diabetes risk, the genotype of an individual (namely all of the alleles on an individual's two sixth chromosomes). For example, an individual with the two DQ haplotypes, DQA1*0501, DQB1*0201 (DR3 associated) and DQA1*0301, DQB1*0302 (DR4 associated) have the highest risk for type 1 diabetes. As each allele codes for a specific molecule and all the combinations of these molecules are possible, such individuals will have four different DQ molecules (DQA*0501/DQB1*0201, DQA1*0301 DQB1*0302) on the surface of their antigen-presenting cells, as expected, but also DQA1*0501/ DQB1*0302 and DQA1*0301/DQB1*0201. Individuals of African descent have a hap-lotype with DQA1*0301/DQB 1*0201 on the same chromosome and, as expected, it confers high risk. In addition to risk, some human leukocyte antigen (HLA) DQ molecules are associated with protection from type 1A diabetes. The DQA1*0102/ DQB1*0602 haplotype is present in approx 20% of most populations but in less than 1% of children developing type 1A diabetes (34). Protection appears to be dominant over susceptibility (35). Patients with DQA1*0102/DQB 1*0602 and diabetes may have unusual variants of diabetes (e.g., type 2 diabetes in children) or may be currently unexplained exceptions. Another remarkably protective DQ haplotype is DQA1*0201/ DQB1*0303, but this haplotype is relatively uncommon.
DR alleles also influence diabetes risk. DRB1*1401 is reported to provide dominant protection similar to DQA1*0102/DQB1*0602, although, again, it is relatively uncommon (36). The DR4 allele, DRB 1*0403, when associated with DQA1*0301/ DQB1*0302 decreases the risk associated with the DQB1*0302. There is no simple "rule" for diabetes risk, and Table 1 summarizes a series of haplotype-associated risks. The observation that aspartic acid at position 57 of the DQ molecule is associated with protection from type 1A diabetes has too many exceptions to be useful (e.g., DQA1*0401/DQB1*0402 and DQA1*0301/DQB1*0401 are both high risk).
At present, the genetic risk associated with HLA alleles is primarily utilized in research studies. A few centers in Europe, the United States, and Canada have begun epidemiologic studies in which thousands of children are HLA typed at birth, the risk of diabetes (and often celiac disease) is determined, and individuals are followed for the development of anti-islet autoantibodies and diabetes (37,38). A child with the highest-risk HLA genotype for type 1 diabetes from the general population has a risk of approx 5% of developing diabetes and will comprise approx 40% of all children developing type 1A diabetes. In contrast, a child with the same HLA DR and DQ alleles who is the offspring of a patient with type 1A diabetes has a risk of approx 20% (39), and a sibling has a risk exceeding 40% of activating anti-islet autoimmunity. This extremely high risk for relatives explains the participation of relatives in most of the trials for the prevention of diabetes.
Alleles of genes within the MHC appear to account for approximately half of the familial aggregation of type 1A diabetes. It is likely that such DR and DQ alleles direct autoimmunity to certain target organs. For instance, although DQA1*0102/ DQB1*0602 protects from type 1A diabetes, it is associated with high risk for multiple sclerosis. A major effort is underway to identify additional genetic risk factors. It is clear that variation of a nucleotide tandem repeat sequence 5' of the insulin gene contributes to diabetes risk (40-42). The group of alleles with the largest number of repeats provides protection from diabetes, and these alleles appear to account for a little less than 10% of the familial aggregation in Western countries. Of note, the tandem repeat associated with diabetes protection is associated with greater thymic insulin messenger RNA (43).
Was this article helpful?