Recent advances in the genetic epidemiology of Type 1, Type 2 and MODY may soon make it possible to distinguish them based on genetic markers. Inheritance in European-origin MODY families usually follows that of an autosomal dominant pattern, with vertical transmission of disease from one generation to the next, and approximately 50% of siblings affected (28). Reports of non-Type 1 in other ethnic groups show vertical transmission in only a subset of such families (4). Work in Europeans and US whites has identified several associations of MODY with specific mutations (29); the HLA-DR and -DQ alleles linked to Type 1 diabetes are not found in MODY patients (27, 65). Indeed, some investigators suggest that patients with early-onset Type 2 diabetes can be distinguished from MODY patients on the basis of inheritance patterns (66).
There is widespread consensus that Type 2 diabetes, in contrast to MODY, is a genetically heterogeneous condition (45, 67). Familial aggregation is common, leading numerous investigators to search for relevant genes with limited success.
An array of polymorphisms have been linked to Type 2 diabetes and obesity in single families or ethnic groups, but no generalizable associations have yet emerged (65, 68, 69).
Among African-origin and Hispanic patients with 'typical' Type 1 diabetes, immunogenetic similarities to European-origin Type 1 diabetes patients have been demonstrated with a few notable exceptions. The strong association of Type 1 diabetes with alleles at the HLA-DR and -DQ loci has been confirmed among both African-origin patients and Latinos (40,70). In a case-control study from Colorado, Mexican American children with Type 1 diabetes were more likely than non-diabetic controls to have HLA-DR3, -DR4, and -DQ alleles which code for an amino acid other than aspartate at position 57 of the B1 subunit (non-Asp-57); similar findings were reported among non-Hispanic whites (71). The associated odds ratios were not significantly different by ethnicity, although fewer Mexican American patients carried DR3. Among blacks, similar strong associations with HLA-DR and -DQ alleles have been reported. Todd et al. (72) reported that the diabetes-associated DR9 haplo-type among Afro-Caribbean Type 1 diabetes patients in England carried a non-Asp codon at position 57 of DQB1, in contrast to DR9 among whites, which appears to be protective against IDDM and codes for Asp at dQb1-57 (73).
A positive family history is found in only 10-15% of Type 1 diabetes patients, while among Type 2 and MODY patients many relatives are affected. The frequency of diabetes among relatives of 9 atypical patients from Winter's clinic series was significantly higher than among relatives of 'typical' black (n = 5) and white (n = 11) Type 1 diabetes patients (74). Diabetes was ascertained in these relatives by patient or family report, and thus may also reflect differences in health care-seeking behaviors or other environmental risks.
A study of African American women reported that HLA-B41 and -DR2 were positively associated with risk of insulin-requiring gestational diabetes mellitus, and with risk of developing Type 2 diabetes in those with previous gestational diabetes (75). This report is interesting in light of the findings that older, ketosis-prone blacks with 'Flatbush' diabetes described by Banerji et al. showed a higher than expected frequency of the Type 1-associated HLA alleles (47). Evidently the strong association of HLA class II alleles with diabetes is not yet fully understood.
Was this article helpful?