3352 Diabetes Volume 65, November 2016 Brennan K. Smith,1 Rebecca J. Ford,1 Eric M. Desjardins,1 Alex E. Green,1 Meghan C. Hughes,2 Vanessa P. Houde,1 Emily A. Day,1 Katarina Marcinko,1 Justin D. Crane,1 Emilio P. Mottillo,1 Christopher G.R. Perry,2 Bruce E. Kemp,3,4 Mark A. Tarnopolsky,5 and Gregory R. Steinberg1,6 Salsalate (Salicylate) Uncouples Mitochondria, Improves Glucose Homeostasis, and Reduces Liver Lipids Independent of AMPK-b1 Diabetes 2016;65:3352–3361 | DOI: 10.2337/db16-0564 Salsalate is a prodrug of salicylate that lowers blood insulin resistance and type 2 diabetes (T2D) (1). Despite glucose in patients with type 2 diabetes (T2D) and the rising prevalence of NAFLD and importance for the reduces nonalcoholic fatty liver disease (NAFLD) in animal development of T2D, there are currently no pharmacolog- models; however, the mechanism mediating these effects is ical approaches for the treatment of this disease (2). unclear. Salicylate directly activates AMPK via the b1sub- Salsalate is a prodrug of salicylate and is hydrolyzed in unit, but whether salsalate requires AMPK-b1toimprove the small intestine to produce two molecules of salic- T2D and NAFLD has not been examined. Therefore, wild-type ylate (3,4). The circulating concentration of salicylate in b – b (WT) and AMPK- 1 knockout (AMPK- 1KO) mice were humans administered salsalate in T2D clinical trials is treated with a salsalate dose resulting in clinically relevant ;1 mmol/L (5–8). Salsalate has also been shown to im- serum salicylate concentrations (∼1 mmol/L). Salsalate treat- prove symptoms of NAFLD (9) and nonalcoholic steato- ment increased VO2, lowered fasting glucose, improved glu- METABOLISM hepatitis in mice (10). The mechanism by which salsalate cose tolerance, and led to an ∼55% reduction in liver lipid improves T2D and NAFLD is currently unclear, although content. These effects were observed in both WT and AMPK- – b1KO mice. To explain these AMPK-independent effects, we multiple mechanisms have been proposed (10 16). The found that salicylate increases oligomycin-insensitive respi- mechanism of action most commonly associated with sal- ration (state 4o) and directly increases mitochondrial proton salate is the direct repressing effect of salicylate on in- k b b conductance at clinical concentrations. This uncoupling ef- hibitor of nuclear factor -B kinase subunit (IKK- )to fl – fect is tightly correlated with the suppression of de novo reduce in ammation (11 13). However, the concentra- lipogenesis. Salicylate is also able to stimulate brown adipose tion of salicylate used in these studies nonspecifically in- tissue respiration independent of uncoupling protein 1. These hibits many protein kinases through direct competition data indicate that the primary mechanism by which salsalate with their ATP binding sites (16–18). In contrast to ki- improves glucose homeostasis and NAFLD is via salicylate- nase inhibition, salicylate has also been shown to directly driven mitochondrial uncoupling. activate AMPK, a metabolic-sensing enzyme important for regulating inflammation (19), liver lipid metabolism (20), and brown fat thermogenesis (21,22). The effect of Nonalcoholic fatty liver disease (NAFLD) is considered salicylate on AMPK occurs via a direct interaction with an important contributing factor to the development of the Ser108 residue of the b1 subunit (16,23). The most 1Division of Endocrinology and Metabolism, Department of Medicine, McMaster Corresponding author: Gregory R. Steinberg, [email protected]. University, Hamilton, Ontario, Canada Received 2 May 2016 and accepted 16 August 2016. 2Muscle Health Research Centre, School of Kinesiology and Health Science, York This article contains Supplementary Data online at http://diabetes University, Toronto, Ontario, Canada .diabetesjournals.org/lookup/suppl/doi:10.2337/db16-0564/-/DC1. 3Protein Chemistry and Metabolism, St Vincent’s Institute and Department of Medicine, University of Melbourne, Melbourne, Victoria, Australia © 2016 by the American Diabetes Association. Readers may use this article as 4Mary MacKillop Institute for Health Research, Australian Catholic University, long as the work is properly cited, the use is educational and not for profit, and the Fitzroy, Victoria, Australia work is not altered. More information is available at http://www.diabetesjournals 5Department of Pediatrics, McMaster University, Hamilton, Ontario, Canada .org/content/license. 6Department of Biochemistry and Biomedical Sciences, McMaster University, Hamilton, Ontario, Canada diabetes.diabetesjournals.org Smith and Associates 3353 recent proposal to explain the mechanism of salicylate body weight) administration was performed as previously suggests that salsalate can activate brown adipose tissue described (26). In a subset of animals, insulin (1 unit/kg) (BAT) through activation of cAMP-dependent protein was administered before sacrifice to examine insulin sig- kinase (15). naling in the liver and 2-deoxy-D-glucose uptake into skeletal Although salicylate directly activates AMPK via the muscle and adipose tissue (27). b1 subunit, daily intraperitoneal injections of salicylate Analytical Measurements (250 mg/kg) improved a marker of HOMA insulin resis- Serum salicylate concentrations were determined from a tance in both wild-type (WT) and AMPK-b1–knockout commercially available kit (Neogen Corporation) follow- (KO) mice fed a high-fat diet (HFD) (16). Because the ing the manufacturer’s instructions. Liver sections were dose of salicylate used in this study results in serum con- stained with hematoxylin and eosin. Liver and tibialis centrations of salicylate more than double the clinical anterior samples were extracted by the Folch method to levels after the oral intake of salsalate (;2.4 mmol/L measure tissue triglyceride levels (28). Primary hepato- compared with ;1.0 mmol/L, respectively) we hypothe- cytes were freshly isolated by collagenase perfusion for sized that the AMPK-b1–independent effects may have the lipogenesis and respiration measurements. Mitochon- been a result of off-target kinase inhibition (16–18). drial membrane potential (Δcm) was measured using The purpose of this study was to investigate whether the tetramethyl rhodamine methyl ester (TMRM) stain oral delivery of clinically relevant concentrations of salsa- (20 nmol/L, nonquenching) (29). Primary hepatocyte de late improves glucose homeostasis and reduces NAFLD novo lipogenesis was measured similar to previous re- through an AMPK-b1–dependent pathway. Salsalate was ports using 3H acetate (PerkinElmer) (20). Quantitative observed to improve whole-body glucose homeostasis, real-time PCR was performed as previously described to de- reduce liver lipid content, and improve adipose tissue termine mRNA expression levels (19). Briefly, epididymal inflammation independently of AMPK-b1. These diverse adipose tissue was lysed in TRIzol reagent (Invitrogen, metabolic effects of salsalate are associated with the pro- Carlsbad, CA) to remove lipid, and the aqueous phase tonophoric effects of salicylate and subsequent mitochon- was applied to an RNeasy kit (Qiagen, Valencia, CA) col- drial uncoupling and increased energy expenditure. These umn for subsequent purification. Relative gene expression data suggest that salicylate-driven mitochondrial uncou- 2DD was calculated using the comparative Ct (2 Ct) method, pling is the primary mechanism mediating the beneficial where values were normalized to the housekeeping gene effects of salsalate therapy on NAFLD and T2D. Ppia. TaqMan primers F4/80 (Emr1, Mm00802529_m1), cluster of differentiation 68 (Cd68, Mm00839636_g1), RESEARCH DESIGN AND METHODS tumor necrosis factor-a (Tnf-a, Mm00443258_m1), che- Study Approval mokine (C-C motif) ligand 2 (CCL2, Mm00441242_m1), All animal procedures were approved by the McMaster and interleukin-1b (Il-1b, Mm00434228_m1), were pur- University Animal Ethics Research Board (AUP #: 12-12- chased from Invitrogen. Western blotting was performed 44; Hamilton, Ontario, Canada) and conform to the Guide similar to the description by Ford et al. (9), and all anti- for the Care and Use of Laboratory Animals published by the bodies were purchased from Cell Signaling. ATP concen- U.S. National Institutes of Health. tration was determined in freeze-clamped liver tissue according to the manufacturer’s instruction (ab113849; Animals Abcam) (30). WT and AMPK-b1KO mice were started on an HFD (60% calories from fat) at 8 weeks of age. At 4 weeks after the Respiration Methods initiation of the HFD, half of the mice continued on the Mitochondrial respiration was measured by high-resolution HFD and the other half were switched to an HFD supple- respirometry (Oxygraph-2k; Oroboros, Innsbruck, Austria) mented with 2.5 g/kg salsalate. These diets were main- at 37°C and room air saturated oxygen tension. Per- tained for 8 weeks until sacrifice (Supplementary Fig. 1A). meabilized primary hepatocyte respiration was per- The glucose tolerance test was performed in 6-h fasted formed in MIRO5 buffer containing EGTA (0.5 mmol/L), mice after an injection of glucose (0.8 g/kg i.p.). The ala- MgCl2*6H2O (3 mmol/L), K-lactobionate (60 mmol/L), nine tolerance test was performed in 16-h fasted mice KH2PO4 (10 mmol/L), HEPES (20 mmol/L), sucrose after an injection of alanine (2 g/kg i.p.) (24). Blood glu- (110 mmol/L), and fatty acid–free BSA (1 g/L). Primary cose levels were determined from a small tail vein nick hepatocytes were scraped into 2 mL of respiration buffer, using a One Touch Ultra Glucometer (LifeScan