ARTICLESResveratrol improves health and survivalof mice on a high-calorie dietJoseph A. Baur1*, Kevin J. Pearson2*, Nathan L. Price2, Hamish A. Jamieson7, Carles Lerin8, Avash Kalra2,Vinayakumar V. Prabhu3, Joanne S. Allard2, Guillermo Lopez-Lluch9, Kaitlyn Lewis2, Paul J. Pistell2, Suresh Poosala4,Kevin G. Becker3, Olivier Boss10, Dana Gwinn11, Mingyi Wang5, Sharan Ramaswamy6, Kenneth W. Fishbein6,Richard G. Spencer6, Edward G. Lakatta5, David Le Couteur7, Reuben J. Shaw11, Placido Navas9, Pere Puigserver8,Donald K. Ingram2,12, Rafael de Cabo2& David A. Sinclair1Resveratrol (3,5,49-trihydroxystilbene) extends the lifespan of diverse species including Saccharomyces cerevisiae,Caenorhabditis elegans and Drosophila melanog aster. In these organisms, lifespan extension is dependent on Sir2, aconserved deacetylase proposed to underlie the beneficial effects of caloric restriction. Here we show that resveratrol shiftsthe physiology of middle-aged mice on a high-calorie diet towards that of mice on a standard diet and significantly increasestheir survival. Resveratrol produces changes associated with longer lifespan, including increased insulin sensitivity, reducedinsulin-like growth factor-1 (IGF-I) levels, increased AMP-activated protein kinase (AMPK) and peroxisomeproliferator-activate d receptor-c coactivator 1a (PGC-1a) activity, increased mitochondrial number, and improved motorfunction. Parametric analysis of gene set enrichment revealed that resveratrol opposed the effects of the high-calorie diet in144 out of 153 significantly altered pathways. These data show that improving general health in mammals using smallmolecules is an attainable goal, and point to new approaches for treating obesity-related disorders and diseases of ageing.The number of overweight individuals worldwide has reached 2.1billion, leading to an explosion of obesity-related health problemsassociated with increased morbidity and mortality1,2. Although theassociation of obesity with increased risk of cardiovascular diseaseand diabetes is well known, it is often under-appreciated that the risksof other age-related diseases, such as cancer and inflammatory dis-orders, are also increased. At the other end of the spectrum, reducingcaloric intake by ,40% below that of ad libitum-fed animals (caloricrestriction) is the most robust and reproducible way to delay age-related diseases and extend lifespan in mammals3,4.Experiments with Saccharomyces cerevisiae and Drosophila mela-nogaster have implicated the sirtuin/Sir2 family of NAD1-dependentdeacetylases and mono-ADP-ribosyltransferases as mediators of thephysiological effects of caloric restriction5. In mammals, seven sirtuingenes have been identified (SIRT1–7). SIRT1 regulates such processesas glucose and insulin production, fat metabolism, and cell survival,leading to speculation that sirtuins might mediate effects of caloricrestriction in mammals5. We previously screened over 20,000molecules to identify ,25 that enhance SIRT1 activity in vitro6.Resveratrol, a molecule produced by a variety of plants in responseto stress, emerged as the most potent.Resveratrol has since been shown to extend the lifespan of evolu-tionarily distant species including S. cerevisiae, C. elegans and D. mela-nogaster in a Sir2-dependent manner6–9. A recent study found thatresveratrol improves health and extends maximum lifespan by59% in a vertebrate fish10. In mammalian cells, resveratrol producesSIRT1-dependent effects that are consistent with improved cellularfunction and organismal health11–15. Whether resveratrol acts directlyor indirectly through Sir2 in vivo is currently a subject of debate16.On the basis of the unprecedented ability of resveratrol to improvehealth and extend lifespan in simple organisms, we have askedwhether it has similar effects in mice. We hypothesized that resver-atrol might shift the physiology of mice on a high-calorie diet towardsthat of mice on a standard diet and provide the associated healthbenefits without the mice having to reduce calorie intake. Cohorts ofmiddle-aged (one-year-old) ma le C57BL/6NIA mice were providedwith either a standard diet (SD) or an otherwise equivalent high-calorie diet (60% of calories from fat, HC) for the remainder of theirlives. To each of the diets, we added resveratrol at two concentrationsthat provided an average of 5.2 6 0.1 and 22.4 6 0.4 mg kg21day21,which are feasible daily doses for humans. After 6 months of treat-ment, there was a clear trend towards increased survival and insulinsensitivity. Because the effects were more prominent in the higherdose (22.4 6 0.4 mg kg21day21, HCR), we initially focused ourresources on this group and present the results of those analysesherein. Analyses of the other groups will be presented at a later date.Increased survivalMice on the HC diet steadily gained weight until ,75 weeks of age,after which average weight slowly declined (Fig. 1a). Although miceon the HCR diet were slightly lighter than the HC mice during theinitial months, there was no significant weight difference between the*These authors contributed equally to this work.1Department of Pathology, Paul F. Glenn Laboratories for the Biological Mechanisms of Aging, Harvard Medical School, 77 Avenue Louis Pasteur, Boston, Massachusetts 02115, USA.2Laboratory of Experimental Gerontology,3Gene Expression and Genomics Unit,4Research Resources Branch,5Laboratory of Cardiovascular Science, and6Laboratory of ClinicalInvestigation, Research Resources Branch of the Gerontology Research Center, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore,Maryland 21224, USA.7Centre for Education and Research on Ageing, and the ANZAC Research Institute University of Sydney, Concord, New South Wales 2139, Australia.8Department of Cell Biology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA.9Centro Andaluz de Biologia del Desarrollo, Universidad Pablo deOlavide-CSIC, 41013 Sevilla, Spain.10Sirtris Pharmaceuticals, Inc., 790 Memorial Drive, Cambridge, Massachusetts 02139, USA.11Molecular and Cell Biology Laboratory, The SalkInstitute, 10010 N Torrey Pines Road, La Jolla, California 92037, USA.12Nutritional Neuroscience and Aging Laboratory, Pennington Biomedical Research Center, Louisiana StateUniversity System, 6400 Perkins Road, Baton Rouge, Louisiana 70808, USA.doi:10.1038/nature053541Nature PublishingGroup ©2006groups from 18–24 months, when most of our