Tumor necrosis factor system activity is associated with insulin resistance and dyslipidemia in myotonic dystrophy

Myotonic dystrophy (MyD) is a multisystem autosomal dominant disorder associated with progressive muscle wasting and weakness. The striking metabolic abnormality in MyD is insulin resistance. The mechanism by which target tissues are insensitive to insulin action remains uncertain. In a recent study, plasma soluble tumor necrosis factor receptor (sTNFR)2 levels were found to be associated with muscle tissue mass and insulin resistance. Given these associations, we speculated that disorders of the muscle cell membrane could lead simultaneously to insulin insensitivity and sTNFR2 leakage in MyD. To test this hypothesis, we measured the levels of circulating sTNFR1 and sTNFR2 and insulin resistance in MyD patients. We studied 22 MyD patients and 24 age-, BMI-, and fat mass-matched control subjects. Both MyD men and women showed higher plasma insulin levels in the presence of comparable glucose concentrations than did control subjects. sTNFR2, but not sTNFR1, levels were approximately 1.5-fold higher in MyD patients. In parallel with these findings, the fasting insulin resistance index (FIRI) was also higher in MyD patients. In fact, in the whole population, fasting insulin and FIRI strongly correlated with sTNFR2 in both men (r = 0.77 and r = 0.81, P<0.0001, respectively) and women (r = 0.67 and r = 0.64, P = 0.001, respectively). sTNFR2 levels were also associated with the insulin sensitivity index (S(I)), calculated from an oral glucose tolerance test (OGTT) according to the method by Cederholm and Wibell (r = -0.43, P = 0.006). We constructed a multiple linear regression to predict FIRI, with BMI, waist-to-hip ratio, and sTNFR2 as independent variables. In this model, both BMI (P = 0.0014) and sTNFR2 (P = 0.0048) levels contributed independently to 46% of the variance of FIRI. In another model, in which FIRI was substituted for S(I) from the OGTT, both BMI (P = 0.0001) and sTNFR2 (P = 0.04) levels contributed independently to 48% of the variance of S(I) from the OGTT. Plasma cholesterol and triglyceride concentrations were significantly increased in MyD patients. sTNFR1 and sTNFR2 levels were found to be strongly associated with plasma cholesterol, LDL cholesterol, and triglycerides. sTNFR1 and sTNFR2 also correlated with serum creatine kinase activity in MyD patients (r = 0.57, P = 0.006; r = 0.75, P<0.0001, respectively). In conclusion, here we describe, for the first time to our knowledge, a relationship between insulin action and plasma sTNFR2 concentration in MyD patients. We have also found increased concentrations of plasma triglycerides and cholesterol levels in parallel with sTNFR1 and sTNFR2 concentrations in MyD patients. We speculate that the latter associations are dependent on, and secondary to, increased tumor necrosis factor (TNF)-alpha action. Whether TNF action is implicated in the pathogenesis of MyD or is a simple marker of disease activity awaits further studies.