Plasma oestrone fatty acid ester levels are correlated with body fat mass in humans

OBJECTIVE The metabolites of steroidal hormones, including sulphate, glucur onide, and fatty acid (FA) ester derivatives, have received little attentio n, although these steroid derivatives are essential components in the globa l assessment of steroid metabolism. The study of FA-derivatives could, in o besity, contribute some insights into factors modulating steroid metabolism and their plasma levels. In a recent study we found that, in rats, an oest rone-fatty acid ester (E-1-FA) was produced by white adipose tissue and rel eased into lipoproteins in the bloodstream. We have examined whether E-1-FA levels correlate with body fat and insulin sensitivity in humans. SUBJECTS A sample of 20 men and 22 women with varying levels of total body fat (mean body mass index (BMI) 29.2 +/- 4.7, range 22.2-35.8 in men; mean BMI 27.6 +/- 6.3, range 16.8-37.9 in women). All participants were healthy. MEASUREMENTS We measured oestrone fatty acid esters (E-1-FA), body fatness, and body fat distribution variables, as well as insulin sensitivity throug h a frequently sampled intravenous glucose tolerance test. Plasma E-1-FA an d serum leptin levels were measured by radioimmunoassay. RESULTS E-1-FA levels strongly correlated with BMI (r = 0.69, P = 0.001 in men; r = 0.75, P < 0.0001, in women) percent body fat (PBF, r = 0.52. P = 0 .018 in men; and r = 0.69, P < 0.0001, in women) and with the sum of 4 fat skinfolds (Sigma skinfolds). E-1-FA level was significantly and positively associated with fasting insulin (r = 0.62, P = 0.003 in men, and r = 0.48, P = 0.023 in women) but not with fasting glucose levels. E-1-FA correlated with insulin sensitivity (SI, r = -0.72 in men; and -0.76, in women, both P < 0.0001). In men, E-1-FA levels also correlated with systolic blood press ure (r = 0.59, P = 0.01), total triglycerides (r = 0.63, P = 0.003), VLDL-t riglycerides (r = 0.62, P = 0.004) and VLDL-cholesterol (r = 0.48, P = 0.03 ), but not with diastolic blood pressure, serum total or LDL-cholesterol, o r total and HDL2 and HDL3 subfractions of HDL cholesterol. After controllin g for fat mass, only the correlation between VLDL-triglycerides and E-1-FA levels remained significant. In women, E-1-FA levels correlated with total triglycerides (r = -0.66, P = 0.001), VLDL-triglycerides (r = 0.65, P = 0.0 01), VLDL-cholesterol (r = 0.63, P = 0.002), LDL-cholesterol (r = 0.57, P = 0.005) and total and HDL2 and HDL3 subfractions of HDL cholesterol (r = -0 .58, -0.48, -0.61, P = 0.004, 0.02 and 0.002, respectively), but not with s ystolic or diastolic blood pressure or total cholesterol. However, covarian ce analysis revealed that controlling for the concomitant variation in body fat mass eliminated all these associations. Fasting plasma E-1-FA concentr ation correlated with serum leptin (r = 0.60, P = 0.005 in men; r = 0.75, P = 0.0001, in women). However, these correlations no longer persisted after controlling for fat mass (r = 0.33 and 0.36, P = NS). Stepwise regression analysis models were tested, with E-1-FA as the dependent variable, and Sig ma skinfolds and SI as independent covariables. Both the Sigma skinfolds (P = 0.03) and S-1 (P = 0.01) entered the equation at a statistically signifi cant level in men. Therefore, insulin sensitivity was related to E1-FA inde pendently of fat in men. In women only Sigma skinfolds (P = 0.04) entered t he regression model at a statistically significantly level. Fifty-seven per cent of the variance in plasma E-1-FA levels in men, and 50% in women, was accounted for using a regression model that combined these variables. CONCLUSIONS Oestrone-fatty acid esters circulate in human blood in proporti on to body fat, independently of gender. Plasma oestrone-fatty acid ester l evels are associated with insulin sensitivity in men, independently of body fat. These findings may widen our perspective on the regulation of insulin action and control of body weight.