LETTERS 358 VOLUME 19 | NUMBER 3 | MARCH 2013 NATURE MEDICINE Hepatic insulin resistance is a driving force in the pathogenesis of type 2 diabetes mellitus (T2DM) and is tightly coupled with excessive storage of fat and the ensuing inflammation within the liver 1–3 . There is compelling evidence that activation of the transcription factor nuclear factor-kB (NF-kB) and downstream inflammatory signaling pathways systemically and in the liver are key events in the etiology of hepatic insulin resistance and b-cell dysfunction, although the molecular mechanisms involved are incompletely understood 3–6 . We here test the hypothesis that receptor activator of NF-kB ligand (RANKL), a prototypic activator of NF-kB, contributes to this process using both an epidemiological and experimental approach. In the prospective population-based Bruneck Study, a high serum concentration of soluble RANKL emerged as a significant (P < 0.001) and independent risk predictor of T2DM manifestation. In close agreement, systemic or hepatic blockage of RANKL signaling in genetic and nutritional mouse models of T2DM resulted in a marked improvement of hepatic insulin sensitivity and amelioration or even normalization of plasma glucose concentrations and glucose tolerance. Overall, this study provides evidence for a role of RANKL signaling in the pathogenesis of T2DM. If so, translation to the clinic may be feasible given current pharmacological strategies to lower RANKL activity to treat osteoporosis. RANKL (also known as TNFSF11) is a member of the tumor necrosis factor superfamily and, after ligation with its cognate receptor RANK (also known as TNFRSF11a), is a potent stimulator of NF-κB. Notably, both RANKL and RANK are expressed in human liver tissue and pancreatic β-cells 7 , and concentrations of the soluble decoy receptor osteoprotegerin (OPG), considered to be a reliable surrogate for the overall activity of this cytokine network, are elevated in patients with T2DM, especially in those with poor glycemic control and compli- cated disease course 8–11 . RANKL exists in both membrane-bound and biologically active soluble forms, with the latter originating from secretion and cleavage 9,12 . Concentrations of soluble RANKL are ele- vated in or predictive of various human diseases, including cardiovas- cular disease, nontraumatic fractures, multiple myeloma, rheumatoid arthritis and inflammatory bowel disease 8–11,13–18 . We determined the distribution of serum concentrations of RANKL and OPG in our study population (n = 844) (Supplementary Fig. 1a). Soluble RANKL concentrations showed associations with insulin resistance assessed by homeostasis model (HOMA-IR) and Gutt Index values and with the number of metabolic syndrome components clustering in an indi- vidual (Supplementary Data) but were not related to most standard population characteristics (Supplementary Table 1). Between 1990 and 2005, 78 of the 844 individuals in the study population (9.2%) developed T2DM (incidence rate, 7.2 per 1,000 person years (95% confidence interval (CI) 5.7–8.9)). We determined the baseline characteristics of subjects with and without incident T2DM (Supplementary Table 2) and found that the concentrations of soluble RANKL differed considerably between the two groups. In a pooled logistic regression analysis adjusted for age, sex and period Blockade of receptor activator of nuclear factor-κB (RANKL) signaling improves hepatic insulin resistance and prevents development of diabetes mellitus Stefan Kiechl 1,16 , Jürgen Wittmann 2 , Andrea Giaccari 3,4 , Michael Knoflach 1 , Peter Willeit 1,5 , Aline Bozec 6 , Alexander R Moschen 7 , Giovanna Muscogiuri 3 , Gian Pio Sorice 3 , Trayana Kireva 6 , Monika Summerer 8 , Stefan Wirtz 9 , Julia Luther 6 , Dirk Mielenz 2 , Ulrike Billmeier 9 , Georg Egger 10 , Agnes Mayr 11 , Friedrich Oberhollenzer 10 , Florian Kronenberg 8 , Michael Orthofer 12 , Josef M Penninger 12 , James B Meigs 13,14 , Enzo Bonora 15 , Herbert Tilg 7 , Johann Willeit 1 & Georg Schett 6,16 1 Department of Neurology, Medical University Innsbruck, Innsbruck, Austria. 2 Department of Internal Medicine 3, Division of Molecular Immunology, University of Erlangen-Nuremberg, Erlangen, Germany. 3 Division of Endocrinology and Metabolic Diseases, Università Cattolica del Sacro Cuore, Policlinico ‘A. Gemelli’, Rome, Italy. 4 Fondazione Don Gnocchi, Milan, Italy. 5 Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK. 6 Department of Internal Medicine 3, University of Erlangen-Nuremberg, Erlangen, Germany. 7 Department of Medicine I, Division of Endocrinology, Gastroenterology and Metabolism, Medical University Innsbruck, Innsbruck, Austria. 8 Department of Medical Genetics, Division of Genetic Epidemiology, Molecular and Clinical Pharmacology, Medical University Innsbruck, Innsbruck Austria. 9 Department of Internal Medicine I, University of Erlangen-Nuremberg, Erlangen, Germany. 10 Department of Internal Medicine, Bruneck Hospital, Bruneck, Italy. 11 Department of Laboratory Medicine, Bruneck Hospital, Bruneck, Italy. 12 Institute of Molecular Biotechnology of the Austrian Academy of Sciences (IMBA), Vienna, Austria. 13 Department of Internal Medicine, Harvard University, Boston, Massachusetts, USA. 14 Department of Medicine, General Medicine Division, Massachusetts General Hospital, Boston, Massachusetts, USA. 15 Division of Endocrinology, Diabetes and Metabolic Diseases, University and Hospital Trust of Verona, Verona, Italy. 16 These authors contributed equally to this work. Correspondence should be addressed to S.K. (stefan.kiechl@i-med.ac.at) or G.S. (georg.schett@uk-erlangen.de). Received 25 October 2012; accepted 8 January 2013; published online 10 February 2013; doi:10.1038/nm.3084 npg © 2013 Nature America, Inc. All rights reserved.