ORIGINAL ARTICLE Role of sex steroids, intrahepatic fat and liver enzymes in the association between SHBG and metabolic features Fabrice Bonnet*, Fritz-Line Velayoudom Cephise†, Alain Gautier*, Severine Dubois‡, Catherine Massart§, Alioune Camara*, Laurent Larifla¶, Beverley Balkau** , †† and Pierre-Henri Ducluzeau‡ *Department of Endocrinology, CHU Rennes, Univ Rennes 1, Inserm UMR 991, Rennes, Department of Endocrinology- Diabetology, University Hospital of Pointe-a-Pitre, Pointe-a-Pitre, Department of Diabetology, University Hospital of Angers, Angers, §Laboratory of hormonology, CHU de Rennes, Inserm 0203 Centre d’Investigation Clinique CIC-P, Univ Rennes 1, Rennes, Research Group, Clinical Epidemiology and Medicine, University Hospital of Guadeloupe, University of Antilles and Guyane, Pointe-a-Pitre, **INSERM CESP Centre for Research in Epidemiology and Population Health, U 1018, Epidemiology of Diabetes, Obesity and Chronic Kidney Disease Over the Lifecourse and ††University Paris Sud 11, UMRS 1018, Villejuif, France Summary Background SHBG and liver enzymes levels are both associated with the risk of type 2 diabetes. However, the relationship between SHBG with liver enzymes and intrahepatic fat content remain poorly understood. Objective To investigate whether SHBG is correlated with glucose and lipids levels and whether this association depends on fatty liver content, liver enzymes or sex hormone concentrations. Design and Patients We studied 233 dysmetabolic men with measures of plasma SHBG, total testosterone, 17b-oestradiol, glucose, adiponectin, liver enzymes and hepatokines. Intrahepat- ic liver fat and visceral fat contents were measured by magnetic resonance imaging in 108 of these individuals. Results After adjustment for age, SHBG concentration was inversely correlated with fasting glucose (b standardized = 021, P = 00007), HbA1c (b standardized = 027, P < 00001), triglyce- rides (b standardized = 019, P = 0003) and positively correlated with HDL-Cholesterol (b standardized = 014, P = 003). These cor- relations persisted after adjustment for either total testosterone or 17b-oestradiol levels. SHBG was not related to either fetuin A or FGF 21 concentrations. The inverse association of SHBG with HbA1c and glycaemia was not altered after adjusting for liver markers but was no longer significant after adjustment for hepa- tic fat content. Conclusion The significant association between SHBG and fasting glycaemia, HbA1c and lipid levels in dysmetabolic men was not related to either sex hormones or markers of liver func- tion, but was dependent on intrahepatic fat. This suggests that intrahepatic fat, but not alterations in liver function markers, may be involved in the association between SHBG and glucose and lipid metabolism. (Received 14 September 2012; returned for revision 23 September 2012; finally revised 30 September 2012; accepted 29 October 2012) Introduction SHBG is a protein transporter of sex steroids and is synthesized by the liver. Low levels of SHBG have been shown to be associ- ated with visceral adiposity, fatty liver and an increased risk of type 2 diabetes. 14 Furthermore, studies applying the Mendelian randomization approach suggest that SHBG may be directly involved in the pathogenesis of type 2 diabetes. 5 However, the pathophysiological mechanisms underlying the inverse associa- tion between SHBG and both the presence of the metabolic syn- drome and the onset of hyperglycaemia remain poorly understood. Hyperinsulinaemia and intrahepatic fat content have been proposed as possible determinants of the molecular expression of SHBG. 6 However, whether the relationship between plasma SHBG concentration and glucose levels depends on hepatic lipogenesis or intrahepatic fat content has not been fully investigated. Furthermore, it has been consistently shown that markers of liver function, specifically cglutamyltransferase (GGT) and ala- nine aminotransferase (ALT), are predictive of the development of type 2 diabetes. 7,8 Sex hormones have a direct impact on the liver and may modulate the release of liver markers. Whereas both a high GGT and a low SHBG concentration are associated with incident type 2 diabetes, it is not known whether liver function markers modulate or interact in the relationship between SHBG and the risk of hyperglycaemia. The aim of this study was therefore to determine whether hepatic fat content, and GGT and ALT levels modulate or Correspondence: Fabrice Bonnet, Department of Endocrinology, University Hospital of Rennes, 2 rue Henri le Guilloux-35033 RENNES CEDEX 9, France. Tel.: +33 2 99 26 71 42; Fax: +33 2 99 26 71 49; E-mail: fabrice.bonnet@chu-rennes.fr © 2012 John Wiley & Sons Ltd 517 Clinical Endocrinology (2013) 79, 517–522 doi: 10.1111/cen.12089