Genetic and non-genetic influences on some metabolic cardiovascular risk factors : quantitative genetic analyses

Abstract

Structural-equation modeling analyses were applied to data from 302 Swedish twin pairs in order to assess the relative importance of genetic and nongenetic influences on several metabolic cardiovascular risk factors. These middle-aged and elderly twins include monozygotic and dizygotic twins reared apart and together, who came from a population-based national twin register.

Genetic factors were found to play a significant role for all study risk factors. The relative importance of genetic effects varied from 34% for diastolic blood pressure, 36% for IGFBP-l, 39% for insulin resistance index, 42% for PAI-l, 44% for systolic blood pressure, 48% for insulin, 63% for IGF-I, to 94% for Lp(a). Shared rearing environmental influences were also evident for DBP and PAI-l. Residual familial environmental factors affected variation in the levels of SBP, Lp(a), and PAI I to some degree. Interestingly, all five principal metabolic components contained in the insulin-resistance syndrome (insulin resistance, body mass index, triglycerides, HDL cholesterol, and systolic blood pressure) were more or less influenced by a single latent genetic factor, whereas only three of the components (triglycerides, insulin resistance, and HDL cholesterol) were influenced by a latent individual specific environmental factor. The genetic factor reflected influences of major importance to BMI and insulin resistance and to a lesser degree to triglycerides, whereas the environmental factors reflected influences in common to triglycerides and HDL cholesterol and to a lesser degree to insulin resistance.

Systolic blood pressure was related to the insulin resistance syndrome, albeit weakly, only through genetic effects. In addition, approximately two-thirds of the phenotypic correlation of -0.50 between IGFBP-l and insulin was attributable to an environmental component, whereas phenotypic associations between IGFBP-l and IGF-I, and between IGF-I and insulin could be entirely explained by environmental factors. Furthermore, a substantial genetic correlation between PAI-l and triglycerides and a moderate genetic correlation between PAI-l and BMI are suggested. In co-twin analyses of monozygotic twins, sex hormone administration was a significant nongenetic predictor for serum Lp(a) levels in women, whereas the influence of alcohol consumption, B-blocker and diuretic administration on Lp(a) appeared not to be significant after controlling for genetic influences.

In conclusion, genetic influences are substantial for Lp(a) and IGF-I, and moderate for insulin, insulin resistance, SBP, DBP, IGFBP-l, and PAI-l. Interestingly, a set of genes in common to all five principal components of IRS (insulin resistance, BMI, triglycerides, HDL-cholesterol, and SBP) are important for covariation among these measures, and may initiate the occurrence of the syndrome. Genetic factors play important roles in the associations between PAI-I and triglycerides and BMI.