A prospective study of leukocyte telomere length and risk of type 2 diabetes in postmenopausal women

Abstract

Telomere length (TL) has been implicated in the pathogenesis of age-related disorders. However, there are no prospective studies directly investigating the role of TL and relevant genes in diabetes development. In the multiethnic Women's Health Initiative, we identified 1,675 incident diabetes case participants in 6 years of follow-up and 2,382 control participants matched by age, ethnicity, clinical center, time of blood draw, and follow-up duration. Leukocyte TL at baseline was measured using quantitative PCR, and Mendelian randomization analysis was conducted to test whether TL is causally associated with diabetes risk. After adjustment for matching and known diabetes risk factors, odds ratios per 1-kilobase increment were 1.00 (95% CI 0.90-1.11) in whites, 0.95 (0.85-1.06) in blacks, 0.96 (0.79-1.17) in Hispanics, and 0.88 (0.70-1.10) in Asians. Of the 80 single nucleotide polymorphisms (SNPs) in nine genes involved in telomere regulation, 14 SNPs were predictive of TL, but none were significantly associated with diabetes risk. Using ethnicity-specific SNPs as randomization instruments, we observed no statistically significant association between TL and diabetes risk (P = 0.52). Although leukocyte TL was weakly associated with diabetes risk, this association was not independent of known risk factors. These prospective findings indicate limited clinical utility of TL in diabetes risk stratification among postmenopausal women.

FIG. 1.

TL (kb) by clinical diabetes status and race/ethnicity. The boxplots show the mean (+), median, and interquartile range. The vertical line indicates the minimum and maximum value of TL in each group. (A high-quality color representation of this figure is available in the online issue.)

FIG. 2.

TL by GRS according to race/ethnicity. The boxplots show the mean (◇), median, and interquartile range. The vertical line indicates the minimum and maximum value >1.5 times the upper quartile of TL in each group. Outliers (○) are displayed when the individual TL is >1.5 times the interquartile range of the upper quartile value. GRS represents the unweighted sum of alleles from the selected SNPs. The selected SNPs for each of the ethnic groups are as follows: 1) whites, rs34368910, rs4888444, rs4975605, rs938886, and rs12880583; 2) blacks, rs872072, rs938886, rs1713458, and rs4387287; 3) Hispanics, rs35276863, rs729421, rs11972248, rs4635969, rs2853669, rs2736098, and rs2853676; and 4) Asians, rs11556640 and rs2297613. (A high-quality color representation of this figure is available in the online issue.)