Objective: In search of functional polymorphisms associated with the genetics of insulin resistance, we studied a variant in the promoter of PIK3CB, the gene coding for the catalytic p110beta subunit of phosphatidylinositol (PI) 3-kinase, a major effector of insulin action.
Research design and methods: The rs361072 C/T variant was selected among single nucleotide polymorphisms of the PIK3CB region because we suspected that its common C allele (allelic frequency approximately 50% in Europeans) could create a GATA-binding motif and was genotyped in five cohorts of obese (n = 1,876) and two cohorts of nonobese (n = 1,490) European children. To estimate insulin resistance in these children, the homeostasis model assessment for insulin resistance (HOMA-IR) index was measured in strict nutritional conditions. GATA-binding and functional effects of rs361072 were explored in transfected cell lines and in lymphocytes from obese children.
Results: The rs361072 C/T variant was associated with HOMA-IR in the obese children cohorts (1.7 x 10(-12) < P < 2 x 10(-4) for C/C vs. T/T using regression analysis). HOMA-IR averaged 3.3 +/- 0.1 in C/C and 4.5 +/- 0.2 in T/T obese children (P = 4.5 x 10(-6) by ANOVA). C/T patients had intermediate values. As shown by the interaction between BMI and genotype (P = 2.1 x 10(-9)), the association of rs361072 with HOMA-IR depended on BMI and was only marginal in nonobese children (P = 0.04). At the molecular level, the C allele of rs361072 was found to create a GATA-binding site able to increase transcription of PIK3CB.
Conclusions: We postulate that the C allele of rs361072 is a causal variant capable of attenuating insulin resistance in obese children through increased expression of p110beta.