Aims/hypothesis
We used recently confirmed type 2 diabetes gene regions to investigate the genetic relationship between type 1 and type 2
diabetes, in an average of 7,606 type 1 diabetic individuals and 8,218 controls, providing >80% power to detect effects as
small as an OR of 1.11 at a false-positive rate of 0.003.
Methods
The single nucleotide polymorphisms (SNPs) with the most convincing evidence of association in 12 type 2 diabetes-associated
gene regions, PPARG, CDKAL1, HNF1B, WFS1, SLC30A8, CDKN2A–CDKN2B, IGF2BP2, KCNJ11, TCF7L2, FTO, HHEX–IDE and THADA, were analysed in type 1 diabetes cases and controls. PPARG and HHEX–IDE were additionally tested for association in 3,851 type 1 diabetes families. Tests for interaction with HLA class II genotypes,
autoantibody status, sex, and age-at-diagnosis of type 1 diabetes were performed with all 12 gene regions.
Results
Only PPARG and HHEX–IDE showed any evidence of association with type 1 diabetes cases and controls (p = 0.004 and p = 0.003, respectively; p > 0.05 for other SNPs). The potential association of PPARG was supported by family analyses (p = 0.003; p
combined = 1.0 × 10−4). No SNPs showed evidence of interaction with any covariate (p > 0.05).
Conclusions/interpretation
We found no convincing genetic link between type 1 and type 2 diabetes. An association of PPARG (rs1801282/Pro12Ala) could be consistent with its known function in inflammation. Hence, our results reinforce evidence suggesting
that type 1 diabetes is a disease of the immune system, rather than being due to inherited defects in beta cell function or
regeneration or insulin resistance.
Keywords Age-at-diagnosis - Association study - Autoantibodies - Genetics -
PPARG
-
SLC30A8
- Type 1 diabetes - Type 2 diabetes