DQA1 and DQB1 heterodimers in insulin-dependent diabetes mellitus: a genetic-epidemiological study in Finland. DiMe Study Group.

Abstract

Susceptibility to insulin-dependent diabetes mellitus (IDDM) correlates with the absence of aspartic acid in position 57 of the DQB1 and/or the presence of arginine in position 52 of the DQA1. It has been postulated that transcomplementation between the DQ alpha and beta chains of the two haplotypes could create new molecules conferring susceptibility to IDDM. Finland has the highest incidence of IDDM in the world (35/100,000). In a nationwide study of IDDM in childhood (DiMe study) HLA genotyping using conventional serology was carried out according to genetic-epidemiological principles. We simulated DQA1 and DQB1 alleles in 707 consecutively diagnosed IDDM probands and 98 non-diabetic children based on serology, restriction fragment length polymorphism results and sequence data assuming no recombination between DQ and DR. In 34% of Finnish children with IDDM all four combinations (two in cis and two in trans) could lead to SS heterodimers. Two-thirds of these combinations were explained by DR3,DR4 heterozygotes. In 50% of IDDM children half and in 11% a quarter of the combinations could lead to heterodimers. In 38 IDDM patients (5%) the formation of hybrid molecules was not possible. In 59% of the controls SS heterodimers were possible and should therefore have an underlying genetic susceptible for IDDM assuming the theory of transcomplementation is correct. These findings, together with the fact that the lowest frequency of DR3,DR4 heterozygosity (21%) was seen in Finland, show that heterozygosity for DQ and DR cannot explain the differences seen in IDDM incidence.