¹Steno Diabetes Center, DK-2820 Gentofte, Denmark

²Department of Diabetes and Metabolic Medicine, Barts and the London, Queen Mary School of Medicine and Dentistry, University of London, UK

Correspondence to: Dr F Pociot, Steno Diabetes Center, Niels Steensensvej 2, DK-2820 Gentofte, Denmark. E-mail: fpoc@novo.dk

This work was in part supported by the EU BioMed 2 Programme (grant no. BMH4-CT97-2311), Novo Nordisk A/S, The Danish Diabetes Association, and the DANDY Foundation. Support from the Juvenile Diabetes Foundation International is also acknowledged.

At least 20 different chromosomal regions have been linked to type 1 diabetes (T1D) susceptibility in humans, using genome screening, candidate gene testing, and studies of human homologues of mouse susceptibility genes. The largest contribution from a single locus (IDDM1) comes from several genes located in the MHC complex on chromosome 6p21.3, accounting for at least 40% of the familial aggregation of this disease. Approximately 30% of T1D patients are heterozygous for HLA-DQA1*0501-DQB1*0201/DQA1*0301-DQB1*0302 alleles (formerly referred to as HLA-DR3/4 and for simplification usually shortened to HLA-DQ2/DQ8), and a particular HLA-DQ6 molecule (HLA-DQA1*0102-DQB1*0602) is associated with dominant protection from the disease. There is evidence that certain residues important for structure and function of both HLA-DQ and DR peptide-binding pockets determine disease susceptibility and resistance. Independent confirmation of the IDDM2 locus on chromosome 11p15.5 has been achieved in both case-control and family-based studies, whereas associations with the other potential IDDM loci have not always been replicated. Several possibilities to explain these variable results from different studies are discussed, and a key factor affecting both linkage and association studies is that the genetic basis of T1D susceptibility may differ between ethnic groups. Some future strategies to address these problems are proposed. These include increasing the sample size in homogenous ethnic groups, high throughput genotyping and genomewide linkage disequilibrium (LD) mapping to establish disease associated ancestral haplotypes. Elucidation of the function of particular genes ('functional genomics') in the pathogenesis of T1D will be a most important element in future studies in this field, in addition to more sophisticated methods of statistical analyses.

Genes and Immunity (2002) 3, 235-249. doi:10.1038/sj.gene.6363875

type 1 diabetes mellitus; T1DM; genetic susceptibility; human leukocyte antigen; genome scan