Erythrocyte membrane phospholipid composition is related to hyperinsulinemia in obese nondiabetic women: Effects of weight loss.

摘要

The complex mechanisms by which obesity predisposes to insulin resistance are not clearly understood. According to a cell membrane hypothesis of insulin resistance, the defects in insulin action could be related to changes in membrane properties. The purpose of this work was to examine the relationship between 2 markers of insulin resistance (fasting plasma insulin [FPI] and homeostasis model assessment [HOMA IR]) and erythrocyte membrane lipid composition. In the first cross-sectional study, 24 premenopausal nondiabetic overweight women (body mass index [BMI], 32.5 +/- 0.9 kg/m(2); age, 35.7 +/- 2.2 years) were compared to 21 lean healthy women (BMI, 21 +/- 0.4 kg/m(2); age, 35.4 +/- 2.2 years). The second study examined whether a 3-month diet-induced weight loss, which usually improves insulin resistance, could also affect the membrane phospholipid (PL) composition and fluidity in the overweight group. Overweight women had significantly higher FPI levels (P <.0001), HOMA IR (P <.0001), membrane sphingomyelin (SM) (P <.05), and cholesterol (P <.05) contents than lean women. Baseline FPI and HOMA IR were positively correlated with membrane SM (P <.005), phosphatidylethanolamine (PE) (P <.005), and phosphatidylcholine (PC) (P <.05) contents, and negatively with phosphatidylinositol (PI) (P <.05) contents in the whole population. Multivariate regression analyses showed that 2 membrane parameters, PE and SM, were among the independent predictors of FPI or HOMA IR in the whole population, but also in the lean and the obese groups separately. Intervention induced a significant reduction in body weight (-5.7% +/- 0.7%), fat mass (-11.3% +/- 1.4%), and FPI (-10.2% +/- 5.4%). An improvement in membrane lipid composition was only observed in the insulin resistant subgroup (FPI > 9.55 mU/L). The reduction in FPI or HOMA IR was directly associated with reduction in SM and PE contents, a finding independent of the reduction in fat mass. A stepwise multiple regression analysis indicated that the changes in SM accounted for 26.6% of the variance in the changes in FPI as an independent predictor, with the changes in fat mass and PE as other determinants (27.8% and 20%, respectively, adjusted r(2) =.704, P <.0001). These results suggest that the abnormalities in the membrane PL composition could be included in the unfavorable lipid constellation of obesity which correlated with impaired insulin sensitivity.