Objective: Tumor necrosis factor (TNF ) locus has been a long-standing type 2 diabetes (T2D) candidate gene. Few studies have been conducted on TNF SNP (single nucleotide polymorphism) as rs1799964 (T-1031C), rs1800630 (A-863C) and rs1799724 (C-857T) in T2D. The purpose of this study is to examine the association of TNF SNP and T2D in a case control study and further explore whether these SNPs influence the clinical efficacy of insulin therapy.
Methods: A total of 109 newly diagnosed type 2 diabetics and 168 healthy individuals were recruited. Three tag SNPs (rs1799964 (T-1031C), rs1800630 (A-863C), rs1799724 (C-857T)) were selected across the TNF locus and polymerase chain reaction (PCR) directed sequencing was performed. The patients received Lispro 25 twice daily to achieve glycemic control and they were followed up for 1 year. Plasma glucose level, lipid profile, homeostatic model assessment for insulin resistance (HOMA-IR) and homeostatic model assessment for beta-cell function (HOMA-β) were compared among groups with different haplotypes of SNPs.
Results: Haplotype of TNF-1031C-863C-857C increased the risk of T2D (OR = 2.7, P < 0.05) . Comparing with homozygote of TNF-1031T-863C-857C diabetics (TCC), those carrying CCC allele had higher fasting serum insulin (16.1(12.0-20.3) mU/L) and HOMA-IR (lnHOMA-IR 1.8 ± 0.4) levels (TCC group: 10.6(8.1-14.3) mU/L and 1.42 ± 0.54 respectively, P < 0.05)). One-year insulin treatment decreased HbA1c effectively in both TCC and CCC groups (P < 0.05). However, higher HOMA-IR was still observed in CCC group than that of TCC after normoglycemia (lnHOMA-IR: 2.5(0.9-3.9) vs 1.1(0.8-1.8) respectively, P < 0.05) . Moreover HOMA-β showed no significant improvement in CCC group as it was in TCC group by the endpoint of follow-up.
Conclusions: TNF-1031C-863C-857C is a risk haplotype for T2D. CCC carrying patients failed to achieve HOMA-β improvement. And it might be due to increased endogenous HbOMA-IR level comparing with TCC homozygote.