The adipose tissue - liver axis in the pathogenesis of vascular diseases

Abstract

Several obesity-related metabolic abnormalities may contribute to the unfavorable cardiovascular risk in obese subjects. These risk factor include dyslipidemia, reflected by low HDL-cholesterol and high triglycerides, hypertension, low-grade inflammation, a prothrombotic state and abnormalities in glucose homeostasis. Dysfunctional adipose tissue, characterized by an increased secretion of pro-inflammatory cytokines, free fatty acids and a decreased secretion of anti-inflammatory adipokines, has been implicated in the development of cardiovascular risk factors. A review of literature presented in this thesis shows that dysfunctional adipose tissue may influence platelet activity, enhance the production of coagulation proteins and attenuate fibrinolysis, thus stimulating a prothrombotic state. The concept of the adipose tissue–liver axis is proposed indicating an interconnection between adipose tissue and the liver through the portal vein, in which visceral adipose tissue(VAT) derived adipokines and cytokines are secreted that may induce the hepatic synthesis of coagulation proteins and pro-inflammatory cytokines. In subsequent in vitro experiments, it was shown that the secretome of (pre)adipocytes stimulated the production of fibrinogen and plasminogen activator inhibitor(PAI)-1 by HepG2 hepatocytes, while there was no effect on tissue factor production. Predominantly interleukin(IL)-6 and IL-1β present in the culture media from (pre)adipocytes were responsible for stimulating fibrinogen and PAI-1 production respectively. To determine the association between the amount and composition of adipose tissue and markers of hemostasis, coagulation and fibrinolysis, a study in 32 patients undergoing aortic surgery was undertaken. Plasma concentrations of thrombin activatable fibrinolysis inhibitor, tissue plasminogen activator, factor VIII and PAI-1 were significantly higher in the inferior mesenteric vein compared with the subclavian vein, suggesting that VAT may contribute to circulating levels of these proteins. Furthermore, not only the amount of subcutaneous adipose tissue(SAT) and VAT was related to prothrombotic proteins, also morphologic characteristics of adipose tissue were associated with these proteins. Also, a broad panel of cytokines, chemokines and adipokines was measured in mesenteric and systemic venous blood samples of this study population. Higher levels of the chemokine interferon-g-inducible protein(IP)-10 and angiogenic hepatic growth factor(HGF) in the inferior mesenteric vein led to the hypothesis that VAT significantly contributes to these proteins. Amount and morphology of adipose tissue were predominantly related to adiponectin levels. The hypothesis that VAT contributes to IP-10 and HGF levels was supported by the significant relation between specifically VAT and these proteins in a large cohort of patients with manifest vascular disease from the SMART-study. In addition, both IP-10 and HGF were independently associated with components of the metabolic syndrome. Even though patients with atherosclerotic disease usually already exhibit a state of low-grade inflammation, still C-reactive protein(CRP) levels were significantly higher with increasing adiposity in these patients. Specifically VAT and not SAT was associated with (hepatically produced) CRP, supporting the concept that pro-inflammatory cytokines produced by VAT stimulate the production of pro-inflammatory proteins by the liver. In conclusion, evidence presented in this thesis show that VAT, being distinct from SAT with respect to secretory characteristics, may directly influence hepatic metabolism by stimulating coagulation and inflammation.