes
 
51
:
545
–551,
2002
16.
Karjalainen J, Salmela P, Ilonen J, Surcel HM, Knip M: A comparison of childhood and adult type I diabetes mellitus.
N Engl J Med
 
320
:
881
–886,
1989
17.
Caillat-Zucman S, Garchon HJ, Timsit J, Assan R, Boitard C, Djilali-Saiah I, Bougneres P, Bach JF: Age-dependent HLA genetic heterogeneity of type 1 insulin-dependent diabetes mellitus.
J Clin Invest
 
90
:
2242
–2250,
1992
18.
Tait BD, Harrison LC, Drummond BP, Stewart V, Varney MD, Honeyman MC: HLA antigens and age at diagnosis of insulin-dependent diabetes mellitus.
Hum Immunol
 
42
:
116
–122,
1995
19.
Dubois-Laforgue D, Timsit J, Djilali-Saiah I, Boitard C, Caillat-Zucman S: Insulin-dependent diabetes mellitus in non-DR3/non-DR4 subjects.
Hum Immunol
 
57
:
104
–109,
1997
20.
Lohmann T, Sessler J, Verlohren HJ, Schroder S, Rötger J, Dähn K, Morgenthaler N, Scherbaum WA: Distinct genetic and immunological features in patients with onset of IDDM before and after age 40.
Diabetes Care
 
20
:
524
–529,
1997
21.
Tuomi T, Carlsson A, Li H, Isomaa B, Miettinen A, Nilsson A, Nissen M, Ehrnstrom BO, Forsen B, Snickars B, Lahti K, Forsblom C, Saloranta C, Taskinen MR, Groop LC: Clinical and genetic characteristics of type 2 diabetes with and without GAD antibodies.
Diabetes
 
48
:
150
–157,
1999
22.
Tanaka S, Kobayashi T, Nakanishi K, Koyama R, Okubo M, Murase T, Odawara M, Inoko H: Association of HLA-DQ genotype in autoantibody-negative and rapid-onset type 1 diabetes.
Diabetes Care
 
25
:
2302
–2307,
2002
23.
Shimizu I, Makino H, Osawa H, Kounoue E, Imagawa A, Hanafusa T, Kawasaki E, Fujii Y: Association of fulminant type 1 diabetes with pregnancy.
Diabetes Res Clin Pract
 
62
:
33
–38,
2003
24.
Expert Committee on the Diagnosis and Classification of Diabetes Mellitus: Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus.
Diabetes Care
 
20
:
1183
–1197,
1997
25.
Alberti KG, Zimmet PZ: Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1. Diagnosis and classification of diabetes mellitus: provisional report of a WHO consultation.
Diabet Med
 
15
:
539
–553,
1998
26.
Takeda H, Kawasaki E, Shimizu I, Konoue E, Fujiyama M, Murao S, Tanaka K, Mori K, Tarumi Y, Seto I, Fujii Y, Kato K, Kondo S, Takada Y, Kitsuki N, Kaino Y, Kida K, Hashimoto N, Yamane Y, Yamawaki T, Onuma H, Nishimiya T, Osawa H, Saito Y, Makino H: Clinical, autoimmune, and genetic characteristics of adult-onset diabetic patients with GAD autoantibodies in Japan (Ehime Study).
Diabetes Care
 
25
:
995
–1001,
2002
27.
Kida K, Mimura G, Kobayashi T, Nakamura K, Sonoda S, Inouye H, Tsuji K: Immunogenetic heterogeneity in type 1 (insulin-dependent) diabetes among Japanese: HLA antigens and organ-specific autoantibodies.
Diabetologia
 
32
:
34
–39,
1989
28.
Kida K, Mimura G, Ito T, Murakami K, Ashkenazi I, Laron Z: Incidence of type 1 diabetes mellitus in children aged 0–14 in Japan,
1986
–1990, including an analysis for seasonality of onset and month of birth: JDS study: the Data Committee for Childhood Diabetes of the Japan Diabetes Society (JDS).
Diabet Med
 
17
:
59
–63,
2000
29.
Matsuda A, Kamata I, Iwamoto Y, Sakamoto Y, Kuzuya T: A comparison of serum C-peptide response to intravenous glucagon, and urine C-peptide, as indexes of insulin dependence.
Diabetes Res Clin Pract
 
1
:
161
–167,
1985
30.
Abiru N, Takino H, Yano M, Kawasaki E, Yamasaki H, Yamaguchi Y, Akazawa S, Nagataki S: Clinical evaluation of non-insulin-dependent diabetes mellitus patients with autoantibodies to glutamic acid decarboxylase.
J Autoimmun
 
9
:
683
–688,
1996
31.
Sera Y, Kawasaki E, Abiru N, Ozaki M, Abe T, Takino H, Kondo H, Yamasaki H, Yamaguchi Y, Akazawa S, Nagataki S, Uchigata Y, Matsuura N, Eguchi K: Autoantibodies to multiple islet autoantigens in patients with abrupt onset type 1 diabetes and diabetes diagnosed with urinary glucose screening.
J Autoimmun
 
13
:
257
–265,
1999
32.
Horton V, Stratton I, Bottazzo GF, Shattock M, Mackay I, Zimmet P, Manley S, Holman R, Turner R: Genetic heterogeneity of autoimmune diabetes: age of presentation in adults is influenced by HLA DRB1 and DQB1 genotypes (UKPDS 43). UK Prospective Diabetes Study (UKPDS) Group.
Diabetologia
 
42
:
608
–616,
1999
33.
Payami H, Joe S, Farid NR, Stenszky V, Chan SH, Yeo PP, Cheah JS, Thomson G: Relative predispositional effects (RPEs) of marker alleles with disease: HLA-DR alleles and Graves disease.
Am J Hum Genet
 
45
:
541
–546,
1989
34.
Imagawa A, Hanafusa T, Uchigata Y, Kanatsuka A, Kawasaki E, Kobayashi T, Shimada A, Shimizu I, Toyoda T, Maruyama T, Makino H: Fulminant type 1 diabetes: a nationwide survey in Japan.
Diabetes Care
 
26
:
2345
–2352,
2003
35.
Baisch JM, Weeks T, Giles R, Hoover M, Stastny P, Capra JD: Analysis of HLA-DQ genotypes and susceptibility in insulin-dependent diabetes mellitus.
N Engl J Med
 
322
:
1836
–1841,
1990
36.
Khalil I, Deschamps I, Lepage V, Al-Daccak R, Degos L, Hors J: Dose effect of cis- and trans-encoded HLA-DQαβ heterodimers in IDDM susceptibility.
Diabetes
 
41
:
378
–384,
1992
37.
Rønningen KS, Spurkland A, Iwe T, Vartdal F, Thorsby E: Distribution of HLA-DRB1, -DQA1 and -DQB1 alleles and DQA1-DQB1 genotypes among Norwegian patients with insulin-dependent diabetes mellitus.
Tissue Antigens
 
37
:
105
–111,
1991
38.
Ludvigsson J, Samuelsson U, Beauforts C, Deschamps I, Dorchy H, Drash A, Francois R, Herz G, New M, Schober E: HLA-DR 3 is associated with a more slowly progressive form of type 1 (insulin-dependent) diabetes.
Diabetologia
 
29
:
207
–210,
1986
39.
Nishimura Y, Oiso M, Fujisao S, Kanai T, Kira J, Chen YZ, Matsushita S: Peptide-based molecular analyses of HLA class II-associated susceptibility to autoimmune diseases.
Int Rev Immunol
 
17
:
229
–262,
1998
40.
Harding CV, Unanue ER: Quantitation of antigen-presenting cell MHC class II/peptide complexes necessary for T-cell stimulation.
Nature
 
346
:
574
–576,
1990
41.
Kawabata Y, Ikegami H, Kawaguchi Y, Fujisawa T, Hotta M, Ueda H, Shintani M, Nojima K, Ono M, Nishino M, Taniguchi H, Noso S, Yamada K, Babaya N, Ogihara T: Age-related association of MHC class I chain-related gene A (MICA) with type 1 (insulin-dependent) diabetes mellitus.
Hum Immunol
 
61
:
624
–629,
2000
42.
Sugihara S, Sakamaki T, Konda S, Murata A, Wataki K, Kobayashi Y, Minamitani K, Miyamoto S, Sasaki N, Niimi H: Association of HLA-DR, DQ genotype with different beta-cell functions at IDDM diagnosis in Japanese children.
Diabetes
 
46
:
1893
–1897,
1997
DIABETES
2004