AMPK- 2 is involved in exercise training-induced adaptations in insulinstimulated metabolism in skeletal muscle following high-fat diet

Transcription

1 J Appl Physiol 7: ,. First published August 7, ; doi:./japplphysiol.8.. AMPK- is involved in exercise training-induced adaptations in insulinstimulated metabolism in skeletal muscle following high-fat diet Marcia J. Abbott, and Lorraine P. Turcotte Department of Biological Sciences, Human and Evolutionary Biology Section, Dana and David Dornsife College of Arts, Letters, and Sciences, University of Southern California, Los Angeles, California; and Crean College of Health and Behavioral Sciences, Chapman University, Orange, California Submitted December ; accepted in final form August Abbott MJ, Turcotte LP. AMPK- is involved in exercise traininginduced adaptations in insulin-stimulated metabolism in skeletal muscle following high-fat diet. J Appl Physiol 7: ,. First published August 7, ; doi:./japplphysiol.8.. AMPactivated protein kinase (AMPK) has been studied extensively and postulated to be a target for the treatment and/or prevention of metabolic disorders such as insulin resistance. Exercise training has been deemed a beneficial treatment for obesity and insulin resistance. Furthermore, exercise is a feasible method to combat high-fat diet (HFD)-induced alterations in insulin sensitivity. The purpose of this study was to determine whether AMPK- activity is required to gain beneficial effects of exercise training with high-fat feeding. Wild-type () and AMPK- dominant-negative () male mice were fed standard diet (), underwent voluntary wheel running (), fed HFD, or trained with HFD ( HFD). By week 6,, irrespective of genotype, decreased blood glucose and increased citrate synthase activity in both diet groups and decreased insulin levels in HFD groups. Hindlimb perfusions were performed, and, in mice with, increased insulin-mediated palmitate uptake (76.7%) and oxidation ( -fold). These training-induced changes were not observed in the mice. With HFD, decreased palmitate oxidation (6 6%) in both and and increased palmitate uptake (%) in the with no effects on palmitate uptake in the. With, increased ERK/ and JNK/ phosphorylation, regardless of genotype. With HFD, reduced JNK/ phosphorylation, regardless of genotype, carnitine palmitoyltransferase expression in, and CD6 expression in both and. These data suggest that low AMPK- signaling disrupts, in part, the exercise training-induced adaptations in insulin-stimulated metabolism in skeletal muscle following HFD. AMPK; exercise training; fatty acid metabolism; glucose uptake; insulin resistance OBESITY IS A GROWING EPIDEMIC in our modern society, resulting in the contribution and development of numerous diseases and disorders. It is widely accepted that exercise training is beneficial for the prevention and treatment of obesity in humans (, 9). Furthermore, it has been established that exercise training, while on a high-fat diet (HFD), has the ability to restore, at least in part, insulin sensitivity in rodent models (, 6). However, the exact mechanisms that occur with exercise training and promote improvements in insulin sensitivity in either obese or high-fat-fed models of insulin resistance remain to be fully elucidated. Address for reprint requests and other correspondence: L. P. Turcotte, Dept. of Biological Sciences, Human and Evolutionary Biology Section, Dana and David Dornsife College of Letters, Arts, and Sciences, Univ. of Southern California, 6 Watt Way, PED 7, Los Angeles, CA ( turcotte@usc.edu). AMP-activated protein kinase (AMPK) has been studied extensively in both acute and chronic exercise studies (,,, ). AMPK is an energy-sensing enzyme that responds to an increase in the concentration of AMP, and possibly ADP, that occurs with physical stress, such as hypoxia or muscle contractions (8,, ). Over the past yr, a great deal of data have suggested that AMPK may have a role in metabolic regulation in health and disease (, 7, 8). In line with some other reports, we have shown that AMPK- signaling is decreased in rodent skeletal muscle following high-fat feeding and that genetic downregulation of AMPK- signaling under these conditions of diet-induced metabolic stress reduces insulin action, especially as it relates to fatty acid (FA) metabolism (,, 8, ). In contrast, it has been suggested that AMPK- may not be necessary for metabolic adaptations that occur with exercise training (), although these data were not collected under diet-induced metabolic stress. Therefore, further study aimed at elucidating the exact mechanisms by which AMPK signaling may regulate exercise-induced improvements in skeletal muscle insulin action under conditions of diet-induced metabolic stress is warranted. If AMPK is essential to obtain the beneficial effects of exercise training while on a HFD, it is unclear which signaling pathways might mediate these effects. Recently, there has been extensive speculation about the relationship of sirtuin (SIRT) and AMPK (6, 7, 9, 8). It has been hypothesized that AMPK may exert its beneficial training-induced effects in part via SIRT (6). We have shown that SIRT activity is increased in skeletal muscle of high-fat-fed transgenic mice expressing a dominant-negative () AMPK- (). Beyond the relationship with SIRT, it has yet to be determined whether AMPK signaling acts through stress kinase signaling pathways to regulate adaptations to exercise and what effects these proposed relationships may play in the prevention of insulin resistance (8, ). The purpose of this study was to determine whether AMPK- activity is necessary to measure an improvement in insulin action and substrate metabolism with voluntary wheel running during high-fat feeding conditions. An AMPK- transgenic mouse model was employed, and animals were subjected to 6 wk of voluntary wheel running while on either a standard diet () or HFD (). Hindlimb perfusion procedures were utilized to analyze insulin-mediated substrate kinetics in the skeletal muscle as a result of the experimental procedures. We hypothesized that AMPK- is a necessary signal in the restoration of insulin stimulation that occurs with voluntary wheel running under high-fat feeding conditions / Copyright the American Physiological Society 869 Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

2 87 AMPK- and Exercise Training Abbott MJ et al. MATERIALS AND METHODS Animal preparation. Male C7BL/6 ( mo old) mice were kept on a :-h light-dark cycle, and animals were randomly divided into chow diet without (;.9 kcal/g, n ) or with training (; n ) or HFD without (HFD; 6% fat,.9 kcal/g, Bio-Serv no. F8, n ) or with training ( HFD; n ) groups and were monitored for 6 wk. Mice were either wild type () or expressed an AMPK- transgene, kindly provided by M. J. Birnbaum (University of Pennsylvania, Philadelphia, PA), and individually housed. The lysine residue, essential for ATP binding and hydrolysis, on an AMPK- ca, underwent in vitro mutagenesis to arginine. The mutated cas were subcloned into an expression vector, driven on the muscle creatine kinase promoter, and were microinjected into fertilized B6SJLF/J mouse eggs. Overall, the animals have unaltered levels of AMPK-, in skeletal and cardiac muscle, compared with mice, but an inactive mutated catalytic subunit replaces the endogenous form (). Food intake and body weight were monitored every week, and blood glucose and plasma insulin were measured every wk by tail vein puncture at 6 h after the active feeding period. Blood glucose was measured using a One-Touch Ultra Glucometer (Lifescan, Milpitas, CA) and an ELISA kit was used to measure plasma insulin (ALPCO, Salem, NH). Voluntary running wheels (Bio-Serv, no. K) were placed in the and HFD groups, and activity was monitored using a Sigma Sport Bicycle odometer (Sigma Sport). At the end of the 6 wk, mice underwent hindlimb perfusion experiments. Running wheels were removed from the cages h before the perfusion protocols were initiated. All procedures for the present study were approved by the Institutional Animal Care and Use Committee at the University of Southern California. Hindlimb perfusion. On the day of the experiment, mice were anesthetized by the administration of an intraperitoneal injection of ketamine-xylazine cocktail ( mg/kg body wt). Surgical preparation for the hindlimb perfusion was performed as previously described (,, 7, ). Before placement of the perfusion catheters, IU of heparin were injected into the inferior vena cava. The mice were then euthanized with an intracardial injection of pentobarbital sodium (. mg/g body wt), catheters were immediately inserted into the descending aorta and ascending vena cava, and the hindlimbs were then washed extensively with saline. The prepared mouse was then placed in a perfusion apparatus for the experimental perfusion periods. The perfusate consisted of Krebs-Henseleit solution, % bovine serum albumin (Millipore, Billerica, MA), M albumin-bound palmitate, Ci of albumin-bound [- C]palmitate, 6 mm glucose, and U/ml insulin. The perfusate was kept at 7 C and was continuously gassed with a mixture of 9% O -% CO with arterial ph levels between 7. and 7.68 and arterial PO and PCO values were between 6 and 9 Torr, respectively. Perfusion pressures were not affected by any of the experimental conditions and averaged. 8.6, ,. 6., and mmhg in the,, HFD, and HFD groups, respectively (P.). The perfusion preparation was equilibrated for min. Perfusion flow rate was maintained at. ml/min for all groups (average:.9. ml min g of perfused muscle). Arterial and venous samples were taken at,,,, and min for further analyses. Following the completion of the -min experimental perfusion period, the gastrocnemius-soleus-plantaris muscle groups were freeze-clamped in situ with precooled aluminum clamps, removed, and stored in liquid N for further analyses. Perfusate sample analyses. Perfusate samples collected during the perfusion were analyzed to determine FA, glucose, and lactate concentrations, as well as radioactive [ C]FA and CO contents. A WAKO NEFA HC kit (WAKO Chemicals, Richmond, VA) was used to measure plasma FA concentrations spectrophotometrically. An YSI- (Yellow Springs Instruments, Yellow Springs, OH) analyzer was used to measure glucose and lactate concentrations in the collected plasma samples. Perfusate [ C]FA and CO radioactivities were measured as previously described (,, ). PCO,PO, and ph were determined by utilizing an ABL- analyzer (Radiometer America, Westlake, OH). Tissue sample preparation. For Western blot analysis, frozen muscle samples ( mg) were powdered under liquid N and homogenized in l of ice-cold RIPA buffer, as previously described (,, ). The total cell homogenate was then transferred to a microcentrifuge tube and vortexed frequently for h, whereupon the samples were centrifuged at, g at C for h. For immunoprecipitation procedures, 9 mg of powdered muscle samples were homogenized in HEPES buffer and centrifuged at, g for min. Supernatants ( g) were incubated with antibodies for AMPK- or AMPK- (Santa Cruz Biotechnology, Santa Cruz, CA) for h at C with gentle agitation (, ). Following the incubation, protein A/G agarose (Santa Cruz, SC-) was added to the tubes and incubated overnight at C with gentle agitation. The immunoprecipitates were collected by centrifugation. Pellets were washed with phosphatebuffered saline buffer, and the final supernatants were resuspended in sucrose homogenizing buffer and stored at 8 C until analysis. Protein concentrations were determined with the Bradford protein assay (BioRad, Hercules, CA). For nuclear extraction procedures, a nuclear extraction kit was used (Pierce, Rockford, IL), and the manufacturer s instructions were followed (). Briefly, mg of muscle samples were homogenized in a cytoplasmic extraction buffer (CERI). The homogenate was vortexed and incubated on ice for min at which time an additional cytoplasmic extraction buffer (CERII) was added to the tube. Following a -min centrifugation step (6, g), the recovered pellet was resuspended in a nuclear extraction buffer (NER). The suspension was incubated on ice ( min) with intermittent vortexing. The tubes were then centrifuged for min (6, g). The nuclear extract was decanted and stored at 8 C until analysis. Western blot analysis. Approximately g of protein from the total cell homogenate preparations were separated on a % gel via S-PAGE (, ). Proteins were transferred onto Immobilon-P polyvinylidene difluoride membranes and blocked with % BSA in Tween-Tris-buffered saline for h. The membrane was then incubated ( C) in % BSA in Tween-Tris-buffered saline with antibodies (:,) against phospho-acc-ser 79, total acetyl-coenzyme A (CoA) carboxylase (ACC) (Cell Signaling, Danvers, MA), carnitine palmitoyltransferase (CPT), CD6, phospho-jnk, total JNK, phospho-erk/, and total ERK/ (Santa Cruz Biotechnology, Santa Cruz, CA). Following this overnight incubation, the membranes were probed with a secondary antibody (anti-rabbit IgG; :,) raised in goats (Pierce, Rockford, IL). Blots were then washed and subjected to enhanced chemiluminescence (Pierce, Rockford, IL). Band density was quantified using Scion Image (National Institutes of Health, Bethesda, MD). All bands were compared with a control sample of nonperfused muscle and expressed as percentage of control. A Ponceau S total protein stain (Sigma, St. Louis, MO) was assessed on the membranes as a loading control, as previously described (6). Activity assays. Citrate synthase activity was measured as previously described with some modifications (). Muscle homogenates ( mg) were added to a 96-well plate containing M,=-dithiobis(-nitrobenzoic acid) and M acetyl-coa. The reaction was initiated with the addition of M oxaloacetate and was monitored in a microplate reader for min. The specific activity was calculated as the absorbance rate per minute divided by the mercaptide extinction coefficient and expressed per muscle weight. AMPK- and - activities were measured using P-ATP incorporation into SAMS peptide (Upstate Signaling, Lake Placid, NY) as described (,, ). Briefly, immunoprecipitates were added to an assay cocktail containing P-ATP and SAMS peptide. Postincubation, an aliquot was spotted onto a piece of Whatman filter paper, and all paper samples were washed with phosphoric acid followed by an acetone wash. Sample papers were analyzed for radioactivity in a Packard scintilla- J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

3 tion counter, and counts were used to calculate phosphotransferase activity. SIRT activity was measured on nuclear extracts with a commercially available histone deacetylase colorimetric activation kit (Active Motif, Carlsbad, CA) (). Briefly, nuclear extracts were added to assay buffers with assay substrates in a 96-well microplate. Trichostatin A was added to the wells to inhibit class I, II, and IV histone deacetylases, and M NAD (Sigma) was added to activate SIRT, which is NAD dependent. Following incubation (7 C, 6 min), the reaction was stopped with the addition of a developing solution. The samples were read in a microplate reader at nm. Calculations and statistics. Palmitate delivery, fractional and total palmitate uptake, and percent and total palmitate oxidation were calculated, as described previously in detail (,, ). Both percent and total FA oxidation were corrected for label fixation by using acetate correction factors (, ). The specific activity for palmitate in the arterial samples was not different between groups and did not vary over time, averaging.67.,..,.67., and.7. Ci/ mol, for the,, HFD, and HFD groups, respectively (P.). Oxygen uptake, glucose uptake, and lactate release were calculated as described (,, ). All uptake and release rates are expressed per gram of perfused hindquarter muscles of both legs, which has been previously determined to be 7% of total body weight (). GraphPad Statmate software was used to calculate sample size with a power of.8. Time effects for glucose, lactate, and FA concentrations and FA kinetic data were analyzed using a two-way ANOVA with repeated measures in each of the experimental groups, with time and genotype as the two factors. If there were no significant differences in the values measured after,,,, or min of perfusion, mean values were used for subsequent analysis. and HFD conditions were analyzed in two separate two-way ANOVAs, and the effects of training and of genotype ( vs. ) were the two factors analyzed using a two-way ANOVA (Statview.). Tukey-Kramer test for post hoc multiple comparisons was performed when appropriate. A significance level of. was chosen for all statistical methods. RESULTS Effects of transgene on physiological characteristics. Body weight, blood glucose, and plasma insulin levels for mice AMPK- and Exercise Training Abbott MJ et al. fed either the or HFD are summarized in Table. The expression of the transgene had no effects on any of the physiological characteristics measured in any of the groups. In mice, plasma insulin levels did not change over time and were not affected by voluntary wheel running. In these animals, blood glucose levels in the voluntary wheel running groups were 9 8% lower (P.) than in the untrained groups by week 6 (Table ). By week, HFD mice body weight was 8 6% higher (P.) in untrained animals than in animals that completed voluntary wheel running by (Table ). In these animals, plasma insulin levels were 7% higher (P.), and blood glucose levels were 9 % higher (P.) in the untrained groups compared with the groups starting at weeks and, respectively. Taken together, knockdown of AMPK- activity did not alter the development of insulin resistance or the beneficial effects of voluntary wheel running with high-fat feeding, as assessed by changes in body weight, plasma insulin, and blood glucose levels. There were no effects of the transgene or diet on overall food consumption throughout the 6-wk protocol. However, training elicited a % increase in food consumption with and a 9% increase with HFD. Voluntary wheel running. Within the groups, the mice averaged.6. km and spent.9.6 h running per night, whereas the mice averaged.. km and spent.8.6 h running during the first week of training (P.) (Fig., A and C). By week, the mice averaged.9. km of running per night, whereas the mice averaged.. km (P.). There were no genotype effects, in regards to time spent running, throughout the 6 wk of voluntary wheel running. In mice, citrate synthase activity increased following voluntary wheel running in both the and mice (9 and 7%, respectively) (P., Fig. E). Within the HFD groups, the mice averaged.6. km and.7. h of running per night, whereas the mice averaged Table. Body weight and blood glucose and plasma insulin levels in mice fed the standard diet and high-fat diet Week Week Week Week 6 Body weight, g Standard diet Untrained High-fat diet Untrained Blood glucose, mg/dl Standard diet Untrained High-fat diet Untrained Plasma insulin, ng/ml Standard diet Untrained High-fat diet Untrained Values are means SE; n 6 mice., wild type;, dominant negative. Blood glucose and plasma insulin levels were measured 6 h after dark period. P. compared with trained group. 87 J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

4 87 AMPK- and Exercise Training Abbott MJ et al. Fig.. Effect of AMP-activated protein kinase (AMPK)- dominant-negative () transgene on voluntary wheel running and citrate synthase activity in mice fed a standard (; A, C, and E) or high-fat diet (HFD; B, D, and F). Voluntary wheel running performance was assessed via the measurement of time spent running (A and B) and total distance run (C and D) per night in mice fed a (A and C) or HFD (B and D). Citrate synthase activity was measured in mice feda(e) or HFD (F). In A D, open squares represent wild-type () mice, and solid squares represent mice. In E and F, open bars represent mice, and solid bars represent mice. Values are means SE for untrained (n ) and (n ) mice and for trained () (n ) and (n ) mice fed the, as well as for untrained (n 6) and (n 6) mice and (n 6) and (n 6) mice fed the HFD. P., training effect. P., genotype effect. A Running Time (hours) C Running Distance (Km) E Citrate Synthase Activity (µmol/min/g) Standard Diet 6 Week 6 Week B Running Time (hours) D Running Distance (Km) F Citrate Synthase Activity (µmol/min/g) 6 Week High Fat Diet 6 Week.6. km and.7.7 h of running at the start of the training period (P., Fig., B and D). At the end of the 6-wk running protocol with HFD, the mice averaged..7 km and.. h of running per night, and the mice averaged..6 km and..6 h of running per night (P.). In HFD mice, citrate synthase activity was % lower in the mice compared with the mice (P., Fig. F), and voluntary wheel running increased citrate synthase activity in both the (9%) and (9%) mice (P.). However, citrate synthase activity remained lower (P.) in the mice following voluntary wheel running. Perfusion characteristics. To verify that perfusion conditions were similar between groups, several physiological variables were measured during the experimental perfusion period Table. Physiological characteristics of hindlimb perfusions Standard Diet Standard Diet High-fat Diet High-fat Diet Oxygen uptake, mol g h FA concentration, mol/l FA delivery, nmol min g Glucose concentration, mmol/l Lactate release, mol g h Values are means SE; n 6 mice. FA, fatty acid. Hindlimbs were perfused with U/ml insulin. J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

5 (Table ). There were no differences in oxygen uptake over time in any of the experimental groups (P.). Palmitate concentration and delivery were not significantly different at any time point in any of the groups or experimental conditions, as dictated by the protocol (P.). Arterial perfusate glucose concentration did not vary over time in any of the groups, and mean glucose concentration was not different between any of the experimental conditions (P.). Additionally, lactate release was not different across time or between any of the experimental conditions during the perfusion (P.). Enzyme activities. There were no differences in AMPK- (pmol min g ) activity in any of the experimental groups (P., Fig., A and B). As expected, AMPK- activity (pmol min g ) was decreased by the expression of the transgene in both diet groups (P., Fig., C and D). Voluntary wheel running increased (P.) AMPK- activity by 66% in mice fed and by 7% in mice fed HFD. Voluntary wheel running did not increase AMPK- activity in mice fed either the or HFD (P.). To verify functional deficits in skeletal muscle AMPK- activity, AMPK- and Exercise Training Abbott MJ et al. 87 we assessed the known downstream target of AMPK, ACC. Total ACC expression was not altered by genotype or voluntary wheel running in either diet condition (P., Fig., E and F). ACC phosphorylation was decreased (P.) by the expression of the transgene in both diet groups (Fig., E and F). Voluntary wheel running increased (P.) ACC phosphorylation in mice fed, but decreased (P.) ACC phosphorylation in mice fed HFD. Voluntary wheel running had no effect (P.) on ACC phosphorylation in mice fed either diet. Insulin-stimulated substrate exchange across the hindlimb. Insulin-stimulated glucose uptake ( mol g h ) did not vary (P.) over time in any of the experimental groups throughout the perfusion period. In mice fed, mean glucose uptake was not affected by genotype or voluntary wheel running (P., Fig. A). In mice fed HFD, voluntary wheel running decreased (P.) glucose uptake in mice (Fig. B). Insulinstimulated palmitate uptake and oxidation (nmol min g ) did not vary (P.) over time in any of the experimental groups throughout the perfusion period. Genotype decreased (P.) mean palmitate uptake by % and increased mean A Standard Diet B High Fat Diet AMPKα Activity C AMPKα Activity E pacc Ser79 ACC AMPKα Activity D AMPKα Activity F HFD pacc Ser79 ACC + HFD Fig.. Effect of AMPK- transgene and voluntary wheel running on AMPK activity and acetyl-coenzyme A carboxylase (ACC) phosphorylation in perfused hindlimb muscles of mice fed a (A, C, and E) orhfd (B, D, and F). The effect of the AMPK- transgene and voluntary wheel running on AMPK- (A and B) and AMPK- (C and D) activity and on ACC phosphorylation (E and F) was measured in perfused hindlimb muscles of mice fed either a (A, C, and E) or HFD (B, D, and F). Open bars represent mice, and solid bars represent mice. Values are means SE for untrained (n ) and (n ) mice and for (n ) and (n ) mice fed the, as well as for untrained (n 6) and (n 6) mice and (n 6) and (n 6) mice fed the HFD. Western blot data are expressed as a percentage of nonperfused control muscle from mice fed the. P., training effect. P., genotype effect. pacc Ser79 pacc Ser79 J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

6 87 AMPK- and Exercise Training Abbott MJ et al. A Standard Diet B High Fat Diet Glucose Uptake (µmol/g/hr) Glucose Uptake (µmol/g/hr) Fig.. Effect of AMPK- transgene and voluntary wheel running on insulin-mediated glucose uptake and palmitate uptake and oxidation in perfused hindlimb muscles of mice feda(a, C, and E) or HFD (B, D, and F). The effect of the AMPK- transgene and voluntary wheel running on glucose uptake (A and B) and palmitate uptake (C and D) and oxidation (E and F) was measured in perfused hindlimb muscles of mice fed either the (A, C, and E) or HFD (B, D, and F). Open bars represent mice, and solid bars represent mice. Values are means SE for untrained (n ) and (n ) mice and for (n ) and (n ) mice fed the, as well as for untrained (n 6) and (n 6) mice and (n 6) and (n 6) mice fed the HFD. P., training effect. P., genotype effect. C Palmitate Uptake (nmol/min/g) E D Palmitate Uptake (nmol/min/g) F Palmitate Oxidation (nmol/min/g) Palmitate Oxidation (nmol/min/g) palmitate oxidation by 7% in mice fed (Fig., C and E). Voluntary wheel running increased (P.) palmitate uptake by 77% and palmitate oxidation by more than twofold in the mice. There were no significant voluntary wheel running-induced changes in palmitate uptake or palmitate oxidation in the mice (P.). As we have previously reported (), insulinstimulated palmitate uptake was higher (P.) in mice fed HFD compared with mice, whereas insulin-stimulated palmitate oxidation was not affected by genotype (Fig., D and F). Voluntary wheel running increased (P.) palmitate uptake in the mice (%), but did not alter palmitate uptake in the mice (P.). Voluntary wheel running decreased palmitate oxidation in both the and mice (6 and 6%, respectively; P., Fig. F). Activation and expression of signaling molecules. In mice fed, voluntary wheel running increased ERK/ (8 %) and JNK/ (6 6%) phosphorylation, regardless of genotype (P., Fig., A and C). In mice fed HFD, ERK/ phosphorylation was % higher (P.) in than mice, but was not affected (P.) by voluntary wheel running, regardless of genotype (Fig. B). In these mice, voluntary wheel running reduced JNK/ phosphorylation ( %), regardless of genotype (P., Fig. D). Overall ERK/ content appeared to increase similarly with training in both and mice with. In mice fed, SIRT activity (pmol min g ) was not affected by genotype or voluntary wheel running (P., Fig. E). Similar to our previous reports (), in mice fed HFD, SIRT activity was higher (76%) in than mice (P., Fig. F). In mice, voluntary wheel running decreased SIRT activity such that it was not different between and mice (P.). In mice fed, CPT and CD6 protein expression were not affected by genotype or voluntary wheel running (P., Fig., A and C). In mice fed HFD, CPT expression was lower (8%) in than mice (P., Fig. B). Voluntary wheel running reduced CPT expression (6%) in the mice and CD6 expression in both (7%) and (%) mice (P., Fig. D). DISCUSSION The data from this study indicate that AMPK- activity, in skeletal muscle, is involved in training-induced adaptations in metabolism and cellular signaling. Furthermore, the impact of low AMPK- activity on training-induced adaptations is dependent on diet. Thus, in -fed mice, AMPK- downregu- J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

7 AMPK- and Exercise Training Abbott MJ et al. 87 Standard Diet High Fat Diet A perk/ B perk/ HFD + HFD perk/ C pjnk/ E SIRT Activity ERK/ pjnk pjnk perk/ D pjnk/ F SIRT Activity ERK/ pjnk pjnk Fig.. Effect of AMPK- transgene and voluntary wheel running on ERK/ and JNK/ phosphorylation and sirtuin (SIRT) activity in perfused hindlimb muscles of mice fed a (A, C, and E) orhfd (B, D, and F). The effect of the AMPK- transgene and voluntary wheel running on ERK/ (A and B) and JNK/ (C and D) phosphorylation and SIRT activity (E and F) was measured in perfused hindlimb muscles of mice fed either the (A, C, and E) or HFD (B, D, and F). SIRT activity is measured in picomoles per minute per microgram of nuclear protein. Open bars represent mice, and solid bars represent mice. Values are means SE for untrained (n ) and (n ) mice and for (n ) and (n ) mice fed the, as well as for untrained (n 6) and (n 6) mice and (n 6) and (n 6) mice fed the HFD. Western blot data are expressed as a percentage of nonperfused control muscle from mice fed the. P., training effect. P., genotype effect. P., different from all groups. lation prevented training-induced increases in FA uptake and oxidation. Conversely, we also show here that, with HFD, AMPK- activity plays a role in training-induced alterations in skeletal muscle FA uptake. However, AMPK- activity is not critical for induction of training-induced changes in FA oxidation. Surprisingly, when HFD is superimposed on low skeletal muscle AMPK- activity, insulin-stimulated glucose uptake is inhibited by voluntary wheel running. Under metabolic challenges, HFD, and voluntary wheel running, AMPK- downregulation also mediated the activity of ERK/ and SIRT, two other key metabolic regulators. Overall, our data provide evidence of the complex role of AMPK- activity in the regulation of fuel metabolism. Taken together, AMPK- activity is especially important when skeletal muscle is faced with metabolic challenges such as those induced by high-fat feeding and voluntary wheel running. One of the aims of this study was to determine the effects of downregulated AMPK- signaling in the regulation of insulin-stimulated substrate utilization following a short duration (6 wk) of exercise training, with or without HFD. We have previously reported () that downregulation of AMPK- activity impairs the ability of skeletal muscle to metabolically respond to insulin stimulation under HFD conditions. Thus we postulated that the beneficial effects of 6 wk of voluntary wheel running on insulin action would also be impacted by downregulation of AMPK- activity, especially when skeletal muscle is simultaneously challenged by HFD. To increase physical activity, a running wheel was installed in individual cages, and each mouse had free access to its own wheel. This form of voluntary wheel running has proven to be an effective mode of exercise training (, ). In line with these studies, our data show that voluntary wheel running, for 6 wk, was a significant J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

8 876 AMPK- and Exercise Training Abbott MJ et al. Fig.. Effect of AMPK- transgene and voluntary wheel running on carnitine palmitoyltransferase (CPT) and CD6 protein expression in perfused hindlimb muscles of mice fed a (A and C) or HFD (B and D). The effect of the AMPK- transgene and voluntary wheel running on CPT (A and B) and CD6 (C and D) protein expression was measured in perfused hindlimb muscles of mice fed either the (A and C) or HFD (B and D). Open bars represent mice, and solid bars represent mice. Values are means SE for untrained (n ) and (n ) mice and for (n ) and (n ) mice fed the, as well as for untrained (n 6) and (n 6) mice and (n 6) and (n 6) mice fed the HFD. Western blot data are expressed as a percentage of nonperfused control muscle from mice fed the. P., training effect. P., genotype effect. P., different from all groups. A CPT C CD6 CPT CD6 Standard Diet B CPT D CD6 CPT CD6 High Fat Diet HFD HFD + HFD + HFD metabolic stimulus, as shown by increases in citrate synthase activity. Furthermore, the voluntary wheel running protocol prevented weight gain, development of hyperglycemia, and hyperinsulinemia in high-fat-fed mice. Several factors have been shown to regulate the effects of exercise and diet on FA oxidation in skeletal muscle (, 7). However, it is generally accepted that AMPK is a central regulatory factor for FA oxidation because it inactivates ACC, which in turn decreases malonyl-coa levels, thus increasing CPT activity, the rate-limiting enzyme for entrance of FA into the mitochondria (, 7). In this regard, it has been demonstrated by us and others that AMPK activity in rodent skeletal muscle correlates positively with ACC phosphorylation and FA oxidation (9, ). In line with this model of metabolic regulation, voluntary wheel running in -fed mice increased FA oxidation that followed an increase in the activation state of AMPK- and phosphorylation of ACC. However, we have previously uncovered a dissociation of AMPK- activity and FA oxidation in skeletal muscle in the muscle (, ). Therefore, it was not surprising to observe high rates of FA oxidation in association with low AMPK- activity and low ACC phosphorylation in -fed mice. Ultimately, our results indicate that, under these metabolic conditions, the measured high rates of FA oxidation cannot be explained by the typical AMPK- /ACC axis and reinforce the notion that alternative regulatory mechanisms exist. It is well accepted that FA oxidative capacity in skeletal muscle increases with endurance training (, ), and here we show that AMPK- is required to observe a traininginduced increase in FA oxidation over the already high basal levels. Whether or not FA oxidation rates are higher or lower with obesity and/or diabetes is still under debate. As such, FA oxidation results are dependent on a variety of physiological factors, including, among others, insulin levels used during experimental measurements, degree of obesity of participants, and whether insulin resistance was nutritionally or genetically induced (, 6,, 6). We have previously shown () that, regardless of AMPK- activity, insulin-stimulated FA oxidation rates are high in skeletal muscle of mice fed a HFD for 6 wk. Here we show that AMPK- is not required to observe the reduction in insulin-stimulated FA oxidation induced by voluntary wheel running while on a HFD. The reduced FA oxidation rates, in both the and mice, can be explained, in part, by the observed reduction in CPT protein expression following voluntary wheel running with HFD. In line with these data, recent studies have suggested that CPT activity may be regulated by factors other than malonyl-coa levels to increase FA entrance into the mitochondria, such as fatty acid translocase (FAT)/CD6 (9). Here we provide indirect support for the potential synergism of CPT and FAT/CD6 in the regulation of FA entrance into the mitochondria and, ultimately, of FA oxidation. We observed decreases in FAT/CD6 protein expression following voluntary wheel running with HFD in both genotypes. Additionally, it has been suggested by some that inhibition of CPT results in a decrease in insulin-stimulated FA oxidation and a subse- J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

9 AMPK- and Exercise Training Abbott MJ et al. quent improvement in insulin action and glucose tolerance (). Therefore, these data suggest that voluntary wheel running is sufficient to correct the high-insulin-mediated FA oxidation rates induced by HFD. Furthermore, it can be postulated that these beneficial alterations in FA oxidation are regulated by the coordination of CPT and FAT/CD6, independent of AMPK- activity. Our data indicate that AMPK- is involved in the regulation of FA uptake in skeletal muscle, irrespective of diet or training status under insulin-stimulated conditions. We have previously observed () a downregulation of insulin-mediated FA uptake in perfused hindlimbs of AMPK- mice that is further supported in the present study. Here we extend these findings and show that AMPK- is a key regulator of insulinstimulated FA uptake, because exercise training was unable to induce an increase in insulin-stimulated FA uptake in mice fed either diet. Recently, it has been determined that AMPK may not be required for regulation of FA uptake via FAT/ CD6 translocation following acute muscle contraction (). These observations suggest that the role of AMPK- in regulating changes in FA uptake is more critical for traininginduced changes than alterations due to an acute bout of muscle contraction. Furthermore, exercise mode, exercise intensity, diet, and diet duration have been shown to independently impact FA uptake and cellular signaling (7,,, ). However, the relative importance of AMPK- in regulating changes in FA uptake, in the presence of insulin, may be highly dependent on the type and duration of metabolic stressors imposed on the muscle. Of note, ERK/ phosphorylation was found to be higher in mice fed the HFD. Given that ERK/ phosphorylation has been linked to the regulation of FA uptake during muscle contraction, our results suggest that ERK/ may have played a role under our experimental conditions (, ). It is interesting to note that, while some of the typical training-induced changes in insulin-mediated glucose and FA uptake and oxidation were not measured in mice with low AMPK- activity, other typical training-induced alterations were observed. Citrate synthase activity was increased by 6 wk of wheel running in both and mice fed either diet, even though the mice ran less distance than mice. With HFD and voluntary wheel running, citrate synthase activity increased, albeit not to the same absolute level as in mice, in mice. Other studies have reported low basal citrate synthase activity and low activity of some, but not all, electron transport chain complexes in other mouse models of AMPK- downregulation, and, as shown here, this reduction did not prevent a traininginduced rise in citrate synthase activity (, 7, ). This suggests that AMPK- is not obligatory to induce an increase in citrate synthase activity (, ). On the other hand, our data indicate that, under high-fat-fed conditions, AMPK- activity appears to be required to maintain citrate synthase activity levels compared with mice. Given that citrate synthase activity is often used as a marker of Krebs cycle activity, our results suggest that oxidative capacity may be reduced in skeletal muscle of HFD-fed mice. Interestingly, FA oxidation, in HFD-fed mice, was not reduced, indicating that other factors compensated for potentially low Krebs cycle activity under our experimental conditions. 877 A commonly observed negative consequence of insulin resistance induced by obesity or HFD is a decrease in insulinmediated glucose uptake and utilization (,, ). We and others have shown that downregulation of AMPK- activity does not affect insulin-mediated glucose uptake in mice fed a control diet (, ). Adding to those findings, we show here that downregulation of AMPK- activity does not impact glucose uptake, in the presence of insulin, in voluntary wheeltrained mice. However, AMPK- activity appears to be a necessary signal for the regulation of insulin-stimulated glucose uptake in trained muscle of mice fed HFD. Our data support previous reports that low AMPK- activity was associated with lower hexokinase content in red muscle of mice (). Taken together, our findings support a role for AMPK- signaling as a necessary regulatory pathway for the regulation of glucose utilization with metabolic challenges, such as voluntary wheel running and HFD. In an attempt to uncover a mechanism for the involvement of AMPK in the regulation of diet- and training-induced alterations in substrate metabolism, we examined two potential pathways, SIRT and stress-related kinases. It has been postulated that AMPK signaling may impact SIRT activity and vice versa (6, 7, 8). In line with growing evidence of a synergy between SIRT and AMPK, we previously reported that low AMPK- activity was associated with a rise in nuclear SIRT activity in mice fed a HFD (, 8). Here we show that the impact of voluntary wheel running on nuclear SIRT activity is dependent on diet. Thus voluntary wheel running did not affect nuclear SIRT activity in mice fed the, regardless of AMPK- activity. However, in mice fed HFD, voluntary wheel running returned nuclear SIRT activity to levels observed in mice. The inability of voluntary wheel running to alter SIRT activity is in line with data showing no change in SIRT activity with chronic muscle use (9). ERK/ and JNK/ are two stress-related kinases whose activation state is known to impact metabolic regulation during exercise and with training (). Our data indicate that the impact of training on ERK/ and JNK/ phosphorylation is diet dependent. Thus training increased ERK/ and JNK/ phosphorylation in mice fed the, decreased JNK/ phosphorylation, and did not alter ERK/ phosphorylation in mice fed the HFD. However, the training-induced increases in ERK/ phosphorylation in mice could be attributed to the increase in overall ERK/ content. We also show that low-muscle AMPK- activity and high-fat feeding increased ERK/ phosphorylation but did not change JNK/ phosphorylation. Overall, our data suggest that ERK/ activation, with high-fat feeding conditions, is dependent on AMPK- signaling. In summary, data presented here support the notion that low AMPK- signaling disrupts exercise training-induced adaptations in insulin-stimulated metabolism in skeletal muscle. Furthermore, disruptions in metabolic processes are amplified when the muscle is simultaneously challenged by high-fat feeding and low AMPK-. Indeed, AMPK- may mediate changes in metabolism via multiple metabolic and mitogenic regulators that include ACC, CPT, ERK/, and SIRT. Further studies are warranted to firmly establish the mechanism of AMPK- -dependent signaling in the regulation of energy metabolism that occurs with exercise training and HFDs. J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.

10 878 AMPK- and Exercise Training Abbott MJ et al. GRANTS The present study was supported by grants from the University of Southern California Women in Science and Engineering (WiSE) program, and by fellowships from the Integrative and Evolutionary Biology Program and the Gold Family Foundation. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONIBUTIONS Author contributions: M.J.A. and L.P.T. conception and design of research; M.J.A. performed experiments; M.J.A. analyzed data; M.J.A. and L.P.T. interpreted results of experiments; M.J.A. and L.P.T. prepared figures; M.J.A. and L.P.T. drafted manuscript; M.J.A. and L.P.T. edited and revised manuscript; M.J.A. and L.P.T. approved final version of manuscript. REFERENCES. Abbott MJ, Bogachus LD, Turcotte LP. AMPK deficiency uncovers time dependency in the regulation of contraction-induced palmitate and glucose uptake in mouse muscle. J Appl Physiol :,.. Abbott MJ, Constantinescu S, Turcotte LP. AMP-activated protein kinase is an essential signal in the regulation of insulin-stimulated fatty acid uptake in control-fed and high-fat-fed mice. Exp Physiol 97: 6 67,.. Abbott MJ, Edelman AM, Turcotte LP. CaMKK is an upstream signal of AMP-activated protein kinase in regulation of substrate metabolism in contracting skeletal muscle. Am J Physiol Regul Integr Comp Physiol 97: R7 R7, 9.. Bandyopadhyay GK, Yu JG, Ofrecio J, Olefsky JM. Increased malonyl-coa levels in muscle from obese and type diabetic subjects lead to decreased fatty acid oxidation and increased lipogenesis; thiazolidinedione treatment reverses these defects. Diabetes : 77 8, 6.. Bruce CR, Thrush AB, Mertz VA, Bezaire V, Chabowski A, Heigenhauser GJF, Dyck DJ. Endurance training in obese humans improves glucose tolerance and mitochondrial fatty acid oxidation and alters muscle lipid content. Am J Physiol Endocrinol Metab 9: E99 E7, Canto C, Gerhart-Hines Z, Feige JN, Lagouge M, Noriega L, Milne JC, Elliott PJ, Puigserver P, Auwerx J. AMPK regulates energy expenditure by modulating NAD metabolism and SIRT activity. Nature 8: 6 6, Cantó C, Jiang LQ, Deshmukh AS, Mataki C, Coste A, Lagouge M, Zierath JR, Auwerx J. Interdependence of AMPK and SIRT for metabolic adaptation to fasting and exercise in skeletal muscle. Cell Metab : 9,. 8. Carling D, Thornton C, Woods A, Sanders MJ. AMP-activated protein kinase: new regulation, new roles? Biochem J : 7,. 9. Chabi B, Adhihetty PJ, O Leary MFN, Menzies KJ, Hood DA. Relationship between Sirt expression and mitochondrial proteins during conditions of chronic muscle use and disuse. J Appl Physiol 7: 7 7, 9.. Durante PE, Mustard KJ, Park SH, Winder WW, Hardie DG. Effects of endurance training on activity and expression of AMP-activated protein kinase isoforms in rat muscles. Am J Physiol Endocrinol Metab 8: E78 E86,.. Eckel RH, Alberti K, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 7: 8 8,.. Fisher JS, Gao J, Han DH, Holloszy JO, Nolte LA. Activation of AMP kinase enhances sensitivity of muscle glucose transport to insulin. Am J Physiol Endocrinol Metab 8: E8 E,.. Frøsig C, Jørgensen SB, Hardie DG, Richter EA, Wojtaszewski JFP. =-AMP-activated protein kinase activity and protein expression are regulated by endurance training in human skeletal muscle. Am J Physiol Endocrinol Metab 86: E E7,.. Fujii N, Ho RC, Manabe Y, Jessen N, Toyoda T, Holland WL, Summers SA, Hirshman MF, Goodyear LJ. Ablation of AMP-activated protein kinase activity exacerbates insulin resistance induced by high-fat feeding of mice. Diabetes 7: , 8.. Hawley JA, Zierath JR. Integration of metabolic and mitogenic signal transduction in skeletal muscle. Exerc Sport Sci Rev : 8,. 6. Hegarty BD, Cooney GJ, Kraegen EW, Furler SM. Increased efficiency of fatty acid uptake contributes to lipid accumulation in skeletal muscle of high fat-fed insulin-resistant rats. Diabetes : 77 8,. 7. Hegarty BD, Turner N, Cooney GJ, Kraegen EW. Insulin resistance and fuel homeostasis: the role of AMP-activated protein kinase. Acta Physiol 96: 9, Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K, Karin M, Hotamisligil GS. A central role for JNK in obesity and insulin resistance. Nature : 6,. 9. Hou X, Xu S, Maitland-Toolan KA, Sato K, Jiang B, Ido Y, Lan F, Walsh K, Wierzbicki M, Verbeuren TJ, Cohen RA, Zang M. SIRT regulates hepatocyte lipid metabolism through activating AMP-activated protein kinase. J Biol Chem 8: 6, 8.. Hulver MW, Berggren JR, Cortright RN, Dudek RW, Thompson RP, Pories WJ, MacDonald KG, Cline GW, Shulman GI, Dohm GL, Houmard JA. Skeletal muscle lipid metabolism with obesity. Am J Physiol Endocrinol Metab 8: E7 E77,.. Ivy JL, Sherman WM, Cutler CL, Katz AL. Exercise and diet reduce muscle insulin resistance in obese Zucker rat. Am J Physiol Endocrinol Metab : E99 E, Jeppesen J, Albers PH, Rose AJ, Birk JB, Schjerling P, Dzamko N, Steinberg GR, Kiens B. Contraction-induced skeletal muscle FAT/CD6 trafficking and FA uptake is AMPK independent. J Lipid Res : 699 7,.. Jørgensen SB, Treebak JT, Viollet B, Schjerling P, Vaulont S, Wojtaszewski JFP, Richter EA. Role of AMPK in basal, training-, and AICAR-induced GLUT, hexokinase II, and mitochondrial protein expression in mouse muscle. Am J Physiol Endocrinol Metab 9: E E9, 7.. Keung W, Ussher JR, Jaswal JS, Raubenheimer M, Lam VHM, Wagg CS, Lopaschuk GD. Inhibition of carnitine palmitoyltransferase- activity alleviates insulin resistance in diet-induced obese mice. Diabetes 6: 7 7,.. Kiens B, Essen-Gustavsson B, Christensen NJ, Saltin B. Skeletal muscle substrate utilization during submaximal exercise in man: effect of endurance training. J Physiol 69: 9 78, Kraegen EW, Storlien LH, Jenkins AB, James DE. Chronic exercise compensates for insulin resistance induced by a high-fat diet in rats. Am J Physiol Endocrinol Metab 6: E E9, Lee-Young RS, Griffee SR, Lynes SE, Bracy DP, Ayala JE, McGuinness OP, Wasserman DH. Skeletal muscle AMP-activated protein kinase is essential for the metabolic response to exercise in vivo. J Biol Chem 8: 9 9, Liu Y, Wan Q, Guan Q, Gao L, Zhao J. High-fat diet feeding impairs both the expression and activity of AMPKa in rats skeletal muscle. Biochem Biophys Res Commun 9: 7 77, Merrill GF, Kurth EJ, Rasmussen BB, Winder WW. Influence of malonyl-coa and palmitate concentration on rate of palmitate oxidation in rat muscle. J Appl Physiol 8: 99 9, Mole P, Holloszy J. Exercise-induced increase in the capacity of skeletal muscle to oxidize palmitate. Proc Soc Exp Biol Med : , 97.. Mu J, Brozinick JT, Valladares O, Bucan M, Birnbaum MJ. A role for AMP-activated protein kinase in contraction- and hypoxia-regulated glucose transport in skeletal muscle. Mol Cell 7: 8 9,.. Nielsen JN, Mustard KJW, Graham DA, Yu H, MacDonald CS, Pilegaard H, Goodyear LJ, Hardie DG, Richter EA, Wojtaszewski JFP. =-AMP-activated protein kinase activity and subunit expression in exercise-trained human skeletal muscle. J Appl Physiol 9: 6 6,.. Raney MA, Turcotte LP. Regulation of contraction-induced FA uptake and oxidation by AMPK and ERK/ is intensity dependent in rodent muscle. Am J Physiol Endocrinol Metab 9: E E7, 6.. Raney MA, Turcotte LP. Evidence for the regulation of contractioninduced fatty acid oxidation via extracellular signal-regulated kinase / activation independent of changes in fatty acid uptake. Metabolism 6: 9, 7.. Röckl KSC, Hirshman MF, Brandauer J, Fujii N, Witters LA, Goodyear LJ. Skeletal muscle adaptation to exercise training: AMP-activated protein kinase mediates muscle fiber type shift. Diabetes 6: 6 69, Romero-Calvo I, Ocón B, Martínez-Moya P, Suárez MD, Zarzuelo A, Martínez-Augustin O, de Medina FS. Reversible Ponceau staining as a loading control alternative to actin in Western blots. Anal Biochem : 8,. 7. Ruderman NB, Houghton CRS, Hems R. Evaulation of the isolated perfused rat hindquarter for the study of muscle metabolism. Biochem Journal : 69 6, 97. J Appl Physiol doi:./japplphysiol.8. Downloaded from by ${individualuser.givennames} ${individualuser.surname} (7..9.7) on February, 8. Copyright American Physiological Society. All rights reserved.