Increased Thrombosis in Male Versus Female Mice Is Mediated by Pulsatile Growth Hormone Secretion Possibly through Inhibition of TFPI Expression in the Vasculature.

Clinical reports suggest significant sex differences in risk for thrombosis-related diseases such as myocardial infarction, stroke, and venous thromboembolism. However, little is known about mechanism for such differences. There is a well-described sexual dimorphism in liver protein synthesis that is growth hormone (GH) dependent. GH secretion from the pituitary is itself highly sexually dimorphic with males (M) secreting in a pulsatile (P) and females (F) a continuous (C) fashion. These patterns induce M- and F-specific signatures of liver gene expression. In the past, we and others have observed significant sex differences in murine thrombosis models. Given that most coagulation proteases and inhibitors are synthesized or modified in the liver, we aimed to test whether sex-specific GH secretion patterns contribute to the observed sex differences in thrombosis. We measured whole blood clotting times (WCT), thrombosis susceptibility in the thromboplastin-mediated pulmonary embolism (PE) model, and hemostasis in the tail bleeding time (BT) model in M and F control (WT) and GH-deficient “little” (LIT) mice. We observed that WT Fs had longer WCTs (mean time 61.38 vs. 56.72 sec) and were significantly protected in the PE model (median survival 232.5 vs 165 sec) as compared to M. There were no differences in the BT model across all experiments. Interestingly, F and M LIT animals both had significantly prolonged WCTs (67.56 and 67.30 sec, respectively) and were substantially protected in the PE model (median survival 900 and 1200 sec) as compared to WT. Next, LIT animals were injected twice daily with GH to simulate the P pattern of GH secretion (LIT+). This resulted in a significant shortening of the F and M WCTs back to WT M levels (53.16 and 50.97 sec). A group of F WT animals were also injected with M pattern GH (WT+). This too resulted in significant shortening of the F WCTs (54.10 sec). To explore for possible mechanisms underlying these differences, we measured activity of coagulation factors II, V, VII, VIII, IX, X, and XI. The average of all factor activity levels was significantly higher in WT M vs F (100 vs. 81.99%), significantly lower and in both M and F LIT (60.85 and 57.97%), and increased to WT M levels in M and F LIT+ animals (106.6 and 99%). To determine whether these changes were mediated by changes in liver gene expression, we measured a panel of 30 coagulation protease and inhibitor genes in liver and vascular tissue by Taqman®. Surprisingly, we found no significant differences in coagulation factor expression, but found that expression of TFPI was significantly increased in F vs M WT vasculature (9431 vs. 7678 gene copy number (GCN)). Expression was increased in M and F LIT animals (10350 and 11710 GCN) and fell to below WT levels in M and F LIT+ animals (4534 and 4194 GCN). These results indicate that sex differences in thrombosis in mice are at least in part mediated by sex differences in GH secretion with F mice relatively protected as compared to M. M and F GH-deficient LIT mice are similarly protected as compared to WT M. Repletion of GH in a P pattern reverts M and F LIT and F WT mice to WT M levels. Finally, P GH secretion may promote increased thrombosis through inhibition of TFPI in the vasculature. This represents a novel mechanism underlying these sex-differences in thrombosis mediated by sexually dimorphic GH secretion and its effect on regulation of TFPI in the vasculature.