IMMUNOGENETICS OF TYPE 1 DIABETES IN MAN

Abstract

Genetic susceptibility to type 1 diabetes is associated with the major histocompatibility complex (MHC) on human chromosome 6. To analyse the contribution of the class I region of the MHC to disease susceptibility, polymorphism of the class Ib molecule HLA-E and several genetic markers within the region was investigated in 378 British Caucasoid patients with type 1 diabetes as defined by the National Diabetes Group, 100 multiplex families from the BDA Warren Repository and 216 unselected normal British Caucasoid controls. PCR-Sequence-Specific Oligonucleotide Probing (PCR-SSOP) was used to analyse polymorphism of the HLA-E gene, Restriction Fragment Length Polymorphism analysis was used to investigate the four genomic probes P3A, P3B, 28L and 66R and finally microsatellite analysis to examine the two markers P 1 and D6S306. The association of the class I region with type 1 diabetes was also investigated with respect to gender and age at onset of type 1 diabetes. Linkage with known MHC class II susceptibility alleles was also analysed and the transmission disequilibrium test was used to analyse the BDA multiplex families. Significant increases in the frequency of the 1.5; 1.5 kilo base (kb) (Pc = 0.0004) and 1.8; 1.5 kb (Pc= 0.008) P3B genotypes were found in the patients compared to the controls. The P3B genotype 1.8; 1.8 was found to be decreased in the patients compared to the controls 26.6% vs 51.5% (Pc = 0.000002). This decrease in the 1.8; 1.8 P3B genotype was particularly apparent in those patients diagnosed before the age of 20 years. There were also differences in the frequency of the P3B genotype between males and female patients. The 1.5 P3B allele was increased and the 1.8 kb allele decreased in patients ( 46.1% and 53.9%) respectively compared to normal controls (27.8% and 72.1 %) respectively (Pc = <0.000001). The 1.8 P3B allele was increased in the <10 years compared to the 10- 20 years age group 60.2% vs 49.1% (Pc = 0.06). Further, the 1.8kb allele was increased in female patients with an age at onset of < 10 years compared to those in the 10-20 years group 63.5% vs 45.8% (Pc = 0.04). Analysis of the BDA multiplex families using the transmission disequilibirum test (TDT) showed that the 1.5kb allele was significantly transmitted from parents to affected offspring (Pc = 0.000002), suggesting that the 1.5 allele is in linkage with a susceptibility locus for type 1 diabetes. There was a highly significant difference in the observed transmission of the 1.5kb allele in the > 10 years group compared to the observed transmission of the 1.8kb allele in the > 10 years group (Pc = 0.000002). Increased frequencies of the HLA-E genotypes 0101;0101 (Pc = 0.0007), 0101;0103 (Pc = 0.025) were found in patients compared to controls. The HLA-E genotypes 0102;0104 (Pc= 0.001) and 0103;0104 (Pc= 0.004) were decreased in patients compared to controls. The 0101;0 101 genotype was increased in patients with an age at onset of 10-20 years compared to patients in the < 10 years group 4 7.0% vs 21 .1% (Pc = 0.001). The HLA-E 0101 allele was increased in patients compared to controls 54.1% vs 28.7% (Pc = 0.000003). In contrast, the alleles 0102 and 0104 were decreased in patients compared to controls (Pc = 0.000003, p = 0.0000 1) respectively. The 0 I 03 allele was increased in patients with an age at onset of < 10 years compared to the 10-20 years group 34.1% vs 18.9% (Pc = 0.02) in contrast, the 0101 allele was increased in the 10-20 years compared to the <10 years 62.9% vs 47.6% (Pc = 0.07). The frequency of the 0103 allele was increased in female patients in the < 10 years compared to the 10-20 years 37.5% vs 16.2% (Pc = 0.04). Haplotype analysis showed the significant presence of the P3A-P3B-28L-HLA-E (4-1.5-3 .8-0101) haplotype in patients with type 1 diabetes, which suggested the possible association between the P3B 1.5kb allele and the HLA-E 0101 allele. This was also supported by the presence of the 3 loci haplotype P3B-28L-HLA-E (1.5-3 .8-0101) that was found significantly present in patients. Combined genotype analysis of the P3B and HLA-E showed that the 0101-0103-1.5 genotype was significantly increased in patients compared to controls (Pc = 0.08). Analysis of the class lI 'diabetes-associated' alleles showed there was no significant linkage/association with either the P3B or HLA-E locus, suggesting that the class I association with type 1 diabetes may be independent of the class II. In conclusion a strong association with type 1 diabetes has been identified within the class I region, localised to the central area of the region. The association is related to gender and age at onset of type 1 diabetes, particularly the 10-20 years age at onset group. Identified HLA-E as a novel susceptibility gene, which is the first report of an association between a class Ib gene and an autoirnmune disease.