Unacylated Ghrelin Reduces Skeletal Muscle Reactive Oxygen Species

874 Diabetes Volume 65, April 2016

Gianluca Gortan Cappellari,1 Michela Zanetti,1 Annamaria Semolic,1 Pierandrea Vinci,1 Giulia Ruozi,2 Antonella Falcione,2 Nicoletta Filigheddu,3 Gianfranco Guarnieri,1 Andrea Graziani,3,4 Mauro Giacca,2 and Rocco Barazzoni1

Unacylated Ghrelin Reduces Skeletal Muscle Reactive Oxygen Species Generation and Inﬂammation and Prevents High-Fat Diet–Induced Hyperglycemia and Whole-Body Insulin Resistance in Rodents

Diabetes 2016;65:874–886 | DOI: 10.2337/db15-1019

Excess reactive oxygen species (ROS) generation and insulin resistance. Stimulated muscle autophagy could inflammation may contribute to obesity-associated contribute to UnAG activities. These findings support UnAG skeletal muscle insulin resistance. Ghrelin is a gastric as a therapeutic strategy for obesity-associated metabolic hormone whose unacylated form (UnAG) is associated alterations. with whole-body insulin sensitivity in humans and may reduce oxidative stress in nonmuscle cells in vitro. We hypothesized that UnAG 1) lowers muscle ROS produc- Clustered metabolic abnormalities, including excess re- tion and inflammation and enhances tissue insulin ac- active oxygen species (ROS) generation and inflammation tion in lean rats and 2) prevents muscle metabolic METABOLISM activation, are proposed contributors to the onset of alterations and normalizes insulin resistance and hyper- skeletal muscle insulin resistance (1–5). Excess muscle – glycemia in high-fat diet (HFD) induced obesity. In ROS production and inflammation are indeed linked 12-week-old lean rats, UnAG (4-day, twice-daily subcu- at the level of inhibitor of kB(IkB)/nuclear factor-kB taneous 200-mg injections) reduced gastrocnemius mi- (NF-kB) activation and may cause insulin resistance by tochondrial ROS generation and inflammatory cytokines inhibiting insulin signaling downstream of insulin recep- while enhancing AKT-dependent signaling and insulin- tor (2,3,5). Ghrelin is a peptide hormone predominantly stimulated glucose uptake. In HFD-treated mice, chronic UnAG overexpression prevented obesity-associated hy- secreted by the stomach, and its acylated form (AG) is a perglycemia and whole-body insulin resistance (insulin major hypothalamic orexigenic signal (6,7). Sustained AG tolerance test) as well as muscle oxidative stress, inflam- administration causes weight gain and hyperglycemia de- mation, and altered insulin signaling. In myotubes, UnAG spite enhanced muscle mitochondrial oxidative capacity consistently lowered mitochondrial ROS production and (8,9) by increasing food intake, hepatic gluconeogenesis, enhanced insulin signaling, whereas UnAG effects were and fat deposition in rodents (10,11). A comprehensive prevented by small interfering RNA–mediated silencing understanding of the metabolic impact of ghrelin, how- of the autophagy mediator ATG5. Thus, UnAG lowers ever, has recently been allowed by reports of independent, mitochondrial ROS production and inflammation while more favorable effects of its unacylated form (UnAG). enhancing insulin action in rodent skeletal muscle. In HFD- Although no specific UnAG receptor has yet been identi- induced obesity, these effects prevent hyperglycemia and fied, UnAG counteracts glucogenic effects of AG as well as

1Department of Medicine, Surgery and Health Sciences, University of Trieste, Received 22 July 2015 and accepted 13 January 2016. Trieste, Italy This article contains Supplementary Data online at http://diabetes 2 Molecular Medicine Laboratory, International Centre for Genetic Engineering and .diabetesjournals.org/lookup/suppl/doi:10.2337/db15-1019/-/DC1. Biotechnology, Trieste, Italy © 2016 by the American Diabetes Association. Readers may use this article as 3Department of Translational Medicine, University of Piemonte Orientale “Amedeo long as the work is properly cited, the use is educational and not for proﬁt, and Avogadro,” Novara, Italy the work is not altered. 4Medical School, Università Vita-Salute San Raffaele, Milan, Italy Corresponding author: Rocco Barazzoni, [email protected]. diabetes.diabetesjournals.org Gortan Cappellari and Associates 875

AG-induced hyperglycemia (10), and negative associations New Brunswick, NJ) and were killed as described previously. have been reported between circulating UnAG and markers Insulin tolerance tests (ITTs) were performed at 15 weeks of of whole-body insulin resistance in humans (12,13). Of treatment by intraperitoneal insulin injection (Humulin R interest, emerging antioxidant effects have been reported 3 nmol/kg; Eli Lilly, Indianapolis, IN) after a 4-h fast. Blood for UnAG in various cell types (14–17), and we have dem- glucose was measured from tail blood (ACCU-CHEK Active; onstrated that UnAG stimulates autophagy in rodent mus- Roche, Basel, Switzerland) immediately before injection and cle, thereby also potentially lowering muscle oxidative stress at 20, 40, 60, and 80 min. through disposal of damaged mitochondria (18). No in- formation is available, however, on 1) the impact of UnAG Myotube Experiments on skeletal muscle ROS generation, inflammation, and C2C12 myoblasts were differentiated in myotubes (20). insulin action and 2) whether UnAG prevents altered ox- After a 4-day incubation with differentiation medium and idative stress, inflammation, and insulin action in obesity an 18-h starvation, cells were treated with AG or UnAG m and diabetes. (0.1, 0.5, 1 mol/L) for 48 h, collected, and processed. In We therefore studied lean rats and a transgenic mouse additional experiments, the potential role of autophagy model of systemic UnAG overproduction (19) to test the in effects of UnAG was investigated by genomic silencing hypothesis that UnAG 1) lowers mitochondrial ROS pro- of the autophagy mediator ATG5 (18). Small interfering duction and inflammation and enhances insulin action in RNA (siRNA) knockdown of ATG5 was performed by re- fi lean rodent muscle and 2) normalizes high-fat diet (HFD)– verse transfection at nal 25 nmol/L concentration with induced muscle metabolic alterations, whole-body insu- mouse ATG5 siRNA (M-064838-02-0005; Dharmacon) or lin resistance, and hyperglycemia. In addition, effects of with nontargeting control siRNA #4 (NT4) (D-001210- UnAG were verified in vitro in myotubes, where we also 04-20; Dharmacon) using Lipofectamine RNAiMAX (Life mechanistically tested the hypothesis that UnAG activi- Technologies). Twenty-four hours after transfection, cul- ties are at least partly mediated by positive modulation ture medium was replaced, and after 36 h, it was differ- of autophagy. entiated, treated, and processed as aforementioned. ATG5 protein levels were quantified by Western blot. RESEARCH DESIGN AND METHODS Analytical Methods Experimental Design Plasma Insulin and Nonesterified Fatty Acids Exogenous UnAG Administration Plasma insulin concentration was measured by ELISA Experiments were approved by the Animal Studies (Ultrasensitive ELISA; DRG, Springfield, NJ). Plasma Committee at the University of Trieste (Trieste, Italy). glucose and nonesterified fatty acid (NEFA) levels were Twenty 12-week-old male Wistar rats (Harlan-Italy, San determined by standard enzymatic colorimetric assays Pietro-al-Natisone, Udine, Italy) were housed for 2 weeks (21,22). in individual cages with a 12-h light/dark cycle at the University of Trieste Animal Facility, with ad libitum Ex Vivo Redox State access to water and standard chow (Harlan 2018, 14.2 kJ/g). Mitochondrial H2O2 production was assessed in isolated Animals were then randomly assigned to 4-day, twice-daily intact mitochondria from tissues and cells by using the m 200-mg subcutaneous injections of UnAG (n = 10; Amplex Red (10 mol/L; Invitrogen, Carlsbad, CA) fi Bachem, Bubendorf, Switzerland) or vehicle (control [Ct]) horseradish peroxidase method, modi ed as previously (n = 10; NaCl 0.9% weight for volume). UnAG dose was reported and normalized by citrate synthase activity in based on previous studies in which equimolar AG mod- the same mitochondrial preparation (22,23). Assay sub- ulated the same parameters (8). Body weight and food strate concentrations (mmol/L) were 8 glutamate, intake were monitored daily; after the last injection, food 4 malate (GM); 10 succinate (S); 4 glutamate, 2 malate, was removed for 3 h followed by anesthesia (tiobutabarbi- 10 succinate (GMS); and 0.05 palmitoyl-L-carnitine, 2 malate tal 100 mg/kg, tiletamine/zolazepam [1:1] 40 mg/kg i.p.). (PCM). Superoxide anion production sources in gastrocne- Gastrocnemius and extensor digitorum longus muscles miusmusclewhole-tissuehomogenatewereassessedbyus- were then surgically isolated, and blood was collected by ing the lucigenin chemiluminescence method as previously heart puncture. described (22) and normalized by protein concentration (BCA Protein Assay Kit; Pierce, Rockford, IL). The impact Transgenic UnAG Overexpression of subsequent addition of specific inhibitors on specific Generation and characteristics of transgenic mice over- substrate-stimulated production rates was used to evaluate expressing UnAG (Tg Myh6/Ghrl) were previously de- relative superoxide production from each source (mito- scribed (19). Selective ghrelin overproduction in the heart, chondria: 5 mmol/L CCCP on succinate; nitric oxide syn- G characterized by negligible acylating activity, results in a thase [NOS]: 10 mmol/L L-N -nitro-L-arginine methyl ester 40-fold increment in circulating UnAG without AG mod- on 10 mmol/L L-arginine; NADPH oxidase: 200 mmol/L ification. Fourteen Tg Myh6/Ghrl and 14 matched wild- diphenylene iodonium on 1 mmol/L NADPH; xanthine ox- type male mice underwent 16-week standard or HFD idase: 200 mmol/L oxypurinol on 500 mmol/L xanthine) as feeding (10% or 60% calories from fat; Research Diets, referenced. 876 Unacylated Ghrelin and Muscle Metabolism Diabetes Volume 65, April 2016

Glutathione and Antioxidant Enzyme Activities pyruvate substituted with 2-DG (1 mmol/L), samples were Total and oxidized glutathione were determined as refer- snap frozen and kept at 280°C. After homogenization enced (24) on ;50 mg gastrocnemius cleaned and homog- in ultrapure water followed by NaOH addition (0.07 N), enized in ice-cold 5% (weight for volume) metaphosphoric enzymes and endogenous NAD(P)H and NAD(P) were acid (20 mL/g tissue). Reduced glutathione (GSH) was inactivated by 45-min incubation at 85°C. Equinormal quan- calculated as total minus oxidized fraction (GSSG). Com- tities of HCl were then added, and samples were cleared of mercial kits were used to measure catalase (Amplex Red debris by centrifugation (10,000g for 5 min) and transferred Catalase Assay Kit; Invitrogen) and glutathione peroxi- to 96-well microplates for incubation (37°C for 60 min) in dase activities (Abcam,Cambridge,U.K.). assay buffer with b-NADP (0.1 mmol/L) and glucose-6- phosphate (G6P) dehydrogenase (20 units/mL) from Protein Analyses Leuconostoc mesenteroides (buffer C) or with b-NAD xMAP. Cytokine profile and insulin signaling protein phos- Y1162/Y1163 S312 (0.1 mmol/L) and G6P dehydrogenase (0.3 units/mL) phorylation (p) at insulin receptor (IR) ,IRS-1 , (buffer D). Concentrations of G6P and G6P + 2-deoxyglucose- S473 bS9 T246 T421/S424 AKT ,GSK-3 ,PRAS40 , and P70S6K levels 6-phosphate (2-DG6P) were quantified by fluorimetrically were measured by xMAP technology (MAGPIX; Luminex measuring (Infinite F200; Tecan, Männedorf, Switzerland) Corporation, Austin, TX) using commercial kits, validated conversion of resazurin to resorufininbuffersCandD, fi by the manufacturer for multiplexing pro ling (LRC0002M, respectively. Values related to 2-DG6P were calculated LHO0001M, LHO0002; Life Technologies, Carlsbad, CA). by subtraction, interpolated on a standard curve of MILLIPLEX Analyst software (Millipore, Billerica, MA) was 2-DG6P, and normalized by protein concentration used for interpolating data to a standard curve. Phosphory- in sample homogenate. Tissue 2-DG6P uptake was lation of each protein is expressed as phosphoprotein units expressed in micromoles of 2-DG per milligram protein per total in picograms. in 30 min. Western Blot. Western blots were performed as previously described (21,22,25). Equal loading was checked ATP Synthesis and Complex-Related ATP Production by Ponceau S staining and GAPDH reprobing. Primary The ATP synthesis rate in tissues and cells was measured antibody dilutions were anti-Mn superoxide dismutase ex vivo in freshly isolated mitochondria by using a luciferin- (SOD) and anti-CuZnSOD (Stressgen, Ann Arbor, MI) luciferase luminometric assay (22). Integrity of mitochon- 1:5,000 and 1:1,000, respectively; anti-IkB (Cell Signaling dria isolated by gentle homogenization was tested by Technology, Beverly, MA) 1:500; anti-pIRS-1Y612 (Abcam) comparison of citrate synthase measurements in sam- 1:500; anti-ATG5 (Cell Signaling Technology) 1:2,000; ples before and after membrane disruption (29). After anti-LC3B (Sigma) 1:1,500; anti-b-actin (Sigma) 1:25,000; signal stabilization and addition of excess substrates, a and anti-GAPDH (Santa Cruz Biotechnology, Dallas, TX) first 10-min kinetic read was performed followed by 1:1,000. addition of 100 mmol/L ADP and a 20-min read. Final respiration substrates composition and reaction concen- Electrophoretic Mobility Shift Assay trations (mmol/L) were 0.25 pyruvate, 0.0125 palmitoyl- NF-kB binding activity was assessed by nonradioactive elec- L-carnitine, 2.5 a-ketoglutarate, 0.25 malate (PPKM); trophoretic mobility shift assay (22) with modifications. 0.025 palmitoyl-L-carnitine, 0.5 malate (PCM); 20 succi- Equal amounts of nuclear protein were loaded for each sam- nate, 0.1 rotenone (SR); and 10 glutamate, 5 malate ple. After incubation with poly(deoxyinosinic-deoxycytidylic) (GM). The impact of complex-related energy flux on m m 9 acid (0.05 g/ L) and double-stranded 3 -biotinylated DNA ATP synthesis was calculated as the difference in pro- probe, electrophoretic separation of nuclear extracts was duction rate induced by the addition, in a subsequent fi performed in 0.8% agarose gel. Band speci city evaluation 20-min read, of a complex-specific inhibitor during state fi and identi cation were performed by running a pooled 3 respiration on excess complex-specificsubstrate.For sample preincubated for 20 min with excess unlabeled probe complex I–related ATP synthesis, substrate and inhibitor 3 m (1,000 ), anti-p65 (2 g; Millipore), or anti-p105/p50 were GM and rotenone (2 mmol/L), whereas for complex m (2 g; Abcam) antibody. Results were calculated from II, these were SR and malonate (1 mmol/L). Mitochon- k – fi optical density of NF- B speci c bands. drial functional integrity in each preparation was con- fi . . Tissue Glucose Uptake rmed by a 80% and 95%decreaseinstate3ATP m Tissue glucose uptake was measured ex vivo with nonra- synthesis after addition of CCCP 30 mol/L and oligomycin m m dioactive 2-deoxyglucose (2-DG) (26). Extensor digitorum 2 g/ L, respectively. Values were then normalized by fi longus muscle is largely metabolically similar to gastroc- ATP synthesis rate with the nonspeci csubstratePPKM, nemius muscle (27) and was used due to smaller diameter and data are presented as the ratio between values – – and better exchange with incubation buffer (28). Two obtained for complex I related over complex II related muscle sections were incubated for 30 min at 37°C under results. constant oxygenation with or without insulin (Humulin R Statistical Analysis 600 pmol/L) in isotonic buffer (pH 7.4) in BSA (1 mg/mL) Groups were compared by using Student t test or one-way and pyruvate (2 mmol/L). After 20-min incubation with ANOVA followed by appropriate post hoc tests. Bonferroni diabetes.diabetesjournals.org Gortan Cappellari and Associates 877 correction for multiple comparisons was applied. P , 0.05 NOS-dependent but not xanthine- or NADPH oxidase– was considered statistically significant. dependent superoxide production was also reduced by UnAG (Fig. 1D–F). UnAG-treated rats also had lower mus- RESULTS cle oxidized-to-total glutathione, a marker of tissue redox Exogenous UnAG Administration state (Fig. 1G and H). Conversely, tissue protein levels of SOD isoforms and activities of antioxidant catalase Animal Characteristics and glutathione peroxidase were not modified by UnAG In lean adult rats, exogenous 4-day UnAG did not (Fig. 1I–L). modify body weight (Ct 319.6 6 3.6 g, UnAG 324.1 6 6.1 g), weight gain during treatment (Ct 13.0 6 1.4 g, fl UnAG 11.6 6 1.1 g), or caloric intake (Ct 76.9 6 2.3 kcal/day, UnAG Lowers Tissue In ammation k k UnAG 73.5 6 1.8 kcal/day). Plasma glucose (Ct 118.6 6 Protein expression of the NF- B inhibitor I Bwashigherin 6.0 mg/dL, UnAG 120.5 6 7.5 mg/dL), insulin (Ct 12.8 6 UnAG-comparedwithsaline-treatedrats,withparallelre- fl k 2.1 mU/mL, UnAG 14.3 6 2.9 mU/mL), and NEFA duction of proin ammatory NF- B p65/p50 nuclear bind- (Ct 0.27 6 0.06 mmol/L, UnAG, 0.21 6 0.03 mmol/L) ing activity (Fig. 2A and B). UnAG also increased p50/p50 concentrations were comparable between groups. homodimer binding activity (Fig. 2B), a transcription acti- vator for anti-inflammatory interleukin-10 (IL-10) (30). UnAG Lowers Skeletal Muscle ROS Production UnAG treatment consistently resulted in anti-inflammatory UnAG lowered gastrocnemius H2O2 and superoxide anion changes in muscle cytokine patterns, with higher IL-10 production rate, and this effect involved mitochondrial expression and lower proinflammatory IL-1a and tumor respiration-dependent ROS generation (Fig. 1A–C). necrosis factor-a (TNF-a) (Fig. 2C–G).

Figure 1—UnAG and skeletal muscle redox state. Effects of UnAG (200 mg subcutaneous injection twice per day) vs. saline (Ct) sustained 4-day treatment on overall (A) and speciﬁc superoxide production from mitochondrial sources in whole-tissue homogenate (B) on intact isolated mitochondrial H2O2 synthesis rate with various respiratory substrates (GMS, S, GM, and PCM) (C), and on superoxide generation from NOS (D), NADPH oxidase (E), and xanthine oxidase (F) in skeletal muscle. Effects of UnAG treatment on total (G) and oxidized (GSSG) over total (GSH: reduced) tissue glutathione (H), on protein expression of CuZnSOD (I) and MnSOD (J) with representative blots, and on enzyme activities of catalase (K) and glutathione peroxidase (GPx) (L) are also shown. Data are mean 6 SEM (n =8–10/group). *P < 0.05 vs. Ct. a.u., arbitrary units; U CS, units of citrate synthase. 878 Unacylated Ghrelin and Muscle Metabolism Diabetes Volume 65, April 2016

Figure 2—UnAG and skeletal muscle inﬂammation. Effects of UnAG (200 mg subcutaneous injection twice per day) vs. saline (Ct) sustained 4-day treatment on the expression of IkB(A), NF-kB binding activity (B) with representative blots, and tissue expression of IL-1a (C), IL-1b (D), TNF-a (E), IL-6 (F), and IL-10 (G) measured by xMAP technology in gastrocnemius muscle. Data are mean 6 SEM (n =8–10/group). *P < 0.05 vs. Ct. Ab, antibody; a.u., arbitrary units.

UnAG Enhances Insulin Signaling and Glucose Uptake not modify body weight (control diet: Ct 31.0 6 2.1 g, UnAG also led to insulin signaling activation with Tg Myh6/Ghrl 28.7 6 2.1 g; HFD: Ct 37.9 6 3.0 g, increased phosphorylation of AKTS473, GSK-3bS9, Tg Myh6/Ghrl 36.6 6 1.1 g) or caloric intake (control PRAS40T246, and P70S6KT421/S424 (Fig. 3A–F), consistent diet: Ct 13.5 6 0.1 kcal/day, Tg Myh6/Ghrl 14.6 6 with the activation of the kinase activity of both mTORC 0.6 kcal/day; HFD: Ct 17.6 6 0.1 kcal/day, Tg Myh6/Ghrl (mammalian target of rapamycin complex) complexes. 17.9 6 0.5 kcal/day) under any dietary regimen (19). Changes in insulin signaling were paralleled by higher Blood glucose (control diet: Ct 106.0 6 7.3 mg/dL, insulin-stimulated muscle glucose uptake (Fig. 3G). These Tg Myh6/Ghrl 98.2 6 8.0 mg/dL), plasma insulin (con- effects were further associated with enhanced IRS-1S312 trol diet: Ct 13.3 6 1.8 mU/mL, Tg Myh6/Ghrl 14.5 6 phosphorylation (Fig. 3B), an mTORC-dependent negative 3.1 mU/mL), and NEFA (control diet: Ct 0.32 6 0.06 mmol/L, feedback mechanism and marker for enhanced insulin sig- Tg Myh6/Ghrl 0.38 6 0.08 mmol/L) were also compara- naling (31). To determine whether activating IRS-1 phos- ble among lean groups. In contrast, blood glucose (HFD: phorylations were also enhanced, we measured pIRS-1Y612 Ct 161.9 6 30.7 mg/dL, Tg Myh6/Ghrl 102.7 6 11.6 mg/dL, and found no stimulation in UnAG-treated animals (Sup- P , 0.05 Ct vs. Tg Myh6/Ghrl) and plasma insulin (HFD: plementary Fig. 1A), further indicating that UnAG-associated Ct 25.1 6 2.2 mU/mL, Tg Myh6/Ghrl 16.5 6 1.6 mU/mL, activation of insulin signaling occurs downstream of mTORC P , 0.05 Ct vs. Tg Myh6/Ghrl) but not NEFA (HFD: complexesbutnotattheIR–IRS-1 level. Ct 0.27 6 0.05 mmol/L, Tg Myh6/Ghrl 0.32 6 0.10 mmol/L, P =notsignificant [NS]) were lower in HFD-obese Tg In Vivo Effects of UnAG Are Tissue Specific fi Myh6/Ghrl compared with both wild-type HFD ani- In liver tissue, a nonstatistically signi cant reduction in mals and lean groups (P =NS,HFDTgMyh6/Ghrlvs.lean mitochondrial superoxide production was observed. This groups). relatively minor change was not associated with altered redox state, inflammation markers, or insulin signaling (Supple- Systemic Circulating UnAG Upregulation Prevents mentary Figs. 2A–I and 3A–F), as previously shown (32,33). Obesity-Associated Hyperglycemia, Whole-Body Insulin Resistance, and Skeletal Muscle Oxidative Stress, Transgenic UnAG Overexpression and HFD-Induced Inflammation, and Impaired AKT Phosphorylation Obesity Consistent with exogenous UnAG administration, circulat- Animal Characteristics ingUnAGupregulationinTgMyh6/Ghrlwascharac- Upregulation of circulating UnAG by myocardial over- terized by lower muscle oxidized-to-total glutathione, expression of the ghrelin gene (Tg Myh6/Ghrl) (19) did lower proinflammatory tissue cytokine profile, and more diabetes.diabetesjournals.org Gortan Cappellari and Associates 879

Figure 3—UnAG and skeletal muscle insulin action. Effects of UnAG (200 mg subcutaneous injection twice per day) vs. saline (Ct) sustained 4-day treatment on the phosphorylation measured by xMAP technology of IRY1162/Y1163 (A), IRS-1S312 (B), AKTS473 (C), GSK-3bS9 (D), PRAS40T246 (E), and P70S6KT421/S424 (F) and on tissue glucose uptake (G) in gastrocnemius muscle. Data are mean 6 SEM (n =8–10/group). *P < 0.05 vs. Ct; †P < 0.05 vs. same treatment without insulin; ‡P < 0.05 vs. other treatment without insulin.

pronounced phosphorylation of AKTS473, GSK-3bS9, In Vitro Myotube Experiments T246 T421/S424 PRAS40 , and P70S6K (Fig. 4A–M). These UnAG Effects on ROS Production and Insulin Signaling effects were associated with higher insulin sensitivity Are Confirmed in C2C12 Myotubes by area under the curve for ITT-induced blood glucose Forty-eight–hour UnAG treatment of C2C12 myotubes changes (Fig. 4N–P). Obese wild-type animals were, as lowered mitochondrial ROS generation, with largely expected, hyperglycemic and insulin resistant (Fig. 4N–P). dose-dependent effects (Fig. 5A). Consistent with in vivo The obese wild-type group also had higher oxidized- data, UnAG treatment also resulted in increased activating to-total glutathione, higher proinflammatory cytokine phosphorylation of mTORC complexes–dependent insulin profile, and reduced phosphorylation of AKTS473 and signaling proteins AKTS473,GSK-3bS9,PRAS40T246,and GSK-3bS9 in gastrocnemius (Fig. 4A–M). Compared with P70S6KT421/S424. Patterns of pIRY1162/Y1163 and pIRS-1S312 lean Tg Myh6/Ghrl, obese Tg Myh6/Ghrl animals had were also comparable in C2C12 and in vivo experiments, sup- moderately higher muscle oxidized-to-total glutathione and porting the lack of activation of IR–IRS-1 (Fig. 5B–G). TNF-a,whichremained,however,lower(P , 0.05) than in obese and comparable (P = NS) to lean wild-type ani- UnAG Effects In Vitro Are Not Shared by AG mals (Fig. 4B and E). In addition, UnAG upregulation C2C12 myotubes do not appear to express the AG receptor growth hormone secretagogue receptor 1 (GHSR1) (34). To prevented obesity-associated increments (P , 0.05 vs. further exclude the possibility that UnAG-induced changes obese wild type) in muscle proinflammatory cytokines result from nonspecific activation of additional AG-regulated IL-1a and IL-1b with lower IL-6, and resulted in normal- pathways, C2C12 experiments were performed with equi- ized activating phosphorylation at AKTS473 and GSK-3bS9 molar AG concentrations. Forty-eight–hour AG incubation levels (P = NS vs. lean Tg Myh6/Ghrl). Of note, obese Tg failed to inhibit ROS production and to activate insulin Myh6/Ghrl were protected from obesity-induced hypergly- signaling, except for less pronounced enhancement of cemia and whole-body insulin resistance (Fig. 4N–P), with pGSK-3bS9 (Fig. 5A–G). both parameters superimposable to lean wild-type animals. Insulin signaling proteins upstream of AKT were not acti- UnAG Effects In Vitro Are Abolished by Silencing vated in Tg Myh6/Ghrl, with patterns of pIRS-1S312 and the Autophagy Mediator ATG5 pIRS-1Y612 comparable to those observed in exogenously In additional experiments, C2C12 myotubes were incu- treated animals (Fig. 4H and I and Supplementary Fig. 1B). bated with UnAG after genomic silencing of the autophagy 880 Unacylated Ghrelin and Muscle Metabolism Diabetes Volume 65, April 2016

Figure 4—Impact of systemic overexpression of UnAG on skeletal muscle redox state, inﬂammation, insulin signaling, and action in lean and obese mice. Effects of UnAG overexpression in Tg Myh6/Ghrl (Tg) vs. wild-type (Wt) mice fed 16 weeks with control diet (CD) or HFD on total glutathione (A) and oxidized (GSSG) over total (GSH: reduced) glutathione (B) and on tissue expression of IL-1a (C), IL-1b (D), TNF-a (E), IL-6 (F), and IL-10 (G) measured by xMAP technology in gastrocnemius muscle. Effects of UnAG overexpression on the phosphorylation of IRY1162/Y1163 (H), IRS-1S312 (I), AKTS473 (J), GSK-3bS9 (K), PRAS40T246 (L), and P70S6KT421/S424 (M) measured by xMAP technology in gastrocnemius muscle. Absolute (N), corresponding area under the curve (AUC) (O), and relative (P) blood glucose values in ITT experiments. Data are mean 6 SEM (n = 7/group). *P < 0.05 vs. Ct; †P < 0.05 vs. same genotype-CD; ‡P < 0.05 vs. other genotype-CD.

mediator ATG5. ATG5 silencing abolished UnAG activities C2C12 myotubes (Fig. 7C). Liver ATP production was not on both mitochondrial ROS production and insulin signal- modified by UnAG (Supplementary Fig. 3G). ing (Fig. 6A–H). Levels of the autophagy activation marker DISCUSSION LC3II/LC3I were also higher in HFD-obese Tg Myh6/Ghrl fi than wild-type mice (Fig. 6I). These studies led to several ndings. First, sustained UnAG administration in vivo leads to 1)lowermuscle Mitochondrial ATP Production ROS production and less oxidized tissue redox state, 2) Effects of UnAG Are Not Associated With Enhanced anti-inflammatory changes in tissue NF-kB activation and Skeletal Muscle Mitochondrial Function cytokine patterns, and 3) enhanced mTORC-dependent Consistent with previous results (22,35), UnAG-induced insulin signaling with higher insulin-stimulated muscle fl changes in redox state, in ammation, and insulin signaling glucose uptake. Second, muscle effects of UnAG are were associated not with an enhanced but, rather, with a reproduced in a model of systemic circulating UnAG lower or an unchanged ATP production rate in vivo and in upregulation with HFD-induced obesity, resulting in vitro, respectively (Fig. 7A–C). Higher skeletal muscle ATP prevention of obesity-associated hyperglycemia and whole- production was observed in obese mice compared with lean body insulin resistance. Third, UnAG effects are dose- counterparts, but UnAG upregulation was also associated dependently confirmed in myotubes; differential effects with lower ATP production rates in obese animals (Fig. 7B). of AG and UnAG are observed in vitro, indicating that UnAG modified muscle respiratory chain complex–related UnAG acts at least partly directly and independently of ATP production by shifting ATP synthesis toward complex AG-regulated pathways. Finally, UnAG effects in vitro I over complex II both in vivo and in vitro (Fig. 7D and E). are abolished by autophagy inhibition, thereby indicating Differently from UnAG, AG enhanced ATP production in mechanistic involvement of autophagy in UnAG activities. diabetes.diabetesjournals.org Gortan Cappellari and Associates 881

Figure 5—In vitro impact of UnAG on cultured myotubes. Effects of 48-h incubation with increasing concentrations of AG or UnAG vs. Ct on isolated mitochondria H2O2 synthesis rate with various respiratory substrates (GMS, S, GM, and PCM) (A) and effects of these treatments on the phosphorylation of IRY1162/Y1163 (B), IRS-1S312 (C), AKTS473 (D), GSK-3bS9 (E), PRAS40T246 (F), and P70S6KT421/S424 (G) measured by xMAP technology in C2C12 myotubes. Data are mean 6 SEM (n = 3/group). *P < 0.05 vs. Ct; †P < 0.05 vs. same hormone 0.1 mmol/L; §P < 0.05 vs. AG, same concentration; $P < 0.05 vs. other hormone 0.5 mmol/L; ¶P < 0.05 vs. other hormone 0.1 mmol/L; #P < 0.05 vs. all other groups. U CS, units of citrate synthase.

The results show that UnAG negatively regulates skele- the current model, indicating lower mitochondrial ROS tal muscle ROS production and inﬂammation, and these generation rather than enhanced antioxidant defenses as effects are indirectly supported by previous in vitro ob- a key mediator of UnAG-induced muscle antioxidant activ- servations in nonmuscle cells (14,16,19). In another ity. Since recently identifying UnAG as a potent inducer of study, UnAG reduced endothelial oxidative stress in models autophagy in cardiomyocytes and myotubes (35), we spec- of peripheral artery disease by restoring SOD expression ulate that enhanced removal of dysfunctional mitochondria (15,16). Skeletal muscle SOD expression and antioxidant could have contributed to lower tissue oxidative load enzyme activities were, however, unchanged by UnAG in in the current experimental setting. This hypothesis was 882 Unacylated Ghrelin and Muscle Metabolism Diabetes Volume 65, April 2016

Figure 6—Role of autophagy in UnAG effects on mitochondrial ROS generation and insulin signaling. Effects of autophagy mediator ATG5 genomic silencing vs. NT4 transfection on C2C12 myotubes after 48-h incubation with increasing concentrations of AG or UnAG vs. Ct on isolated mitochondria H2O2 synthesis rate with various respiratory substrates (GMS, S, GM, and PCM) (A) and cell protein expression of ATG5 after transfection with the two siRNAs (B). Effects of these treatments on the phosphorylation of IRY1162/Y1163 (C), IRS-1S312 (D), AKTS473 (E), GSK-3bS9 (F), PRAS40T246 (G), and P70S6KT421/S424 (H) measured by xMAP technology. Autophagy activation marker LC3II/ LC3I as measured by Western blot in the gastrocnemius muscle of mice with UnAG upregulation (Tg Myh6/Ghrl [Tg]) vs. wild type (Wt) fed 16 weeks with control diet (CD) or HFD (n = 7/group) (I) with representative blot. Data are mean 6 SEM (n = 3/group). *P < 0.05 vs. NT4, same UnAG concentration; †P < 0.05 vs. same siRNA, no UnAG; ‡P < 0.05 vs. other siRNA, no UnAG; $P < 0.05 vs. other siRNA, UnAG 0.1 mmol/L; #P < 0.05 vs. all other groups. U CS, units of citrate synthase.

confirmed in myotube experiments using siRNA-mediated whilenotattheIR–IRS-1 level, and these effects were autophagy inhibition. Among less quantitatively rele- paralleled by increased insulin-stimulated muscle glu- vant ROS sources (36), UnAG selectively inhibited coseuptake.Thesechangesagreewithandprovidea NOS-dependent superoxide production. This finding is molecularbasisforclinicalobservationslinkingUnAG consistent with emerging colocalization and functional with preserved whole-body insulin sensitivity in hu- interactions among NOS, nitric oxide (NO), and muscle mans (12,13). Of note, autophagy inhibition in vitro mitochondria (37,38). In particular, NO production has abolished UnAG activities on both mitochondrial ROS been reported to enhance mitochondrial ROS generation production and insulin signaling. These observations (37), whereas UnAG was reported to reduce NO produc- provide further strong support for a causal negative tion induced by proinflammatory cytokines in various impact of mitochondrial ROS production on AKT- settings (39). Potential interactions among UnAG, NO, dependent insulin signaling, which agrees with previous and mitochondrial ROS generation should be directly observations (1–5). UnAG effects were associated with investigated in future studies. enhanced inhibitory IRS-1S312 phosphorylation; this Sustained UnAG administration enhanced skeletal mus- seemingly paradoxical observation is, however, consis- cle insulin signaling downstream of mTORC complexes tent with reports of IRS-1S312 phosphorylation as a diabetes.diabetesjournals.org Gortan Cappellari and Associates 883

Figure 7—Impact of UnAG on mitochondrial ATP synthesis. Effects on muscle ATP synthesis rate with various respiratory substrates (PPKM, PCM, GM, and SR) (A) in isolated mitochondria from rat gastrocnemius muscle after UnAG (200 mg subcutaneous injection twice per day) vs. saline (Ct) sustained 4-day (n =8–10/group) treatment in isolated mitochondria from gastrocnemius muscle of mice with UnAG upregulation (Tg Myh6/Ghrl [Tg]) vs. wild type (Wt) fed 16 weeks with control diet (CD) or HFD (n = 7/group) (B) and in C2C12 myotubes after 48-h incubation with increasing concentrations of AG or UnAG vs. Ct (n = 3/treatment) (C). Complex I– over complex II–related ATP synthesis rate ratio in rat gastrocnemius muscle after sustained treatment (D) and in cultured myotubes (E). Data are mean 6 SEM. *P < 0.05 vs. Ct or Wt; †P < 0.05 vs. same genotype-CD; ‡P < 0.05 vs. other genotype-CD; §P < 0.05 vs. AG, same concentration; #P < 0.05 vs. all other groups. U CS, units of citrate synthase.

physiological negative feedback modulation following Potential mechanisms underlying differential regulation downstream signaling activation (31). of PRAS40T246 and P70S6KT421/S424 phosphorylation in Results in Tg Myh6/Ghrl mice with chronic systemic obese versus lean models of UnAG exposure should be UnAG overexposure (19) confirmed effects of exogenous investigated in future studies. Overall, results in the UnAG administration, and these results are supported by HFD-obesity model demonstrated that effects of UnAG higher insulin sensitivity in a lean UnAG adipose trans- translate into beneficial metabolic changes in a clinically genic model (17). Because plasma AG and IGF-I are un- relevant model of dietary-induced insulin resistance and changed in Tg Myh6/Ghrl (19), the current findings hyperglycemia, thereby providing a strong rationale for further confirm that UnAG effects are independent of therapeutic strategies to increase UnAG availability in changes in AG and its potential impact on growth hormone- obese, insulin resistant, and type 2 diabetic conditions. IGF-I through GHSR1 (7,19). Most importantly, circulating Myotube experiments agreed with in vivo studies by UnAG upregulation prevented HFD-induced hyperglycemia showing superimposable effects of UnAG on ROS pro- and systemic insulin resistance, whereas muscle oxidative duction and insulin signaling that were not induced stress markers, inflammation, and impaired insulin signal- by equimolar AG concentrations. These observations strongly ing were preserved overall at levels comparable with lean indicate that UnAG directly stimulates skeletal muscle Tg Myh6/Ghrl or wild-type animals. UnAG-dependent insulin signaling and are consistent with previously stimulation of muscle autophagy was also confirmed in reported UnAG signaling and antiatrophic activities vivo in HFD-fed obese Tg Myh6/Ghrl by a higher LC3II/ in skeletal muscle of both wild-type and GHSR1 knockout LC3I ratio and, therefore, could have potentially directly mice (19). These findings overall provide strong support contributed to beneficial effects of UnAG overexpression to the hypothesis that UnAG effects in skeletal muscle are (40). Of note, obese Tg Myh6/Ghrl animals showed independent of GHSR1 and are mediated by alternative, no increments in pIRS-1S312 compared with wild-type yet unidentified, UnAG receptors. Both AG and UnAG counterparts, and lack of effect was associated with also stimulate differentiation of C2C12 myoblasts (34), lack of stimulation of insulin signaling activation at and both ghrelin forms enhance mTORC2-mediated P70S6K levels. These combined observations are consis- antiatrophic signaling under acute experimental condi- tent with the hypothesis that IRS-1S312 phosphorylation tions in C2C12 myotubes as well as in vivo in skeletal is at least partly mediated by this feedback loop (31). muscle of GHSR knockout mice (19,32). In the current 884 Unacylated Ghrelin and Muscle Metabolism Diabetes Volume 65, April 2016 studies with prolonged hormone incubation, highest AG vivo. Unchanged mitochondrial ATP production in vitro, doses selectively induced a moderate increase of GSK-3bS9 however, does not support a direct role of UnAG to inhibit phosphorylation but failed to reduce ROS generation and mitochondrial function, while it further indicates that to enhance downstream insulin signaling. Also consistent reduced mitochondrial function is not a prerequisite for with these findings, AG is a weaker autophagy inducer reduced ROS generation. Of note, UnAG modified complex- than UnAG and fails to stimulate both mitophagy (35) related ATP production by favoring complex I– over com- and ischemia-induced skeletal muscle regeneration (15). plex II–related synthesis in vitro and in vivo, potentially On the basis of the available knowledge, differential muscle reflecting preferential glucose- over fat-derived substrate effects of ghrelin forms may depend on still uninvestigated oxidation (46). Because glucose-related substrate oxidation acylation-selective and time-dependent AG activities. Over- may lower mitochondrial ROS generation (47,48), this all, differential effects of ghrelin forms on muscle insulin mechanism could also contribute to inhibiting ROS pro- signaling are fully consistent with clinical observations duction. The current results warrant further studies on linking UnAG, but not AG, to whole-body insulin sen- interactions between UnAG and muscle mitochondrial sitivity in humans (12,13). function. Finally, UnAG-induced lower ROS production, lower In conclusion, these studies demonstrate a novel role of inflammation, and enhanced insulin signaling were associ- UnAG to modulate skeletal muscle redox state, inflamma- ated with reduced or unchanged ATP production. In agree- tion, and insulin signaling. UnAG-treated rat muscle is ment with previous studies, HFD-fed animals conversely characterized by lower mitochondrial ROS production, showed higher mitochondrial ATP production despite lower inflammation, and enhanced insulin signaling and higher oxidative stress markers and insulin resistance action. These effects are tissue specific, appear to be (32), and this alteration could involve enhanced substrate direct and independent of acylated hormone, and could availability through feed-forward mechanisms (32). The be at least partly mediated by UnAG-dependent stimu- current results, therefore, provide further evidence against lation of autophagy (Fig. 8). UnAG overexpression also a role for low mitochondrial function to primarily cause prevents obesity-associated hyperglycemia and systemic insulin resistance (41–45), conversely indicating UnAG as insulin resistance as well as muscle oxidative stress, in- a novel modulator of muscle mitochondrial activity with flammation activation, and impaired insulin signaling. a negative impact on both ATP and ROS production in The current findings collectively indicate UnAG as a

Figure 8—Proposed interactions between UnAG and clustered obesity-associated metabolic alterations in skeletal muscle of HFD-induced obese rodents: higher mitochondrial production of ROS, higher inflammation, and lower insulin signaling activation are normalized by chronic UnAG overexposure. The findings further demonstrate a direct effect of UnAG to lower mitochondrial ROS production through stimulated autophagy, which may directly lead to lower inflammation and enhanced insulin signaling. Potential parallel UnAG activities to directly lower inflammation and enhance insulin signaling should be investigated further. diabetes.diabetesjournals.org Gortan Cappellari and Associates 885 potential novel treatment for obesity-associated meta- 12. Barazzoni R, Zanetti M, Ferreira C, et al. Relationships between desacylated bolic alterations. and acylated ghrelin and insulin sensitivity in the metabolic syndrome. J Clin Endocrinol Metab 2007;92:3935–3940 13. Barazzoni R, Gortan Cappellari G, Semolic A, et al. Plasma total and un- Acknowledgments. The authors thank Marco Stebel and Davide Barbetta acylated ghrelin predict 5-year changes in insulin resistance. Clin Nutr 2015; (University of Trieste Animal Facility, Trieste, Italy) and Margherita De Nardo S0261-5614(15)00256-3 and Lorenza Mamolo (Department of Medicine, Surgery and Health Sciences, 14. Dieci E, Casati L, Pagani F, Celotti F, Sibilia V. Acylated and unacylated University of Trieste) for assistance with the in vivo procedures. Manuela ghrelin protect MC3T3-E1 cells against tert-butyl hydroperoxide-induced Boschelle and Chiara Matilde Boccato (Department of Medicine, Surgery and oxidative injury: pharmacological characterization of ghrelin receptor and Health Sciences, University of Trieste) are acknowledged for technical assistance possible epigenetic involvement. Amino Acids 2014;46:1715–1725 in ex vivo and in vitro experiments. 15. Togliatto G, Trombetta A, Dentelli P, et al. Unacylated ghrelin promotes Funding. This work was funded by the European Society for Clinical Nutrition skeletal muscle regeneration following hindlimb ischemia via SOD-2-mediated and Metabolism (ESPEN) through a research fellowship to G.G.C. miR-221/222 expression. J Am Heart Assoc 2013;2:e000376 Duality of Interest. No potential conflicts of interest relevant to this article 16. Togliatto G, Trombetta A, Dentelli P, et al. Unacylated ghrelin induces oxi- were reported. dative stress resistance in a glucose intolerance and peripheral artery disease Author Contributions. G.G.C. contributed to the study design, experi- mouse model by restoring endothelial cell miR-126 expression. Diabetes 2015; ments, data research and analysis, and writing and final approval of the manuscript. 64:1370–1382 M.Z. contributed to the discussion and review, editing, and final approval of the 17. Zhang Q, Huang WD, Lv XY, Yang YM. Ghrelin protects H9c2 cells from manuscript. A.S. contributed to the experiments, data analysis, discussion, and final hydrogen peroxide-induced apoptosis through NF-kB and mitochondria-mediated approval of the manuscript. P.V. and G.G. contributed to the discussion and final signaling. Eur J Pharmacol 2011;654:142–149 approval of the manuscript. G.R. and A.F. contributed to the experiments, 18. Ryter SW, Koo JK, Choi AM. Molecular regulation of autophagy and its discussion, and final approval of the manuscript. N.F. generated the transgenic implications for metabolic diseases. Curr Opin Clin Nutr Metab Care 2014;17: mice and contributed to the discussion and final approval of the manuscript. A.G. 329–337 generated the transgenic mice and contributed to the discussion and review, 19. Porporato PE, Filigheddu N, Reano S, et al. Acylated and unacylated ghrelin editing, and final approval of the manuscript. M.G. contributed to discussion and impair skeletal muscle atrophy in mice. J Clin Invest 2013;123:611–622 review, editing, and final approval of the manuscript. R.B. contributed to the study 20. Lovric J, Mano M, Zentilin L, Eulalio A, Zacchigna S, Giacca M. Terminal design, discussion, and writing and final approval of the manuscript. R.B. is the differentiation of cardiac and skeletal myocytes induces permissivity to AAV guarantor of this work and, as such, had full access to all the data in the study and transduction by relieving inhibition imposed by DNA damage response proteins. takes responsibility for the integrity of the data and the accuracy of the data Mol Ther 2012;20:2087–2097 analysis. 21. Gortan Cappellari G, Losurdo P, Mazzucco S, et al. Treatment with n-3 polyunsaturated fatty acids reverses endothelial dysfunction and oxidative stress References in experimental menopause. J Nutr Biochem 2013;24:371–379 1. Bonnard C, Durand A, Peyrol S, et al. Mitochondrial dysfunction results from 22. Barazzoni R, Zanetti M, Gortan Cappellari G, et al. Fatty acids acutely en- oxidative stress in the skeletal muscle of diet-induced insulin-resistant mice. hance insulin-induced oxidative stress and cause insulin resistance by increasing – J Clin Invest 2008;118:789 800 mitochondrial reactive oxygen species (ROS) generation and nuclear factor-kB 2. Morino K, Petersen KF, Shulman GI. Molecular mechanisms of insulin re- inhibitor (IkB)-nuclear factor-kB (NFkB) activation in rat muscle, in the absence sistance in humans and their potential links with mitochondrial dysfunction. of mitochondrial dysfunction. Diabetologia 2012;55:773–782 – Diabetes 2006;55(Suppl. 2):S9 S15 23. Barazzoni R, Zanetti M, Bosutti A, et al. Moderate caloric restriction, 3. Schenk S, Saberi M, Olefsky JM. Insulin sensitivity: modulation by nutrients but not physiological hyperleptinemia per se, enhances mitochondrial oxi- fl – and in ammation. J Clin Invest 2008;118:2992 3002 dative capacity in rat liver and skeletal muscle–tissue-specificimpacton 4. Víctor VM, Espulgues JV, Hernández-Mijares A, Rocha M. Oxidative stress tissue triglyceride content and AKT activation. Endocrinology 2005;146: and mitochondrial dysfunction in sepsis: a potential therapy with mitochondria- 2098–2106 targeted antioxidants. Infect Disord Drug Targets 2009;9:376–389 24. Rahman I, Kode A, Biswas SK. Assay for quantitative determination of 5. Wei Y, Sowers JR, Clark SE, Li W, Ferrario CM, Stump CS. Angiotensin II- glutathione and glutathione disulfide levels using enzymatic recycling method. induced skeletal muscle insulin resistance mediated by NF-kappaB activation via Nat Protoc 2006;1:3159–3165 NADPH oxidase. Am J Physiol Endocrinol Metab 2008;294:E345–E351 25. Zanetti M, Gortan Cappellari G, Burekovic I, Barazzoni R, Stebel M, Guarnieri 6. Nakazato M, Murakami N, Date Y, et al. A role for ghrelin in the central G. Caloric restriction improves endothelial dysfunction during vascular aging: regulation of feeding. Nature 2001;409:194–198 effects on nitric oxide synthase isoforms and oxidative stress in rat aorta. Exp 7. Chen CY, Asakawa A, Fujimiya M, Lee SD, Inui A. Ghrelin gene products and Gerontol 2010;45:848–855 the regulation of food intake and gut motility. Pharmacol Rev 2009;61:430–481 26. Yamamoto N, Kawasaki K, Kawabata K, Ashida H. An enzymatic fluorimetric 8. Barazzoni R, Bosutti A, Stebel M, et al. Ghrelin regulates mitochondrial-lipid assay to quantitate 2-deoxyglucose and 2-deoxyglucose-6-phosphate for in vitro metabolism gene expression and tissue fat distribution in liver and skeletal and in vivo use. Anal Biochem 2010;404:238–240 muscle [published correction appears in Am J Physiol Endocrinol Metab 2006; 27. Clerk LH, Rattigan S, Clark MG. Lipid infusion impairs physiologic insulin- 291:E428]. Am J Physiol Endocrinol Metab 2005;288:E228–E235 mediated capillary recruitment and muscle glucose uptake in vivo. Diabetes 9. Barazzoni R, Zhu X, Deboer M, et al. Combined effects of ghrelin and higher 2002;51:1138–1145 food intake enhance skeletal muscle mitochondrial oxidative capacity and AKT 28. Bonen A, Tan MH, Watson-Wright WM. Insulin binding and glucose uptake phosphorylation in rats with chronic kidney disease. Kidney Int 2010;77:23–28 differences in rodent skeletal muscles. Diabetes 1981;30:702–704 10. Delhanty PJ, Neggers SJ, van der Lely AJ. Mechanisms in endocrinology: 29. Lanza IR, Nair KS. Functional assessment of isolated mitochondria in vitro. ghrelin: the differences between acyl- and des-acyl ghrelin. Eur J Endocrinol Methods Enzymol 2009;457:349–372 2012;167:601–608 30. Cao S, Zhang X, Edwards JP, Mosser DM. NF-kappaB1 (p50) homodimers 11. Tschöp M, Smiley DL, Heiman ML. Ghrelin induces adiposity in rodents. differentially regulate pro- and anti-inflammatory cytokines in macrophages. Nature 2000;407:908–913 J Biol Chem 2006;281:26041–26050 886 Unacylated Ghrelin and Muscle Metabolism Diabetes Volume 65, April 2016

31. Hançer NJ, Qiu W, Cherella C, Li Y, Copps KD, White MF. Insulin and extracellular signal-regulated kinase 1/2, and phosphatidyl inositol 3-kinase/Akt metabolic stress stimulate multisite serine/threonine phosphorylation of insulin signaling. Endocrinology 2007;148:512–529 receptor substrate 1 and inhibit tyrosine phosphorylation. J Biol Chem 2014;289: 40. Liu Y, Palanivel R, Rai E, et al. Adiponectin stimulates autophagy and re- 12467–12484 duces oxidative stress to enhance insulin sensitivity during high-fat diet feeding 32. Barazzoni R, Zanetti M, Cattin MR, et al. Ghrelin enhances in vivo skeletal in mice. Diabetes 2015;64:36–48 muscle but not liver AKT signaling in rats. Obesity (Silver Spring) 2007;15:2614– 41. Holloszy JO. “Deﬁciency” of mitochondria in muscle does not cause insulin 2623 resistance. Diabetes 2013;62:1036–1040 33. Gortan Cappellari G, Zanetti M, Semolic A, et al. Unacylated ghrelin does not 42. Fisher-Wellman KH, Weber TM, Cathey BL, et al. Mitochondrial respiratory alter mitochondrial function, redox state and triglyceride content in rat liver in capacity and content are normal in young insulin-resistant obese humans. vivo. Clin Nutr Exp 2015;4:1–7 Diabetes 2014;63:132–141 34. Filigheddu N, Gnocchi VF, Coscia M, et al. Ghrelin and des-acyl ghrelin 43. Nair KS, Bigelow ML, Asmann YW, et al. Asian Indians have enhanced promote differentiation and fusion of C2C12 skeletal muscle cells. Mol Biol Cell skeletal muscle mitochondrial capacity to produce ATP in association with severe 2007;18:986–994 insulin resistance. Diabetes 2008;57:1166–1175 35. Ruozi G, Bortolotti F, Falcione A, et al. AAV-mediated in vivo functional 44. Sreekumar R, Unnikrishnan J, Fu A, et al. Impact of high-fat diet and an- selection of tissue-protective factors against ischaemia. Nat Commun 2015;6: tioxidant supplement on mitochondrial functions and gene transcripts in rat 7388 muscle. Am J Physiol Endocrinol Metab 2002;282:E1055–E1061 36. Barbieri E, Sestili P. Reactive oxygen species in skeletal muscle signaling. 45. Barazzoni R. Skeletal muscle mitochondrial protein metabolism and function J Signal Transduct 2012;2012:982794 in ageing and type 2 diabetes. Curr Opin Clin Nutr Metab Care 2004;7:97–102 37. Sorarù G, Vergani L, Fedrizzi L, et al. Activities of mitochondrial complexes 46. Fink BD, O’Malley Y, Dake BL, Ross NC, Prisinzano TE, Sivitz WI. Mito- correlate with nNOS amount in muscle from ALS patients. Neuropathol Appl chondrial targeted coenzyme Q, superoxide, and fuel selectivity in endothelial Neurobiol 2007;33:204–211 cells. PLoS One 2009;4:e4250 38. Brookes PS, Levonen AL, Shiva S, Sarti P, Darley-Usmar VM. Mitochondria: 47. Anderson EJ, Yamazaki H, Neufer PD. Induction of endogenous uncoupling regulators of signal transduction by reactive oxygen and nitrogen species. Free protein 3 suppresses mitochondrial oxidant emission during fatty acid-supported Radic Biol Med 2002;33:755–764 respiration. J Biol Chem 2007;282:31257–31266 39. Granata R, Settanni F, Biancone L, et al. Acylated and unacylated ghrelin 48. St-Pierre J, Buckingham JA, Roebuck SJ, Brand MD. Topology of super- promote proliferation and inhibit apoptosis of pancreatic beta-cells and human oxide production from different sites in the mitochondrial electron transport chain. islets: involvement of 39,59-cyclic adenosine monophosphate/protein kinase A, J Biol Chem 2002;277:44784–44790