Associations of middle-aged mother's but not father's body mass index with 18-year-old son's waist circumferences, birth weight, and serum hepatic enzyme levels

Abstract

Mitochondrial dysfunction has been reported to contribute to insulin resistance (IR) in the elderly and type 2 diabetes. To test this hypothesis, we examined relations of insulin resistance in young men to their mother's body mass index (BMI) and compared with those to their father's BMI, because as a rule, mitochondrial DNA is exclusively maternally inherited and because mitochondria are fundamental in mediating effects on energy dissipation. We measured heights, weights, waist circumference, systolic and diastolic blood pressure (BP), and biochemical variables in sera from 193 male college students aged 18 to 20 years after an overnight fast. Birth weight was available from 184 students. Self-reported heights and weights of their parents were obtained from 148 students. Insulin resistance and insulin secretion were estimated using homeostasis model assessment (HOMA-IR and HOMA- β, respectively). Mother's BMI was associated with their son's birth weight ( r = 0.23, P = .008), BMI ( r = 0.37, P < .0001), waist circumference ( r = 0.42, P < .0001), fasting insulin ( r = 0.19, P = .02), and HOMA-IR ( r = 0.18, P = .03) but not with fasting glucose, HOMA- β, and systolic and diastolic BP. In addition, high-density lipoprotein cholesterol and lipoprotein(a) [Lp(a)] were inversely associated with mother's BMI ( r = −0.21, P = .01 and r = −0.17, P = .03, respectively). Furthermore, there were significant associations with aspartate ( r = 0.20, P = .01) and alanine ( r = 0.28, P = .0008) aminotransferase and γ-glutamyl transpeptidase ( r = 0.30, P = .0003), all of which are associated with mitochondrial function. In contrast, none of those variables were associated with father's BMI, except for Lp(a), which showed a significant and inverse association ( r = −0.17, P = .05). After adjustment for sons' BMI, waist circumference and 3 hepatic enzymes were associated with mother's BMI, whereas Lp(a) was associated with both mother's and father's BMI. In multiple regression analysis for HOMA-IR as a dependent variable, BMI of their own ( β = .10, P < .0001) and of their mothers ( β = .04, P =.10) and birth weight ( β = −.27, P = .10) emerged as determinants of HOMA-IR of the students( R 2 = 0.30). Our results are consistent with clinical observations of a greater risk of transmission of type 2 diabetes from the mother than the father and suggest that son's IR may be influenced by maternal effect as well as their adiposity.