epinephrine depletion exacerbates the fasting-induced...

Epinephrine depletion exacerbates the fasting-induced protein breakdownin fast-twitch skeletal muscles

Flávia A. Graça,1 Dawit A. P. Gonçalves,1 Wilian A. Silveira,1 Eduardo C. Lira,1 Valéria Ernestânia Chaves,3

Neusa M. Zanon,1 Maria Antonieta R. Garófalo,1 Isis C. Kettelhut,1,2 and Luiz C. C. Navegantes1

1Department of Physiology, Ribeirão Preto Medical School, University of São Paulo, São Paulo, Brazil; 2Department ofBiochemistry/Immunology, Ribeirão Preto Medical School, University of São Paulo, São Paulo, Brazil; and 3Laboratory ofPhysiology and Pharmacology, Federal University of São João del Rei, Divinópolis, Minas Gerais, Brazil

Submitted 14 May 2013; accepted in final form 22 October 2013

Graça FA, Gonçalves DA, Silveira WA, Lira EC, Chaves VE,Zanon NM, Garófalo MA, Kettelhut IC, Navegantes LC. Epi-nephrine depletion exacerbates the fasting-induced protein break-down in fast-twitch skeletal muscles. Am J Physiol EndocrinolMetab 305: E1483–E1494, 2013. First published October 22, 2013;doi:10.1152/ajpendo.00267.2013.—The physiological role of epi-nephrine in the regulation of skeletal muscle protein metabolismunder fasting is unknown. We examined the effects of plasma epi-nephrine depletion, induced by adrenodemedullation (ADMX), onmuscle protein metabolism in fed and 2-day-fasted rats. In fed rats,ADMX for 10 days reduced muscle mass, the cross-sectional area ofextensor digitorum longus (EDL) muscle fibers, and the phosphory-lation levels of Akt. In addition, ADMX led to a compensatoryincrease in muscle sympathetic activity, as estimated by the rate ofnorepinephrine turnover; this increase was accompanied by high ratesof muscle protein synthesis. In fasted rats, ADMX exacerbated fast-ing-induced proteolysis in EDL but did not affect the low rates ofprotein synthesis. Accordingly, ADMX activated lysosomal proteol-ysis and further increased the activity of the ubiquitin (Ub)-protea-some system (UPS). Moreover, expression of the atrophy-related Ubligases atrogin-1 and MuRF1 and the autophagy-related genes LC3band GABARAPl1 were upregulated in EDL muscles from ADMX-fasted rats compared with sham-fasted rats, and ADMX reducedcAMP levels and increased fasting-induced Akt dephosphorylation.Unlike that observed for EDL muscles, soleus muscle proteolysis andAkt phosphorylation levels were not affected by ADMX. In isolatedEDL, epinephrine reduced the basal UPS activity and suppressedoverall proteolysis and atrogin-1 and MuRF1 induction followingfasting. These data suggest that epinephrine released from the adrenalmedulla inhibits fasting-induced protein breakdown in fast-twitchskeletal muscles, and these antiproteolytic effects on the UPS andlysosomal system are apparently mediated through a cAMP-Akt-dependent pathway, which suppresses ubiquitination and autophagy.

epinephrine; fasting; muscle atrophy; protein degradation; cAMP

ACTIVATION OF THE SYMPATHETIC NERVOUS SYSTEM (SNS) results inthe targeted release of norepinephrine (NE) from peripheralnerve endings as well as the hormonal release of epinephrinefrom the adrenal medulla. The precise contribution of theneural and hormonal components to the regulation of cate-cholamine-dependent processes depends on the context andtype of the stimulus (37). It is well established that sympatheticactivity in different tissues is reduced by fasting, an effect thatmay contribute to a decrease in the metabolic rate underconditions of caloric restriction. Conversely, the adrenal me-dulla is stimulated or remains unaltered during fasting (50).

Although this stimulation of epinephrine secretion is modest, itmay contribute to substrate mobilization, particularly by facil-itating the hydrolysis of triglycerides in adipose tissue and byaiding glycogenolysis in liver and skeletal muscle (6). How-ever, the physiological role of epinephrine in the regulation ofskeletal muscle protein metabolism during fasting is unknown.

Important adaptations of skeletal muscle under fasting con-ditions include the activation of proteolysis and the reductionof protein synthesis to support liver gluconeogenesis and theenergy requirements of the organism. These adaptive changesin muscle protein metabolism are signaled by low levels ofcirculating insulin and high levels of glucocorticoids (20) andare more pronounced in fast-twitch muscles [e.g., extensordigitorum longus (EDL) and tibialis anterior] than in slow-twitch muscles (e.g., soleus) (25). When fasting is prolonged,protein catabolism decreases to protect against excessive mus-cle wasting, and energy is derived from lipolysis. However, themechanisms underlying this protein-sparing process have notbeen completely elucidated.

The ubiquitin (Ub)-proteasome system (UPS) is the mainproteolytic system activated in starvation, and it catalyzes thebulk of muscle protein degradation, especially that of myofi-brillar components (24). During starvation conditions, muscleprotein is rapidly mobilized through a common program ofchanges in gene expression, affecting �100 atrophy-relatedgenes, or atrogenes (23). Muscle-specific Ub-ligases (E3s),atrogin-1/MAFbx (muscle atrophy F-box), and MuRF1 (mus-cle RING finger-1) are atrogenes that are induced prior to theonset of muscle weight loss (24) and have been suggested asnecessary for rapid atrophy (5, 13). Recent evidence indicatesthat atrogene regulation can vary between fast and slow mus-cles. For instance, the greater sensitivity of the fast-twitchmuscles to fasting can be explained, at least in part, by theirlower content of PGC-1� (peroxisome proliferator-activatedreceptor-� coactivator 1�), known to inhibit atrogin-1 expres-sion and fiber atrophy (26, 39). Consistent with a fiber type-dependent distribution of atrogenes, Moriscot et al. (33) haveshown in mice that MuRF1 is preferentially induced in type IImuscle fibers after denervation compared with soleus musclewith mixed fiber-types.

In addition to atrogin-1 and MuRF1, autophagy-relatedgenes [microtubule-associated protein-1 light chain 3� (LC3b)and �-aminobutyric acid (GABA) receptor-associated protein-like 1 (GABARAPl1)] and an increased capacity for lysosomalproteolysis (4) have been implicated in many forms of atrophy,including fasting (51).

It is well established that the expression of atrogenes andautophagy-related genes is inhibited by the insulin-like growth

Address for reprint requests and other correspondence: L. C. C. Navegantes,Dept. of Physiology, FMRP/USP, Ribeirão Preto, Av. Bandeirantes 3900, SãoPaulo, Brazil 14049-900 (e-mail: [email protected]).

Am J Physiol Endocrinol Metab 305: E1483–E1494, 2013.First published October 22, 2013; doi:10.1152/ajpendo.00267.2013.

0193-1849/13 Copyright © 2013 the American Physiological Societyhttp://www.ajpendo.org E1483

by 10.220.32.246 on Novem

ber 7, 2017http://ajpendo.physiology.org/

Dow

nloaded from

mailto:[email protected]

http://ajpendo.physiology.org/

factor I (IGF-I)-insulin-phosphatidylinositol 3-kinase (PI3K)-Akt signaling pathway through the phosphorylation and nu-clear exclusion of forkhead box class o (Foxo) transcriptionfactors (38, 42). Although cAMP-dependent protein kinase(PKA) is the most commonly examined epinephrine effector inskeletal muscle (31), recent in vitro studies have shown that theproduction of cAMP by epinephrine in fast-twitch skeletalmuscle (3) may activate Akt and inhibit Foxo transcriptionfactors. Furthermore, the acute and chronic administration ofclenbuterol, a selective �2-agonist, to rodents suppressed thegene expression of atrogin-1 and MuRF1 in fast-twitch mus-cles following fasting (14) and denervation (15), and theseeffects were likely mediated by cAMP. Drugs that induce anincrease in the intracellular concentrations of cAMP, such asthe selective and nonselective cAMP-phosphodiesterase (PDE)inhibitors, have been shown to prevent muscle atrophy in vivo(18) and to reduce UPS activity and atrogin-1 mRNA levels inmuscles from normal and fasted rats in vitro (28). On the basisof these findings, we propose that epinephrine may negativelyregulate the fasting-induced activation of UPS through acAMP-Akt-dependent pathway.

To investigate the specific role of epinephrine in the controlof skeletal muscle protein metabolism in rodents, we have beenusing a model of surgical sympathectomy in which the adrenalmedulla is removed. These animals are completely deficient inepinephrine but do not show any changes in the activities of thelysosomal and UPS systems under basal energy conditions(35). The purpose of this study was to examine the role ofepinephrine in the setting of fasting, particularly with regard toprotein metabolism, the expression of genes and proteinsinvolved in muscle atrophy and autophagy, and Akt/Foxosignaling. We found that plasma epinephrine depletion isassociated with abnormal fasting-induced protein breakdownand atrogene expression in fast-twitch skeletal muscles, whichmay be consistent with a defect in Akt signaling.

MATERIALS AND METHODS

Animals and adrenodemedullation surgery. The incubation proce-dure required intact muscles of a sufficient thinness to allow foradequate diffusion of metabolites and oxygen; thus, fed 4-wk-oldmale Wistar rats (�100 g) were used for all experiments. A separategroup of rats (�250 g) was used for the microdialysis experiments. Thefollowing four experimental groups were used: 1) sham-operated fed rats(fed), 2) adrenodemedullated fed rats (ADMX fed), 3) sham-operatedfasted rats (fasted), and 4) adrenodemedullated fasted rats (ADMXfasted). The fasted groups of rats were left without food for 48 h. Thisanimal model of muscle atrophy was chosen because it inducessignificant muscle weight loss that is associated with a drastic increasein Ub-ligase mRNA levels. A separated group of animals was killedafter 24 or 36 h of fasting for the analysis of muscle mass andcross-sectional area of muscle fibers. ADMX was performed bilater-ally under xylazine-ketamine anesthesia (10 and 85 mg/kg body wt ip,respectively) 10 days prior to use of the rats in experiments. Eachadrenal medulla was squeezed through a nick made on its capsula. Theanimals did not require saline in their drinking water after surgery.Sham-operated rats were used as controls. Food consumption was notaltered as a result of surgery. All animals were housed in a room witha 12:12-h light-dark cycle and were given free access to water and anormal chow diet. All experiments and protocols were performed inaccordance with the ethical principles of animal research adopted bythe Brazilian College of Animal Experimentation (COBEA) and wereapproved by the Ribeirão Preto Medical School of the University of

São Paulo - Ethical Commission of Ethics in Animal Research(CETEA; no 087/2010).

Isolated skeletal muscles. The EDL and soleus muscles wererapidly dissected, weighed, and maintained at approximately theirresting length by securing the tendons in aluminum wire supports.Tissues were incubated at 37°C in Krebs-Ringer bicarbonate buffer(pH 7.4) equilibrated with 95% oxygen and 5% carbon dioxide; thebuffer contained 5 mM glucose.

Rates of protein synthesis. EDL and soleus muscles were incubatedas described above in a buffer that contained all amino acids atconcentrations similar to those of rat plasma (40). After a 1-hequilibration period, L-[U-14C]tyrosine (0.05 �Ci/ml) was added tothe replacement medium, and the muscles were incubated for anadditional 2 h. At the end of this period, the specific activity ofacid-soluble tyrosine (intracellular tyrosine pool) in each muscle wasestimated by measuring the radioactivity and the concentration oftyrosine in this pool, which was determined by the method describedby Waalkes and Udenfriend (47). After measuring the radioactivityincorporated into the protein of the same muscle, the rate of synthesiswas calculated using the specific activity of the intracellular pool oftyrosine for each muscle, assuming that there was no recycling of thelabel during the incubation period (11).

Protein degradation measurements. Briefly, overall proteolysis andthe activities of the proteolytic systems (UPS, lysosomal, and Ca�2

dependent) were measured by tracking the tyrosine released into themedium in the presence of cycloheximide (0.5 mM), which preventedprotein synthesis and reincorporation of tyrosine. Tissues were pre-incubated for 1 h and then incubated for 2 h in identical fresh medium.Because muscle cannot synthesize or degrade tyrosine, its releasereflects protein breakdown. Overall proteolysis was evaluated bymeasuring the amount of tyrosine released into the medium. Tomeasure UPS activity, muscles from one limb were incubated inconditions that prevent activation of the lysosomal (10 mM methyl-amine and 1 U/ml insulin), branched-chain amino acid-dependent(170 �M leucine, 100 �M isoleucine, and 200 �M valine), andCa2�-dependent (Ca2�-free medium with cysteine-protease inhibi-tors, including 50 �M leupeptin) proteolytic systems. In addition,muscles from the contralateral limb were incubated with the protea-some inhibitor MG132 (20 �M). For lysosomal proteolysis measure-ments, muscles from one limb were incubated in the absence ofmethylamine, insulin, and branched-chain amino acids, conditionsthat activate the lysosomal system. Contralateral muscles were incu-bated in the presence of insulin (1 U/ml), leucine (170 �M), isoleu-cine (100 �M), valine (200 �M), and methylamine (10 mM). Meth-ylamine is a weak base that is transported across the cellular mem-branes through the ammonium transporter AmtB (48) and increasesintralysosomal pH and inhibits lysosomal proteolysis. To measureCa2�-dependent proteolytic activity, muscles from one limb wereincubated in the presence of Ca2� and inhibitors of the lysosomalsystem (methylamine, insulin, and branched-chain amino acids),whereas contralateral muscles were incubated in a Ca2�-free mediumthat contained lysosomal inhibitors and cysteine-protease inhibitor (25�M leupeptin). UPS, lysosomal, and Ca2�-dependent proteolyticactivities were calculated according to the difference in releasedtyrosine between the left and right muscles. Tyrosine release wasassayed using the previously described fluorometric method (47).

Quantitative PCR. EDL and soleus muscles were harvested andimmediately frozen in liquid nitrogen. RNA was subsequently isolatedfrom individual skeletal muscles using TRIzol (Invitrogen, Carlsbad,CA). Reverse transcription of RNA to cDNA was performed using 2�g of total cellular RNA, 20 pmol of oligo(dT) primer (Invitrogen),and Advantage ImProm-II reverse transcriptase (Promega, Madison,WI). Real-time PCR was performed using an ABI 7000 sequencedetection system (Applied Biosystems, Foster City, CA), a Super-Script III Platinum SYBR Green One-Step RT-qPCR Kit with ROX(Invitrogen), and primers for rat atrogin-1 (forward 5=-GCA GAGAGT CGG CAA GTC-3= and reverse 5=-CAG GTC GGT GAT CGT

E1484 EPINEPHRINE AND PROTEOLYSIS

AJP-Endocrinol Metab • doi:10.1152/ajpendo.00267.2013 • www.ajpendo.org

by 10.220.32.246 on Novem


Dow

nloaded from


GAG-3=), MuRF1 (forward 5=-TCG ACA TCT ACA AGC AGGAA-3= and reverse 5=-CTG TCC TTG GAA GAT GCT TT-3=),GABARAPl1 (forward 5=-CCC AGT TGT GGC AGT AGA CA-3=and reverse 5=-GAC TGA TCC TGA GGC TCC TG-3=), LC3b(forward 5=-TTT GTA AGG GCG GTT CTG AC-3= and reverse5=-CAG GTA GCA GGA AGC AGA GG-3=), or cyclophilin B(forward 5=-GCA TAC AGG TCC TGG CAT CT-3= and reverse5=-CTT CCC AAA GAC CAC ATG CT-3=). The relative quantifica-tion of mRNA levels was plotted as the fold increase compared withthe values of the fed group. Transcripts of interest were normalized tocyclophilin B and RPL39 levels, and target transcript mRNA levelswere calculated using the standard curve method (8).

Western blotting analysis. EDL and soleus muscles were homog-enized in 50 mM Tris·HCl buffer (pH 7.4) containing 150 mM NaCl,1 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 1% SDS,10 mM sodium pyrophosphate, 100 mM sodium fluoride, 10 mMsodium orthovanadate, 5 �g/ml aprotinin, 1 mg/ml leupeptin, and 1mM phenylmethylsulfonyl fluoride at 4°C. The homogenate wascentrifuged at 21,000 g at 4°C for 20 min, and the supernatant wasretained. Protein content was determined using BSA as a standard(30). An equal volume of sample buffer (20% glycerol, 125 mMTris·HCl, 4% SDS, 100 mM dithiothreitol, 0.02% bromophenol blue,pH 6.8) was added to the supernatant, and the mixture was boiled.Fifty to one hundred micrograms of total protein was separated bySDS-PAGE, transferred to nitrocellulose membranes, and blottedwith anti-Akt (1:750), anti-phospho (p)-Ser473-Akt (1:750), anti-p-Thr308-Akt (1:500), anti-Foxo3a (1:500), anti-p-Ser253-Foxo3a(1:750), anti-GABARAP (1:750), anti-LC3 (1:1,000), anti-atrogin-1(1:1,000), and anti-�-actin (1:2,000) antibodies (Abs). Primary Abswere detected using peroxidase-conjugated secondary Abs [(1:1,000)for GABARAP, Foxo3a, p-Ser253-Foxo3a, Akt, p-Thr308-Akt, andp-Ser473-Akt; and 1:5,000 for the other primary Abs] and visualizedusing ECL reagents on film or an ImageQuant 350 detection system(GE Healthcare, Piscataway, NJ). Band intensities were quantifiedusing Image Lab software (v. 3.0; Bio-Rad, Hercules, CA).

Epinephrine and muscle proteolysis studies. To investigate the invitro effect of epinephrine on the rate of overall proteolysis, Ubligases and autophagy-related gene mRNA expression, soleus andEDL muscles from fed and/or fasted rats were incubated in thepresence of varying epinephrine concentrations (10�6, 10�5, or 10�4

M) using the same procedure described above.Microdialysis studies. Fasted and ADMX fasted rats (�250 g) were

anesthetized with sodium thionembutal (50 mg/kg body mass ip) andplaced on heating pads to maintain a temperature of 37°C. The tracheawas cannulated to facilitate respiration. A polyethylene catheter (PE-50; Becton-Dickinson, Franklin Lakes, NJ) was placed in the leftcarotid artery for the withdrawal of blood samples and to obtainmeasurements of blood pressure. A microdialysis probe was insertedinto the tibialis anterior muscle, and an equilibration period of 30 minwas allowed. The principle of microdialysis has been describedpreviously in detail (46). In our study, catheters of single dialysistubing (18 � 0.3 mm, Gambro, Cuprophane, 3,000 molecular masscutoff, Sweden) were glued to polyethylene tubing (standardizedlength of 50 mm). After connecting the catheter inlet to a microin-jection pump (Insight EFF 311; Insight Equipments, Ribeirão Preto,Brazil), the system was perfused with 0.5% bovine albumin, 1 mMglucose, and 50 �M tyrosine in isotonic saline at a rate of 1 �l/min.Muscle dialysate (40 �l) and arterial blood (100 �l) samples werecollected for measurements at the end of the 90-min microdialysissession. Tyrosine was assayed using a fluorometric method (47).Because skeletal muscle cannot synthesize or degrade tyrosine, itsconcentration in muscle reflects the net balance between proteinsynthesis and degradation (16). An increase in the concentration oftyrosine in the interstitium would indicate a shift in the balance towardnet degradation. An in vivo probe was used to assess recoveredtyrosine in all samples according to the internal reference calibrationtechnique (29).

Hemodynamic parameters. MEAN ARTERIAL PRESSURE. The leftcarotid artery was cannulated and connected to a pressure transducer(Braille Biomedical, São José do Rio Preto, Brazil) to measure meanarterial pressure (MAP) every 20 min throughout the experiment.

MUSCLE BLOOD FLOW (MBF). Blood flow changes around themicrodialysis probe were measured using the previously describedethanol method (17). Each experiment included 120 min of perfusionwith isotonic saline containing 0.5% bovine albumin, 1 mM glucose,and 5 mM ethanol. Dialysate samples were collected every 15 minthroughout the experiment. The ethanol concentration was determined usingthe YSI 2700 select biochemical analyzer (Yellow Springs, OH).

NE turnover rates. Muscle NE turnover rate was measured in vivoaccording to the decline in tissue NE levels following inhibition ofcatecholamine synthesis after treatment with DL-�-methyl-p-tyrosinemethyl ester (�-MT, Sigma). After 0, 6, and 12 h of �-MT intraperi-toneal administration the rats were euthanized, and the EDL andsoleus muscles were removed to determine NE content determinationby HLPC (12). Briefly, tissues were homogenized in 0.2 N perchloricacid containing the antioxidants EDTA and sodium metabisulfite.Dihydroxybenzylamine was used as internal standard. After proteinswere removed by centrifugation, catecholamines were adsorbed inalumina, eluted with 0.1 N perchloric acid, and isolated via HPLC(LC-7A, Shimadzu Instruments) with a Spherisorb ODS-2 (5 �m,Sigma-Aldrich) reversed-phase column. NE and the internal standardwere quantified using an electrochemical detector (LC-ECD-6A,Shimadzu).

Catecholamine, glucose, hormone, and muscle cAMP levels. Agroup of rats was killed by decapitation to determine the plasma levelsof catecholamines, hormones, and glucose. Plasma was stored at�70°C prior to analysis. Catecholamines were assayed using theHPLC method described above (LC-7A, Shimadzu Instruments) witha Spherisorb ODS-2 (5 �m, Sigma-Aldrich) reversed-phase column.Glucose concentrations were determined using the YSI 2700 selectbiochemical analyzer (Yellow Springs, OH). Free fatty acids (FFA)were determined by spectrophotometry using a Randox NEFA kit(Randox Laboratories, Crumlin, County Antrim, UK). Insulin andcorticosterone were measured by radioimmunoassay, and muscularcAMP levels were measured using a method based on a competitiveenzyme immunoassay system (GE Healthcare).

Skeletal muscle cross-sectional area measurements. Rats werekilled, and EDL muscles were carefully harvested, snap-frozen inisopentane, and stored in liquid nitrogen at �80°C depending on theintended experiments. Muscles were cut into 5-�m-thick transversesections with a Leica CM1850 UV cryostat at �25°C (Leica Micro-systems, Wetzlar, Germany); slices were then placed on 26 � 76-mmslides and stained with hematoxylin and eosin. Muscle fiber sizes werecalculated using ImageJ software (v. 1.45s, National Institutes ofHealth). Approximately 300 EDL muscle fibers from each rat weremeasured, and the images were captured using an optical microscope(Leica DM 2500) with a digital video camera (Leica DFC 300FX)connected to a computer.

Antibodies, drugs, and reagents. Rabbit polyclonal anti-p-Ser473-Akt, anti-p-Thr308-Akt, anti-Akt, anti-p-Ser253-Foxo3a, anti-Foxo3a,and anti-p-Ser133-CREB Abs were purchased from Cell SignalingTechnology (Danvers, MA). Rabbit anti-atrogin-1, anti-GABARAP,and anti-�-actin Abs were purchased from Santa Cruz Biotechnology(Santa Cruz, CA). Rabbit anti-LC3 Ab was purchased from Medicaland Biological Laboratories (Nagoya, Aichi, Japan). Rabbit anti-goatIgG and goat anti-rabbit IgG antibodies (secondary antibodies) werepurchased from Santa Cruz Biotechnology and Cell Signaling, respec-tively. All drugs and reagents were purchased from Sigma-Aldrich(St. Louis, MO), Thermo Scientific HyClone (Pittsburgh, PA), Invit-rogen (Carlsbad, CA), Calbiochem EMD Biosciences (La Jolla, CA),or Amersham Biosciences (Piscataway, NJ).

Calculations. The recovery factor is the permeability of the micro-dialysis catheter and is calculated from the concentration in the

E1485EPINEPHRINE AND PROTEOLYSIS


by 10.220.32.246 on Novem


Dow

nloaded from


dialysate/concentration outside the catheter. In the present study, theinternal reference technique was used for in vivo calibration ofmicrodialysis catheters. This method has been validated in othermicrodialysis studies (27, 29) and predicts that the outflux fraction ofthe labeled metabolite added in the perfusate is the same as its relativerecovery. Briefly, the in vivo recovery from the microdialysis cathe-ters was determined for all samples and was calculated according tothe ratio of the tyrosine (Tyr) concentration in the dialysate to that inthe perfusate. To calibrate the catheters, [14C]tyrosine (�2.500 dpm,0.05 mmol/l) was added to the perfusate, and the fractional extractionof radioactivity (%recovered at each period) was measured.

The following formula was used to calculate the interstitial Tyrconcentration:

�Tyr�interstitium ��Tyr�dialysate � �Tyr�perfusate�

recovery factor� �Tyr�dialysate

Changes in blood flow are expressed as the ethanol outflow/inflowratio (e.g., the ethanol concentration in the dialysate with respect tothe perfusate).

Statistical analysis. The data are presented as means SE. Mul-tiple comparisons were made using a one-way ANOVA followed bya Student-Newman-Keuls post hoc test or a two-way ANOVA fol-lowed by a Holm-Sidak post hoc test. P � 0.05 was considered torepresent significance. In the turnover experiments, the data weresubjected to ANOVA and ANCOVA, with P 0.05 as criterion ofsignificance. The slope (fractional NE turnover rate, k) of the declinein endogenous NE over time after �-MT injection was calculatedusing method of least squares. Comparison of regression line slopeswas made with ANOVA. Fractional turnover rates were comparedusing ANCOVA. NE turnover rates were calculated as the product offractional turnover rate and the endogenous NE content at time zero.

Fig. 1. Effects of adrennomedullation (ADMX)for 10 days on mass loss of EDL (A) andsoleus (B) muscle and the cross-sectionalarea (CSA) of EDL muscle fibers (C and D)at different times of fasting. Representativehematoxylin and eosin (H&E)-stained crosssections of the same region within the EDLmuscle for each experimental group. Scalebar, 50 �m (C). Values are expressed aspercent change from control. *P � 0.05,fasted vs. fed; #P � 0.05, ADMX fed vs.fed; &P � 0.05, ADMX fasted vs. fasted.

Table 1. Effects of ADMX for 10 days on tissue mass, hormone levels, and metabolic parameters in fedand 2-day-fasted rats

Fed Fasted 48 h

Sham ADMX Sham ADMX

Body mass, g 133 2 (16) 137 3 (16) 99 2* (16) 103 2 (16)EWAT, g/100 g 753 66 (7) 456 26# (7) 641 23* (7) 397 33& (7)RWAT, g/100 g 546 41 (7) 241 24# (7) 368 47* (7) 228 22& (7)Glucose, mg/dl 146 5 (8) 148 2 (8) 84 3* (7) 91 3 (10)Insulin, �U/ml 19 3 (6) 18 2 (6) ND NDCorticosterone, �U/dl 11 2 (8) 10 1 (8) 22 1* (7) 21 1 (10)FFA, nmol/ml 1.6 0.05 (7) 3.4 0.4# (7) 6.6 0.6* (7) 4.1 0.9 (7)Muscle cAMP, fmol/mg 611 45 (7) 695 58 (7) 871 60* (7) 518 38& (7)

Values are means SE; nos. in parentheses, n. ADMX, adrenomedullation; EWAT, epididymal white adipose tissue; RWAT, retroperitonial white adiposetissue; FFA, free fatty acids; ND, not detectable. *P � 0.05, Fasted vs. Fed; #P � 0.05, ADMX Fed vs. Fed; &P � 0.05 ADMX Fasted vs. Fasted.



by 10.220.32.246 on Novem


Dow

nloaded from


For comparison of turnover rates, 95% confidence intervals weredetermined as previously described (45).

RESULTS

Effects of ADMX on body, fat, and skeletal muscle mass. Thebody mass in fed rats was not affected by ADMX (Table 1).Compared with the fed sham-operated group, the EDL masswas decreased significantly by 10% in the ADMX fed group(Fig. 1, C and D), whereas soleus muscle mass did not change(data not shown). After 48 h of fasting, all animals hadsurvived, and the ADMX and sham-operated animals had bothlost a similar degree of body mass (�40%) and fat mass(�25%). As shown in Fig. 1A, notable EDL muscle loss wasobserved with fasting in the sham-operated animals (6, 8, and20% after 24, 36, and 48 h of fasting, respectively). Soleusmuscle mass in the ADMX and sham-operated groups was notchanged after 24 h of fasting, but it decreased significantly by9% after 36 and 48 h of fasting in both groups (Fig. 1B).ADMX in rats fasted for 36 h amplified the EDL muscle massloss (Fig. 1B) and the lower cross-sectional area of musclefibers (Fig. 1, C and D) but did not interfere with the effects offasting for 24 and 48 h on muscle mass in EDL (Fig. 1A).

Effects of ADMX on metabolic and hormonal parameters.ADMX in fed rats induced a marked reduction in the wet mass ofepididymal and retroperitoneal white adipose tissue as well as anincrease in plasma FFA levels, which suggests that these ratsunderwent active lipolysis (Table 1). The fasting state in sham-operated animals decreased plasma glucose levels (�40%), de-pleted insulin levels, and increased FFA levels (�3-fold), corti-costerone levels (�2-fold), and cAMP levels in the EDL muscle(�42%). Similar responses were observed in the fasted ADMXgroup compared with the fed ADMX group, with the exception ofthe fasting-induced increase in muscle cAMP levels that was notobserved in the ADMX group and the FFA that was already highin the fed ADMX group (Table 1).

Effect of ADMX on plasma catecholamines and NE turnoverin skeletal muscles. The effect of ADMX was confirmed by adrastic reduction in plasma epinephrine concentration (�90%)but not NE concentration (Fig. 2A). However, caution shouldbe taken in asserting that plasma NE levels were not affectedby ADMX, given that the blood was collected by decapitation,a stressful condition that can activate the sympathetic activityin different tissues and increase the circulating pool of NE (22).To further characterize the functional compensation of theadrenal medulla by the sympathetic nerves (44), the effects ofADMX on NE turnover in the EDL and soleus muscles werestudied in fed and fasted rats. There was no difference inbaseline sympathetic activity between the EDL and soleusmuscles in the sham group (Fig. 3). However, the rate of NEturnover in both the EDL and soleus muscles in fed rats wassignificantly increased following bilateral adrenal removal(Fig. 3, A and C), which indicates that plasma epinephrinedepletion for 10 days in rats leads to compensatory activationof sympathetic nerve terminals in skeletal muscles under basalenergy conditions. Fasting did not affect the plasma concen-trations of catecholamines (Fig. 2) but significantly decreasedsympathetic activity in the EDL (Fig. 3B). NE turnover in thesoleus tended to decrease in fasted animals, but this differencedid not attain statistical significance. In ADMX rats, the sym-

pathetic activity in both the EDL and soleus muscles wassimilarly decreased due to fasting (Fig. 3, B and D; Table 2).

Effects of ADMX on rates of protein synthesis and proteindegradation. Skeletal muscle protein metabolism in ADMXrats varied according to the nutritional state of the animals andthe muscle type examined. As expected, fasting increased thelevels of proteolysis (�30%) in the EDL (Fig. 4A) and de-creased the rate of protein synthesis in both the soleus (33%)and EDL (50%) muscles of sham-operated rats (Fig. 4B).Epinephrine depletion in fed rats did not affect the release oftyrosine in the EDL but led to an additional increase (�30%)in overall fasting-induced proteolysis (Fig. 4A). ADMX in fedrats induced a 20% increase in the rate of protein synthesis inboth the EDL and soleus muscles (Fig. 4B) but did not affectthe low rate of protein synthesis in the muscles of fasted rats(Fig. 4B). The rate of protein degradation in the soleus musclewas not affected by either fasting or ADMX (Fig. 4A).

Effects of ADMX on proteolytic pathways and markers ofatrophy and autophagy. To further investigate the mechanismsunderlying ADMX-induced exacerbation of overall proteolysisin EDL muscle under fasting conditions, the activity of pro-teolytic pathways and the mRNA and protein levels of atrophy-related Ub-ligases atrogin-1 and MuRF1 and autophagy-relatedproteins LC3 and GABARAP were assessed. The increased ratesof protein degradation observed in EDL muscle from fasted ratswere accompanied by elevated UPS activity (Fig. 5A). ADMX

Fig. 2. Effects of ADMX for 10 days on plasma concentrations of epinephrine(A) and norepinephrine (B) in fed and 2-day-fasted rats. #P � 0.05, ADMX fedvs. fed; &P � 0.05, ADMX fasted vs. fasted.



by 10.220.32.246 on Novem


Dow

nloaded from


treatment amplified the proteolytic effect of fasting on UPSactivity in EDL muscle �2-fold (Fig. 5A) but had no effect onthe fasting-induced increase in UPS activity in soleus muscle(0.196 0.015 in ADMX fasted vs. 0.177 0.011 nmolTyr·mg�1·2 h�1 in fasted rats; n � 7, P � 0.05). Furthermore,lysosomal proteolytic activity in EDL muscle from the shamgroup was not altered as a result of fasting (Fig. 5A) but wassignificantly increased by �50% in EDL muscles from ADMXfasted rats (Fig. 5A). No significant effect was observed onCa�2-dependent proteolytic activity in skeletal muscles fromany group (Fig. 5A). As shown in Fig. 5, fasting increased the

mRNA levels (Fig. 5B) and protein contents (Fig. 5, C and D)of atrogin-1, MuRF1, LC3, and GABARAP, whereas ADMXdid not affect the expression of atrophy and autophagy markersin muscles from fed rats. However, ADMX led to a furtherincrease in the mRNA levels (Fig. 5B) of atrogin-1 (�3-fold),MuRF1 (�3-fold), LC3b (�5-fold), and GABARAPl1 (�4-fold) and in the protein content (Fig. 5C) of MuRF1 (�3-fold)and the active, autophagosome membrane-bound form, LC3-II(�5-fold) in EDL muscles from fasted rats. These effects ongene expression were not accompanied by alterations in theprotein levels of atrogin-1 and GABARAP (Fig. 5, C and D).

Fig. 3. Effects of ADMX for 10 days on thedisappearance of endogenous norepineph-rine (NE) in EDL muscle (A and B) andsoleus muscle (C and D) from fed and 2-day-fasted rats. �-MT, DL-�-methyl-p-tyrosinemethyl ester. Data are presented as means SE for 6 rats at each time point. See text forthe protocol and Table 2 for fractional turn-over rates (k) and rates of NE turnover (TR).Half-time of NE disappearance (t1/2) �0.693/k.

Table 2. Effects of ADMX for 10 days on NE content, fractional turnover rate (k), turnover rate (TR), and half-time of NEdisappearance (t1/2) in EDL and soleus muscle from fed and 2-day-fasted rats

NE, ng/g k, %/h TR, ng/g·h t1/2

EDLFed

Sham 9.51 0.50 8.28 0.67 0.79 (0.68–0.90) 8.37ADMX 7.62 0.55 11.93 1.15# 0.91 (0.76–1.07) 5.81#

FastedSham 10.00 0.64 2.29 0.49* 0.23 (0.17–0.30) 30.26*ADMX 11.98 1.13 2.83 0.07 0.34 (0.30–0.38) 24.51

SoleusFed

Sham 9.26 0.57 6.80 0.97 0.63 (0.51–0.76) 10.19ADMX 9.78 0.38 10.37 0.41# 1.01 (0.93–1.48) 6.68#

FastedSham 11.28 0.94 4.45 0.99 0.50 (0.36–0.66) 15.40ADMX 10.81 1.07 3.31 0.05 0.36 (0.27–0.46) 20.90

Values are means SE. *P � 0.05. Fasted vs. Fed; #P � 0.05. ADMX Fed vs. Fed.



by 10.220.32.246 on Novem


Dow

nloaded from


In soleus, ADMX did not affect the fasting-induced increase inatrogin-1 and LC3b mRNA levels (data not shown). These datademonstrate that plasma epinephrine depletion exacerbates thefasting-induced activation of UPS and the autophagy/lyso-somal proteolytic systems in fast-twitch muscles.

Effects of ADMX on Akt/Foxo signaling. Because Akt/Foxosignaling has recently been proposed to mediate the actions ofepinephrine and �2-agonists on protein metabolism (15, 21),we analyzed the phosphorylation status of Akt and Foxo3a inEDL and soleus muscles by immunoblotting. The results indi-cated that fasting reduced the phosphorylation levels of Akt (atThr but not Ser) in EDL muscle (Fig. 6, A and B), and ADMXalso decreased the phosphorylation levels of Akt (at both sites)in EDL muscle from fed rats and further decreased the phos-phorylation levels in muscles from fasted rats (Fig. 6, A and B).Interestingly, this effect was not observed in soleus musclesfrom either ADMX fed or fasted rats (data not shown). Thephosphorylation levels of Foxo3a in EDL muscle from shamrats were also reduced as a result of fasting, but ADMXtreatment had no effect on either fed or fasted rats (Fig. 6, Aand B). Together, these findings indicate that plasma epineph-rine depletion amplifies the decrease in the Akt phosphoryla-tion levels induced by fasting in EDL muscle.

Effects of ADMX on in vivo protein metabolism balance, aktsignaling, and hemodynamic parameters. Microdialysis exper-iments in the fasted and ADMX fasted groups were conductedto verify whether plasma epinephrine depletion could affect the

in vivo protein metabolism balance, Akt signaling, muscleblood flow, and/or MAP in adult rats. In both experimentalgroups, skeletal muscle interstitial levels of tyrosine were signif-icantly higher than arterial plasma tyrosine levels (Fig. 7A),indicating a net release of tyrosine from muscles. In addition,ADMX fasted rats had significantly higher concentrations ofinterstitial tyrosine (50%) compared with fasted rats (Fig. 7A),although arterial plasma tyrosine levels did not differ between thegroups (Fig. 7A). Moreover, the difference between interstitial andarterial plasma tyrosine concentrations in skeletal muscle wassignificantly increased (�2-fold) in ADMX fasted rats comparedwith fasted rats (Fig. 7A). ADMX also decreased the phosphory-lation of Akt (at Ser473 residue) in the tibialis muscle of fastedrats relative to that in sham-operated fasted rats (Fig. 7B).MBF, estimated by the ethanol outflow/inflow concentrationratio, and MAP were not altered by ADMX treatment. Cumu-latively, these data suggest that ADMX treatment increasesprotein catabolism in fasted adult rats independently of hemo-dynamic changes, and this effect is associated with low levelsof Akt phosphorylation.

In vitro effects of epinephrine on rates of protein degrada-tion and on mRNA expression of atrogenes and autophagy-related genes. Various concentrations of epinephrine were addedto the incubation medium of EDL muscle isolated from fastedrats. As shown in Fig. 8, A and B, fasting increased the rates ofprotein degradation and the mRNA levels of atrogin-1 (�25-fold), MuRF1 (�10-fold), LC3b (�4-fold), and GABARAPl1(�6-fold). Furthermore, the addition of 10�6, 10�5, and 10�4

M epinephrine to the incubation medium reduced the rates oftotal protein degradation by 15, 30, and 25%, respectively (Fig.8A). All concentrations of epinephrine tested significantlydownregulated the fasting-induced increase of atrogin-1mRNA levels, whereas the MuRF1 mRNA levels returned tobasal levels only in response to the highest concentration of thehormone (10�4 M; Fig. 8B). LC3b and GABARAPl1 expres-sion in muscles from fasted rats incubated with epinephrinetended to be at a lower level, although this decrease in expres-sion was not statistically significant compared with fasted ratmuscle incubated in the absence of epinephrine. In addition,10�6, 10�5, and 10�4 M epinephrine inhibited the fasting-induced activity of UPS in EDL muscle by 50, 35, and 28%,respectively, whereas the highest concentration of epinephrine(10�4 M) did not interfere with this proteolytic activity insoleus muscle (0.123 0.018 in fasted�ADR vs. 0.144 0.012 nmol Tyr·mg�1·2 h�1 in fasted rats, n � 7, P � 0.05).These in vitro results are consistent with the in vivo data anddemonstrate that epinephrine is able to directly suppress theUPS activity and the expression of atrogin-1 and MuRF1mRNA in EDL muscle induced as a result of 48 h of fooddeprivation.

DISCUSSION

This study is the first to show that chronic depletion ofepinephrine in juvenile and adult rats exacerbates long-termfasting-induced protein catabolism in fast-twitch skeletal mus-cles. In accord with previous studies (20, 49), our data dem-onstrate that the loss of muscle mass following 48 h of fastingin sham-operated rats was due to the activation of proteolysisand a reduction in protein synthesis. These well-known cata-bolic events were more pronounced in the EDL than in the

Fig. 4. Effects of ADMX for 10 days on overall proteolysis (A) and proteinsynthesis (B) in EDL and soleus muscle from fed and 2-day-fasted rats. Dataare presented as means SE of 7 muscles. *P � 0.05, fasted vs. fed; #P �0.05, ADMX fed vs. fed; &P � 0.05, ADMX fasted vs. fasted.



by 10.220.32.246 on Novem


Dow

nloaded from


soleus muscle and were associated in EDL with the suppres-sion of muscle sympathetic activity, as estimated by NEturnover. Although these data correlate well with the loss ofmuscle mass that was higher in EDL than in soleus, they do notallow us to make any conclusion that the fall in the EDLsympathetic activity might contribute to the muscle wasting.One possible rationale for such a pattern of sympathoadrenalresponse is that the reduction in sympathetic activity in fast-twitch muscles reduces the rate of energy utilization, whereasthe maintenance of adrenomedullary secretion and cAMP in-crease sustains essential catecholamine-dependent processes,such as lypolysis and glycogenolysis, at a lower net energy cost

(22). However, the role of epinephrine in the control of muscleprotein metabolism in the setting of starvation is not known.We observed that the plasma depletion of epinephrine byADMX in fasted rats amplified overall proteolysis and furtherincreased the activity of UPS. Epinephrine depletion alsoincreased the mRNA levels of atrogin-1 and MuRF1 and led toelevated MuRF1 protein levels in EDL muscle. In contrast toMuRF1, the protein content of atrogin-1 was not affected byADMX, most likely because of the inherent instability of itsmRNA. Indeed, atrogin-1 mRNA is relatively short-lived (half-life of 1 h), whereas MuRF1 mRNA is relatively long-lived(41). The amplification of fasting-induced catabolic effects on

Fig. 5. Effects of ADMX for 10 days onproteolytic activities of the ubiquitin (Ub)-proteasome system (UPS), lysosomal, andCa2�-dependent proteolytic systems (A),mRNA expression levels (B), and proteinexpression levels (C) in EDL muscles fromfed and 2-day-fasted rats. Proteolytic activitywas estimated as the difference between theamount of tyrosine released from musclesincubated in the presence or absence of spe-cific proteolytic system inhibitors (see MA-TERIALS AND METHODS). Gene expressionlevels were analyzed using cyclophilin B asan endogenous control. For immunoblotting,membranes were stripped and reprobed for�-actin as a loading control. Densitometryvalues (C) and representative immunoblots(D) are shown in top and botton, respectivelyData are presented as means SE of 7muscles. *P � 0.05, fasted vs. fed; &P �0.05, ADMX fasted vs. fasted.

Fig. 6. Effects of ADMX for 10 days onphosphorylated and total protein levels ofAkt (p-Ser473 and p-Thr308) and Foxo3a (p-Ser253) in EDL muscle from fed and 2-day-fasted rats. Membranes were stripped andreprobed for �-actin as a loading control.Representative immunoblots (A) and densi-tometry values (B) are shown at left andright, respectively. Data are presented asmeans SE of 6 muscles. *P � 0.05, fastedvs. fed; #P � 0.05, ADMX fed vs. fed; and&P � 0.05, ADMX fasted vs. fasted.



by 10.220.32.246 on Novem


Dow

nloaded from


proteolysis in ADMX rats occurred without any changes inmuscle blood flow, as estimated by microdialysis, hypoglyce-mia, or plasma insulin and glucocorticoid levels. Thus, thecatabolic effects were likely the direct consequence of a re-duction in plasma epinephrine levels. This mechanism is con-sistent with the inhibition of fasting-induced proteolysis andatrogene expression observed upon the addition of epinephrinein vitro and with the repeatedly demonstrated inhibitory effectof epinephrine and �2-agonists on muscle proteolysis in hu-mans (31) and rodents (14). The inhibitory effects of epineph-rine on UPS seem to be mediated by cAMP, and similar effectson overall proteolysis have been observed with dibutyryl-cAMP, which is a cAMP analog (35), and 6-BNZ-cAMP,which is a selective PKA agonist (15). This mechanism is alsoconsistent with the present finding that muscle cAMP levels,which were already high in the fasted group, were significantlyreduced by ADMX as well as being consistent with previousstudies showing that the increase in cAMP levels that isinduced by cAMP-PDE inhibitors prevents the fasting-inducedincrease in UPS in isolated muscles (3, 24) and blocks dexa-methasone-induced atrogin-1 expression in cultured musclecells (14). Other in vivo studies have shown that daily admin-istration of either clenbuterol or pentoxifylline prevents muscleatrophy by suppressing increased UPS activity in Yoshidasarcoma-bearing (9), diabetic (1), and denervated rats (15). Thepresent data also show that removal of the medulla in fastedrats not only amplified the UPS activity but also resulted inhyperactivation of the lysosomal system (Fig. 5A), which wasaltered neither in sham-operated fasted rats nor in ADMX fed

rats. This effect was accompanied by the additional upregula-tion of the autophagy-related genes LC3b and GABARAPl1and the active form of LC3, LC3-II. A similar increase in UPSactivity and the hyperactivation of lysosomal proteolysis bychemical sympathectomy have been previously reported formuscles from diabetic rats (2). Collectively, these findingssuggest that epinephrine plays a role in suppressing UPS andlysosomal/autophagic activity during fasting conditions viacAMP/PKA signaling, most likely through a decrease in atro-gene expression. It is noteworthy that the EDL of ADMX ratsdid not demonstrate further atrophy after 48 h of food depri-vation even when confronted with increased UPS activity andatrogene expression. However, we observed that the plasmadepletion of epinephrine by ADMX in fasted rats for 36 hamplified the muscle mass loss, suggesting that the EDLmuscle loss upon 48 h had reached maximum values in ADMXfasted and sham fasted animals.

It is well established that insulin negatively regulates theinduction of UPS and lysosomal/autophagic activities via Aktsignaling and the inactivation of Foxo3a and Foxo1 (38, 42).More recently, we and others have demonstrated that �2-adrenergic agonist treatment activates Akt and its downstreamtargets to induce muscle growth (21) and suppress muscle lossduring unloading and denervation (15, 21). This muscle-spar-ing effect occurs through the inhibition of lysosomal and UPSproteolysis, which is accompanied by the suppression of genescontrolling atrophy (15). Thus, we hypothesized that the in-hibitory effect of epinephrine on atrogenes is mediated throughcAMP, which leads to the activation of Akt and the inhibition

Fig. 7. Effects of ADMX for 10 days oninterstitial, arterial plasma, interstitial minusarterial (I � A) plasma tyrosine concentra-tions (A), phosphorylated and total proteinlevels of Akt (p-Ser473; B), ethanol outflow/inflow concentration ratio (C) in skeletalmuscle, and mean arterial pressure (MAP;D). Microdialysis catheters were insertedinto tibialis anterior muscles from 2-day-fasted and ADMX fasted rats (�220 g) andperfused with saline containing ethanol (5mmol/l) for 90 min. Following, muscleswere harvested for immunoblotting. Data arepresented as means SE of 7 rats. &P �0.05, ADMX fasted vs. fasted.



by 10.220.32.246 on Novem


Dow

nloaded from


of the forkhead transcriptional program. Accordingly, the lowlevels of phosphorylated Akt (at a Thr residue) in muscles fromfasted rats were further reduced by ADMX (Fig. 6). Thisadditional decrease in Akt phosphorylation was a direct con-sequence of plasma epinephrine depletion, which was indi-cated in previous reports. First, epinephrine addition in vitrowas shown not only to induce basal Akt and Foxo3a phosphor-ylation but also to dramatically amplify insulin-stimulatedphosphorylation of these intermediates in isolated EDL muscle(3) and cardiomyocytes (43). Second, Akt phosphorylation hasbeen observed in muscles from clenbuterol-injected wild-typefed mice but not in �2-adrenoceptor knockout mice (14). Thesedata indicate that insulin and epinephrine use similar mecha-nisms, likely via Akt, to exert inhibitory effects on UPS andlysosomal/autophagic pathways and suggest that the two hor-mones interact to control these proteolytic processes. Althoughour data show that ADMX treatment did not affect the lowphosphorylation levels of Foxo3a in EDL muscle from fastedrats, we cannot conclude that epinephrine suppresses Ub-ligaselevels via a Foxo-independent mechanism. It is likely that afurther decrease in phosphorylated Foxo occurred prior to the48 h of fasting, and this supposition is supported by the findingthat the phosphorylated levels of Akt and Foxo3a increased inmuscles from fasted mice as soon as 4 h after clenbuterolinjection (14).

Despite the recognized importance of Akt signaling in theregulation of growth and protein synthesis (10, 34), our dataindicate that the fasting-induced decrease in protein synthesis

was comparable between sham-operated and ADMX rats.These data are difficult to interpret, given that the rate ofmuscle protein synthesis in fed animals was already increasedby ADMX. The observed increase cannot be explained byincreased insulin sensitivity because p-Akt levels in the EDLwere decreased by ADMX. Because sympathetic innervationhas been shown to stimulate the basal rate of protein synthesisin skeletal muscles (36), we hypothesized that acceleration ofNE turnover after ADMX could account for the higher rate ofprotein synthesis in the muscles of fed animals. In fact, thepresent data demonstrate for the first time that noradrenergicnerves of skeletal muscles are activated by ADMX in fedanimals as a compensatory mechanism. These findings alignwith those of previous studies that found similar results in othertissues, including the pancreas and brown adipose tissue (44);our results also support the finding that ADMX fed rats exhibitan increase in active lipolysis, thus providing indirect evidenceof increased noradrenergic activity in white adipose tissue.Notwithstanding the increased protein synthesis, plasma epi-nephrine depletion in fed rats led to a decrease in the mass andcross-sectional area of the EDL, which is consistent with aprevious observation that Gs�-deficient mice developed mus-cle atrophy (7). Despite these effects, muscle cAMP levels inADMX fed rats were similar to those in sham fed rats. Perhapsthis lack of effect could be explained by the compensatorysympathetic activity induced by ADMX that was able to rescuecAMP levels but not p-Akt and muscle atrophy. The findingthat basal levels of Akt phosphorylation were also reduced in

Fig. 8. In vitro effects of various epinephrine(EPI) concentrations on overall proteolysis(A) and mRNA expression of atrogenes (B)in EDL and on UPS activity in EDL fromfasted rats (C). Muscles from 2-day-fastedrats were isolated and incubated with epi-nephrine for 2 h. Gene expression levelswere analyzed using RPL39 as an endoge-nous control. Data are presented as means SE of 6 muscles. *P � 0.05, fasted vs. fed;†P � 0.05, fasted � EPI vs. fasted.



by 10.220.32.246 on Novem


Dow

nloaded from


the EDL under basal conditions reinforces the hypothesis thatepinephrine regulates protein metabolism through Akt signal-ing. The finding that the protein degradation rate was notaltered by ADMX treatment in fed rats does not rule out thepossibility that an increase in proteolysis and/or a decrease inprotein synthesis occurred prior to the end of the 10-day fast,which was the only time interval examined in the presentstudy. Indeed, we have previously shown that ADMX treat-ment for 3 days induces a rapid and reversible increase inoverall proteolysis (35).

One interesting finding of this study was that most of thedeleterious effects of ADMX on proteolysis and Akt signalingoccurred only in EDL muscle and not in soleus muscle. UnlikeEDL, soleus muscle from fasted animals did not exhibit anamplification of proteolytic activities or a defect in Akt signal-ing after ADMX. Furthermore, epinephrine did not decreasethe UPS activity in soleus muscle from fasted rats, and ADMXdid not affect the muscle mass or the phosphorylation levels ofAkt in soleus muscle from fed animals. It is well establishedthat all muscles do not respond identically to stimulation byepinephrine or �-adrenergic agonists. For instance, fast-twitchmuscles are more responsive to the hypertrophic effects of�-adrenoceptor agonists than slow-twitch muscles (31). Thisconclusion is consistent with a previous report indicating thatAkt is activated by �2-agonists in EDL muscle but not in soleusmuscle (15). Furthermore, it is likely that this lack of Aktresponsiveness in soleus muscle results from the activity ofPKA, which presents approximately two- to threefold higherexpression and activity in the soleus than in the EDL muscle(19) and also inhibits Akt phosphorylation (3, 32).

In summary, the present data suggest that epinephrine re-leased from the adrenal medulla may be of particular impor-tance for decreasing protein breakdown in fast-twitch skeletalmuscles during long-term fasting. These antiproteolytic effectson the UPS and lysosomal systems in the setting of fasting arelikely mediated via a cAMP-Akt-dependent pathway and maylead to the suppression of ubiquitination and autophagic flux.Finally, we show there is a functional compensatory mecha-nism between the sympathetic nerves and the adrenal medullain skeletal muscles under basal conditions. Therefore, studiesthat do not take into account the relative importance of sym-pathetic nerves and adrenal medulla may underestimate thecontributions of catecholamines in a given situation.

ACKNOWLEDGMENTS

We thank Sebastião Lázaro Brandão Filho (Laboratory of Endocrinology,HCFMRP-USP) for the determination of plasma corticosterone and AnaCláudia Mattiello Sverzut for providing the research laboratory facilities at theDepartment of Pathology, FMRP-USP. We are also indebted to Dr. ClaudiaFerreira da Rosa Sobreira, Danilo Lustrino, Elza Aparecida Filippin, LilianZorzenon, and Victor Diaz Galban for technical assistance.

GRANTS

This work was supported by grants from the Fundação de Amparo aPesquisa do Estado de São Paulo (Fapesp 08/06694-6, 09/07584-2, 10/11083-6, 10/11015-0, and 12/24524-6) and from the Conselho Nacional dePesquisa (CNPq 140094/07-5, 306101/09-2, 303786/08-6, and 305149/2012-1).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: F.A.G., D.A.G., W.A.S., I.C.K., and L.C.N. concep-tion and design of research; F.A.G., D.A.G., W.A.S., E.C.L., N.M.Z., and

M.A.R.G. performed experiments; F.A.G., E.C.L., V.E.C., and L.C.N. ana-lyzed data; F.A.G., D.A.G., V.E.C., I.C.K., and L.C.N. interpreted results ofexperiments; F.A.G. and D.A.G. prepared figures; F.A.G. and L.C.N. draftedmanuscript; F.A.G., D.A.G., W.A.S., E.C.L., V.E.C., I.C.K., and L.C.N. editedand revised manuscript; F.A.G., D.A.G., W.A.S., and L.C.N. approved finalversion of manuscript.

REFERENCES

1. Baviera AM, Zanon NM, Carvalho Navegantes LC, Migliorini RH, doCarmo Kettelhut I. Pentoxifylline inhibits Ca2�-dependent and ATPproteasome-dependent proteolysis in skeletal muscle from acutely diabeticrats. Am J Physiol Endocrinol Metab 292: E702–E708, 2007.

2. Baviera AM, Zanon NM, Navegantes LCC, Migliorini RH, KettelhutIC. Chemical sympathectomy further increases mucle protein degradationof acutely diabetic rats. Muscle Nerve 38: 1027–1035, 2008.

3. Baviera AM, Zanon NM, Navegantes LC, Kettelhut IC. Involvement ofcAMP/Epac/PI3K-dependent pathway in the antiproteolytic effect of epi-nephrine on rat skeletal muscle. Mol Cell Endocrinol 315: 104–112, 2010.

4. Bechet D, Tassa A, Taillandier D, Combaret L, Attaix D. Lysosomalproteolysis in skeletal muscle. Int J Biochem Cell Biol 37: 2098–2114,2005.

5. Bodine SC, Latres E, Baumhueter S, Nunez L, Clarke BA, Pou-eymirou WT, Panaro FJ, Na E, Dharmarajan K, Pan ZQ, ValenzuelaDM, DeChiara TM, Stitt TN, Yancopoulos GD, Glass DJ. Identifica-tion of ubiquitin ligases required for skeletal muscle atrophy. Science 294:1704–1708, 2001.

6. Brito NA, Brito MN, Kettelhut IC, Migliorini RH. Intra-ventromedialhypothalamic injection of cholinergic agents induces rapid hyperglycemia,hyperlactatemia and gluconeogenesis activation in fed, conscious rats.Brain Res 626: 339–342, 1993.

7. Chen J, Feng HZ, Gupta Kelleher J, Dickerson KE, Wang J, Hunt D,Jour W, Gavrilova O, Jin JP, Weinstein LS. Gs� deficiency in skeletalmuscle leads to reduced muscle mass, fibers-type switching, and glucoseintolerance without insulin resistance or deficiency. Am J Physiol CellPhysiol 296: C930–C940, 2009.

8. Cikos S, Bukovska A, Koppel J. Relative quantification of mRNA:comparison of methods currently used for real-time PCR data analysis.BMC Mol Biol 8: 113–127, 2007.

9. Combaret L, Rallière C, Taillandier D, Tanaka K, Attaix D. Manip-ulation of the ubiquitin-proteasome pathway in cachexia: pentoxifyllinesuppresses the activation of 20S and 26S proteasomes in muscle fromtumor-bearing rats. Mol Biol Rep 26: 95–101, 1999.

10. Frost RA, Lang CH. Protein kinase B/Akt: a nexus of growth factor andcytokine signaling in determining muscle mass. J Appl Physiol 103:378–387, 2007.

11. Fulks RM, Li JB, Goldberg AL. Effects of insulin, glucose, and aminoacids on protein turnover in rat diaphragm. J Biol Chem 250: 290–298,1975.

12. Garófalo MAR, Kettelhut IC, Roselino JES, Migliorini RH. Effect ofacute cold exposure on norepinephrine turnover rates in rat White adiposetissue. J Autonomic Nerv Syst 60: 206–208, 1996.

13. Gomes MD, Lecker SH, Jagoe RT, Navon A, Goldberg AL. Atrogin-1,a muscle-specific F-box protein highly expressed during muscle atropy.Proc Natl Acad Sci USA 98: 14440–14445, 2001.

14. Gonçalves DA, Lira EC, Baviera AM, Cao P, Zanon NM, Arany Z,Bedard N, Tanksale P, Wing SS, Lecker SH, Kettelhut IC, navegantesLC. Mechanisms involved in 3=,5=-cyclic adenosine monophosphate-mediated inhibition of the ubiquitin-proteasome system in skeletal muscle.Endocrinology 150: 5395–5404, 2009.

15. Gonçalves DA, Silveira WA, Lira EC, Graça FA, Paula-Gomes S,Zanon NM, Kettelhut IC, Navegantes LC. Clenbuterol suppressesproteasomal and lysosomal proteolysis and atrophy-related genes in de-nervated rat soleus muscles independently of Akt. Am J Physiol Endocri-nol Metab 302: E123–E133, 2012.

16. Henriksson J. Microdialysis of skeletal muscle at rest. Proc Nutr Soc 58:919–923, 1999.

17. Hickner RC, Rosdahl H, Borg L, Ungerstedt U, Jorfeldt L, Hen-riksson J. The ethanol technique of monitoring local blood flow changesin rat skeletal muscle: implications for microdialysis. Acta Physiol Scand176: 87–97, 1992.

18. Hinkle RT, Dolan E, Cody DB, Bauer MB, Isfort RJ. Phosphodiester-ase 4 inhibition reduces skeletal muscle atrophy. Muscle Nerve 32:775–781, 2005.



by 10.220.32.246 on Novem


Dow

nloaded from


19. Hoover F, Mathiesen I, Skålhegg BS, Lømo T, Taskén K. Differentialexpression and regulation of the PKA signalling pathway in fast and slowskeletal muscle. Anat Embryol (Berl) 203: 193–201, 2001.

20. Kettelhut IC, Pepato MT, Migliorini RH, Medina R, Goldberg AL.Regulation of different proteolytic pathways in skeletal muscle in fastingand diabetes mellitus. Brazilian J Med Biol Res 27: 981–993, 1994.

21. Kline WO, Panaro FJ, Yang H, Bodine SC. Rapamycin inhits thegrowth and muscle-sparing effects of clenbuterol. J Appl Physiol 102:740–747, 2007.

22. Landsberg L, Young JB. Catecholamines and the adrenal medulla. In:Willians Textbook of Endocrinology, 8th ed., edited by Wilson JD, FosterDW. Philadelphia, PA: WB Saunders, 1992.

23. Lecker SH, Jagoe RT, Gilbert A, Gomes M, Baracos V, Bailey J, PriceSR, Mitch WE, Goldberg AL. Multiple types of skeletal muscle atrophyinvolve a common program of changes in gene expression. FASEB J 18:39–51, 2004.

24. Lecker SH, Goldberg AL, Mitch WE. Protein degradation by theubiquitin-proteasome pathway in normal and disease states. J Am SocNephrol 17: 1807–1819, 2006.

25. Li JB, Goldberg AL. Effects of food deprivation on protein synthesis anddegradation in rat skeletal muscles. Am J Physiol 231: 441–448, 1976.

26. Lin J, Wu H, Tarr PT, Zhang CY, Wu Z, Boss O, Michael LF,Puigserver P, Isotani E, Olson EN, Lowell BB, Bassel-Dubyk R,Spiegelman BM. Transcriptional co-activator PGC-1� drives the forma-tion os slow-twitch muscle fibres. Nature 418: 797–801, 2002.

27. Lira EC, Graça Fa Gonçalves DA, Zanon NM, Baviera AM, Strind-berg L, Lönnroth P, Migliorini Kettelhut IC RH, Navegantes LCC.Cyclic adenosine monophosphate-phosphodiesterase inhibitors reduceskeletal muscle protein catabolism in septic rats. Shock 27: 687–694, 2007.

28. Lira EC, Gonçalves DA, Parreiras-ESilva LT, Zanon NM, KettelhutIC, Navegantes LC. Phosphodiesterase-4 inhibition reduces proteolysisand atrogenes expression in rat skeletal muscles. Muscle Nerve 44:371–81, 2011.

29. Lönnroth P, Strindberg L. Validation of the internal reference techniquefor calibrating microdialysis catheters in situ. Acta Physiol Scand 153:375–380, 1995.

30. Lowry OH, Rosebrough NJ, Farr AL, Randall RF. Protein measure-ment with the folin phenol reagent. J Biol Chem 193: 265–275, 1951.

31. Lynch GS, Ryall JG. Role of �-adrenoceptor signaling in skeletalmuscle: implications for muscle wasting and disease. Physiol Rev 88:729–767, 2008.

32. Mei FC, Qiao J, Tsygankova OM, Meinkoth JL, Quilliam LA, ChengX. Differential signaling of cyclic AMP: opposing effects of exchangeprotein directly activated by cyclic AMP and cAMP-dependent proteinkinase on protein kinase B activation. J Biol Chem 277: 11497–11504,2002.

33. Moriscot AS, Baptista IL, Bogomolovas J, Witt C, Hirner S, GranzierH, Labeit S. MuRF1 is a muscle fiber-type II associated factor andtogether with MuRF2 regulates type-II fiber trophicity and maintenance. JStruct Biol 170: 344–353, 2010.

34. Nader GA. Molecular determinants of skeletal muscle mass: getting the“AKT” together. Int J Biochem Cell Biol 37: 1985–1996, 2005.

35. Navegantes LC, Resano NM, Migliorini RH, Kettelhut IC. Role ofadrenoceptors and cAMP on the catecholamine-induced inhibition of

proteolysis in rat skeletal muscle. Am J Physiol Endocrinol Metab 280:E663–E668, 2001.

36. Navegantes LC, Resano NM, Baviera AM, Migliorini RH, KettelhutIC. Effect of sympathetic denervation on the rate of protein synthesis in ratskeletal muscle. Am J Physiol Endocrinol Metab 286: E642–E647, 2004.

37. Roatta S, Farina D. Sympathetic actions on the skeletal muscle. ExercSport Sci Rev 38: 31–35, 2012.

38. Sandri M, Sandri C, Gilbert A, Skurk C, Calabria E, Picard A, WalshK, Schiaffino S, Lecker SH, Goldberg AL. Foxo transcription factorsinduce the atrophy-related ubiquitin ligase atrogin-1 and cause skeletalmuscle atrophy. Cell 117: 399–412, 2004.

39. Sandri M, Lin J, Handschi C, Yang W, Arany ZP, Lecker SH,Goldberg AL, Spiegelman BM. PGC-1� protects skeletal muscle formatrophy by suppressing FoxO3 action and atrophy specific gene transcrip-trion. Proc Natl Acad Sci USA 103: 16260–16265, 2006.

40. Scharff R, Wool IG. Effect of diabetes on the concentration of aminoacids in plasma and heart muscle of rats. Biochem J 99: 173–178, 1966.

41. Sacheck JM, Ohtsuka A, Mclary SC, Goldberg AL. IGF-I stimulatesmuscle growth by suppressing protein breakdown and expression ofatrophy-related ubiquitin ligases, atrogin-1 and MuRF1. Am J PhysiolEndocrinol Metab 287: E591–E601, 2004.

42. Stitt TN, Drujan D, Clarke BA, Panaro F, Timofeyva Y, Kline WO,Gonzalez M, Yancopoulos GD, Glass DJ. The IGF-1/PI3K/Akt pathwayprevents expression of muscle atrophy-induced ubiquitin ligases by inhib-iting FOXO transcription factors. Mol Cell 14: 395- 626, 2004.

43. Stuenaes JT, Bolling A, Ingvaldsen A, Rommundstad C, Sudar E, LinFC, Lai YC, Jensen J. Beta-adrenoceptor stimulation potentiates insulin-stimulated PKB phosphorylation in rat cardiomyocytes via cAMP andPKA. Br J Pharmacol 160: 116–129, 2010.

44. Takahashi A, Ikarashi Y, Ishimaru H, Maruyana Y. Compensationbetween sympathetic nerves and adrenalmedullary activity: effects ofadrenodemedullation and sympathectomy on catecholamine turnover. LifeSci 53: 1567–1572, 1993.

45. Taubin HL, Djahanguiri B, Landsberg L. Noradrenaline concentrationand turnover in different regions of the gastrointestinal tract of the rat: anapproach to the evalutation of sympathetic activity in the gut. GUT 13:790–795, 1972.

46. Ungerstedt U. Microdialysis—principles and application for studies inanimals and man. J Intern Med 230: 365–373, 1991.

47. Waalkes TP, Udenfriend SA. A fluorimetric method for estimation oftyrosine in plasma and tissue. J Lab Clin Med 50: 733–736, 1957.

48. Wang J, Yang H, Zuo Z, Yan X, Wang Y, Luo X, Jiang H, Chen K,Zhu W. Molecular dynamics simulations on the mechanism of transport-ing methylamine and ammonia by ammonium transporter AmtB. J PhysChem B 114: 15172–15179, 2010.

49. Wing SS, Goldberg AL. Glucocorticoides activate the ATP-ubiquitin-dependent proteolytic system in skeletal muscle during fasting. Am JPhysiol Endocrinol Metab 264: E668–E676, 1993.

50. Young JB, Rosa RM, Landsberg L. Dissociation of sympathetic nervoussystem and adrenal medullary responses. Am J Physiol Endocrinol Metab247: E35–E40, 1984.

51. Zhao J, Brault JJ, Schild A, Cao P, Sandri M, Schiaffino S, LeckerSH, Goldberg AL. FoxO3 coordinately activates protein degradation bythe autophagic/lysosomal and proteasomal pathways in atrophying musclecells. Cell Metab 6: 472–483, 2007.



by 10.220.32.246 on Novem


Dow

nloaded from


epinephrine depletion exacerbates the fasting-induced...

Documents