Retraction and Correction · Increased muscle PGC-1 expression protects from sarcopenia and metabolic disease during aging Tina Wenza, Susana G. Rossib, Richard L. Rotundob, Bruce

Retraction and Correction

RETRACTION

MEDICAL SCIENCESRetraction for “Dominant suppression of inflammation by glycan-hydrolyzed IgG,” by Kutty Selva Nandakumar, Mattias Collin,Kaisa E. Happonen, Allyson M. Croxford, Susanna L. Lundström,Roman A. Zubarev, Merrill J. Rowley, Anna M. Blom, andRikard Holmdahl, which appeared in issue 25, June 18, 2013, ofProc Natl Acad Sci USA (110:10252–10257; first published May13, 2013; 10.1073/pnas.1301480110).The authors wish to note the following: “Using studies of IgG

hydrolyzed by the streptococcal glycan hydrolyzing enzyme EndoS,we found that treatment of mice with hydrolyzed IgG blockedantibody mediated arthritis. As an explanation for this observa-tion, we suggested that EndoS-hydrolyzed IgG per se dominantlyblocks local immune complex formation.“With new data from our own follow up experiments, we have

now found that this conclusion was incorrect.“Our new data shows that injection of EndoS is much more

potent in vivo than we could logically anticipate, as i.v. injectionof doses containing less than 0.1 μg EndoS mixed with IgGsuppressed arthritis using the same model as the one reported inthe initial paper (collagen antibody-induced arthritis). We pre-viously excluded the possibility that contaminating EndoS couldplay a role, as this contaminating amount was not detected usingstandard methods in the hydrolyzed IgG fraction we used in theexperiments. Furthermore, much higher doses of EndoS injectedin the same mouse strain as a control experiment did not affectcollagen induced arthritis in earlier experiments. The correctinterpretation of our collective data is that EndoS operates verypotently in vivo on an immune complex-mediated disease, pos-sibly by accumulating within immune complexes. Because thisinterpretation is different from our major conclusion of the pub-lished paper, the authors have unanimously decided to retract thispaper to be able to publish the data connected with a correct in-terpretation. We sincerely apologize to readers of this paper, whomight have been misled by our earlier interpretation.”

Kutty Selva NandakumarMattias CollinKaisa E. HapponenAllyson M. CroxfordSusanna L. LundströmRoman A. ZubarevMerrill J. RowleyAnna M. BlomRikard Holmdahl

www.pnas.org/cgi/doi/10.1073/pnas.1419043111

CORRECTION

MEDICAL SCIENCESCorrection for “Increased muscle PGC-1α expression protects fromsarcopenia and metabolic disease during aging,” by Tina Wenz,Susana G. Rossi, Richard L. Rotundo, Bruce M. Spiegelman, andCarlos T. Moraes, which appeared in issue 48, December 1, 2009,of Proc Natl Acad Sci USA (106:20405–20410; first publishedNovember 16, 2009; 10.1073/pnas.0911570106).The authors note that the α-tubulin loading control blot in

Fig. 4D appeared incorrectly. The corrected figure and its legendappear below.

www.pnas.org/cgi/doi/10.1073/pnas.1419095111

Fig. 4. Increased PGC-1α levels in aging muscle prevent degradative pro-cesses. (A) Immunohistochemistry of biceps femoris using anti-active caspase3 antibody to detect apoptosis. (B) Apoptotic index in skeletal muscle ho-mogenates of wild-type and MCK-PGC-1α of different age-groups based onnucleosome fragmentation (n = 6 for each group). *, P < 0.05, **, P < 0.01,***, P < 0.001. (C) Western blot of Bax and Bcl-2 in skeletal muscle homo-genates. (D) Western blot of the 20S subunit of the proteasome and tubulinin skeletal muscle homogenates. (E) Western blot of LC3-I and LC3-II inskeletal muscle homogenates.

www.pnas.org PNAS | November 4, 2014 | vol. 111 | no. 44 | 15851

RETR

ACT

IONAND

CORR

ECTION

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Dow

nloa

ded

by g

uest

on

Janu

ary

27, 2

021

Retraction

MEDICAL SCIENCESRetraction for “Increased muscle PGC-1α expression protectsfrom sarcopenia and metabolic disease during aging,” by TinaWenz, Susana G. Rossi, Richard L. Rotundo, Bruce M. Spiegelmanand Carlos T. Moraes, which appeared in issue 48, December 1,2009, of Proc Natl Acad Sci USA (106:20405–20410; first publishedNovember 16, 2009; 10.1073/pnas.0911570106).The authors wish to note the following, “This article describes

improved systemic health in mice overexpressing PGC-1α inmuscle. Following an investigation on Tina Wenz by the Universityof Cologne, it was concluded that there was scientific misconductwith respect to the above paper. Specifically, blots in Figs. 2D and3B also appear in an unrelated publication (1). In addition, therewas a previously detected duplication of a tubulin blot betweenFigs. 4D and 5D. Even if its main conclusions may be correct, giventhat the scientific integrity of the paper was compromised, we seethe retraction of the paper as the best corrective action. Wesincerely apologize to the scientific community.”

1. Noe N, et al. (2013) Bezafibrate improves mitochondrial function in the CNS of a mousemodel of mitochondrial encephalopathy. Mitochondrion 13(5):417–426.

www.pnas.org/cgi/doi/10.1073/pnas.1619713114

E8502 | PNAS | December 27, 2016 | vol. 113 | no. 52 www.pnas.org

Increased muscle PGC-1� expression protects fromsarcopenia and metabolic disease during agingTina Wenza, Susana G. Rossib, Richard L. Rotundob, Bruce M. Spiegelmanc, and Carlos T. Moraesa,b,1

Departments of aNeurology and bCell Biology and Anatomy, University of Miami School of Medicine, Miami, FL33136 ; and cDepartment of Cell Biology,Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA 02115

Contributed by Bruce M. Spiegelman, October 8, 2009 (sent for review September 1, 2009)

Aging is a major risk factor for metabolic disease and loss of skeletalmuscle mass and strength, a condition known as sarcopenia. Bothconditions present a major health burden to the elderly population.Here, we analyzed the effect of mildly increased PGC-1� expressionin skeletal muscle during aging. We found that transgenic MCK-PGC-1� animals had preserved mitochondrial function, neuromuscu-lar junctions, and muscle integrity during aging. Increased PGC-1�levels in skeletal muscle prevented muscle wasting by reducingapoptosis, autophagy, and proteasome degradation. The preserva-tion of muscle integrity and function in MCK-PGC-1� animals resultedin significantly improved whole-body health; both the loss of bonemineral density and the increase of systemic chronic inflammation,observed during normal aging, were prevented. Importantly, MCK-PGC-1� animals also showed improved metabolic responses as evi-dent by increased insulin sensitivity and insulin signaling in agedmice. Our results highlight the importance of intact muscle functionand metabolism for whole-body homeostasis and indicate that mod-ulation of PGC-1� levels in skeletal muscle presents an avenue for theprevention and treatment of a group of age-related disorders.

mitochondria � PGC-1�

Aging is a multifactorial condition characterized by ener-getic deficits and decreased stress tolerance resulting in

tissue degeneration and malfunction. One of the major tissuesaffected during aging is skeletal muscle. Besides being funda-mental for movement and mobility, skeletal muscle is alsocrucial for overall energy balance and metabolism (1). Agingcauses a gradual loss of skeletal muscle mass, a conditionknown as sarcopenia (2). This age-associated muscle wastingresults in muscle weakness and hence has a significant effecton physical activity and life quality in the elderly. While theexact causes of sarcopenia are not clear, and multiple factorslikely define the disease mechanism (3), several lines ofevidence suggest mitochondrial involvement in this degener-ative condition (4). Apoptosis as well as autophagy have beenimplicated in the disease mechanism (4). Therapeutic strate-gies for sarcopenia like endurance exercise (5) and caloricrestriction (4) result in increased mitochondrial capacity in themuscle suggesting that mitochondrial dysfunction has a criticalrole in the muscle loss.

A key player controlling mitochondrial function is the peroxi-some proliferator-activated receptor � coactivator � (PGC-1�), amaster regulator of mitochondrial biogenesis (6). In skeletal mus-cle, PGC-1� can also prevent muscle wasting by regulating auto-phagy (7) and stabilization of the neuromuscular junction (NMJ)program (8) in the context of muscle atrophy during disease.Thereby, PGC-1� links mitochondrial function to muscle integrity(7). PGC-1� levels in skeletal muscle decrease during aging (9). Thehealth promoting effects of increased PGC-1� expression in skel-etal muscle could be shown in different mouse models with affectedmuscle such as Duchenne muscular dystrophy (8), denervation-induced atrophy (7), and mitochondrial myopathy (10).

ResultsIncreased PGC-1� Expression in Muscle Prevents Age-AssociatedWeight Gain and Improves Exercise Capacity. We have recently shownthat activation of PGC-1� delays the onset of a myopathy caused by

a mitochondrial dysfunction and extends life span and health span,most likely by increasing overall ATP generating capacity (10, 11).Here, we tested if elevated PGC-1� expression in muscle has abeneficial effect on the development and progression of sarcopeniaand its metabolic consequences. Therefore, we studied muscleaging in a mouse model with a MCK-PGC-1� transgene, which isexpressed in all skeletal muscles and results in PGC-1� proteinlevels equal to those found endogenously in type 1 muscle (6). Here,we observed that the MCK-PGC-1� animals did not show theage-associated weight gain observed in the wild-type control ani-mals (Fig. 1A and Fig. S1A). At 22 months of age, MCK-PGC-1�animals had �18% reduction in fat mass and �8% increase in leanmass compared to their age-matched control littermates (Fig. 1 Band C). We further observed that the hind limb muscle mass relativeto the body weight was significantly decreased in the aged wild-typeanimals compared to their age-matched MCK-PGC-1� littermates(Fig. 1D), indicating that PGC-1� might be in involved in regulatingmuscle mass during aging. Expression of PGC-1� also improved theexercise performance during aging as evidenced by increasedendurance (Fig. S1B) and improved performance in a high-performance exercise test compared to wild-type animals (Fig. 1E),indicating improved muscle function in the aging MCK-PGC-1�mice. Increased PGC-1� expression was reported to increaseendurance capacity in young mice (10, 12). Importantly, we alsoobserved that MCK-PGC-1� animals had increased levels of bonemineral density compared to their aging wild-type littermates (Fig.S1C), most likely because of the improved muscle function in thetransgenic animals (13). Moreover, we observed a longer lifespanin MCK-PGC-1� mice (Fig. S1D). Studies of larger groups ofanimals will be to be necessary to define the extent and significanceof this life-prolonging effect. We also observed that the expressionof the NAD� deacetylase Sirt1, which is implicated in longevity(14), is significantly increased in muscle from MCK-PGC-1� mice(Fig. S1E).

Elevated PGC-1� Levels Increases Mitochondrial Mass and PreserveMitochondrial Oxidative Phosphorylation Capacity in Muscle DuringAging. We next analyzed the effect of PGC-1� expression onmitochondrial function during aging by assessing the specific activ-ity of different mitochondrial enzymes in skeletal muscle mitochon-dria and homogenates. At 3 months of age, no significant differ-ences in the oxidative phosphorylation (OXPHOS) enzymecytochrome c oxidase (COX) in mitochondria were observedbetween wild-type and MCK-PGC-1� animals. Samples obtainedfrom 12-month-old wild-type animals showed a decline in COXactivity, while samples obtained from age-matched MCK-PGC-1�animals maintained their COX activity. At 22 months of age,

Author contributions: T.W., S.G.R., R.L.R., B.M.S., and C.T.M. designed research; T.W. andS.G.R. performed research; R.L.R. and B.M.S. contributed new reagents/analytic tools; T.W.,S.G.R., R.L.R., and C.T.M. analyzed data; and T.W., R.L.R., B.M.S., and C.T.M. wrote thepaper.

The authors declare no conflict of interest.

1To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/cgi/content/full/0911570106/DCSupplemental.

www.pnas.org�cgi�doi�10.1073�pnas.0911570106 PNAS � December 1, 2009 � vol. 106 � no. 48 � 20405–20410

MED

ICA

LSC

IEN

CES

See Retraction Published December 19, 2016

wild-type samples had only �60% of the COX activity observed inthe younger animals, whereas samples derived from MCK-PGC-1�animals were virtually indistinguishable from those of 3-month-oldanimals (Fig. 2A). Similar results were obtained for the mitochon-drial matrix protein citrate synthase (CS) (Fig. S2A) and OXPHOScomplexes I�III (CI�III) (Fig. S2B).

When assessing mitochondrial enzymes in muscle homogenates,samples from MCK-PGC-1� animals had increased activities: At 3months of age, COX activity in muscle homogenates from MCK-PGC-1� animals was �1.5-fold increased over activitiy observed inwild-type controls (Fig. 2B). Increased CS and CI�III activitieswere also observed at this age in muscle homogenates. Since COX,CS, and CI�III activity in mitochondria at 3 months is notsignificantly different between MCK-PGC-1� and wild-type, theincrease in the activity can be attributed to an increase in mito-chondrial mass induced by the PGC-1� expression. In both MCK-PGC-1� and wild-type animals, COX activity declined over time.This decline was more pronounced in muscle homogenates fromwild-type samples, where COX activity decreased at 12 months ofage to �60% of the activity seen in 3-month-old wild-type animalsand decreased further to �30% at 22 months of age. A similardecline in muscle COX activity was reported before during aging(15). In contrast, at 12 months of age, muscle homogenates fromMCK-PGC-1� animals had �140% of COX activity observed in3-month-old control animals and �100% at 22 months of age.Similar trends were observed for CS and CI�III activities in musclehomogenates (Fig. S2 C and D).

The observed decrease in OXPHOS in muscle mitochondria andhomogenates during aging in wild-type animals was also evident inWestern blots. When mitochondria and muscle homogenates wereprobed for different subunits of the individual OXPHOS com-plexes, samples from aged wild-type animals showed decreasedlevels of the probed subunits compared to the young wild-typeanimals. Samples from aged MCK-PGC-1� animals showed little orno decline in subunit levels compared to the young MCK-PGC-1�samples (Fig. 2 C and D). Histochemical staining of muscle sectionswere in agreement with this finding of preserved mitochondrialfunction in aging MCK-PGC-1� mice (Fig. 2E). Maintenance ofoxidative capacity also allowed the aged MCK-PGC-1� animals tosupport their ATP demand during exercise training comparable toyoung animals. Aged wild-type animals showed a dramatic increase

in serum lactate levels and ATP depletion in muscle after exercise,indicating their susceptibility to metabolic stress caused by thedeclined mitochondrial function (Fig. 2 F and G).

RT-PCR experiments using skeletal muscle cDNA from 3-and 22-month-old wild-type mice revealed that PGC-1� levelsdeclined during aging by �60% (Fig. S2E). In addition, mito-chondrial transcription factor A (TFAM) and nuclear respira-tory factor 1 (NRF1), key transcription factors in mitochondrialbiosynthesis (16), are also reduced during aging (Fig. S3E). Wealso observed a decrease in a mitochondrial ribosome compo-nent suggesting a concomitant decrease in mitochondrial proteinsynthesis (Fig. S2F).

Increased PGC-1� Levels in Aging Muscle Enhances Anti-OxidantDefense and Prevents Oxidative Damages. Maintenance of func-tional mitochondria relies on coordination of mitochondrial bio-genesis and degradation. The observed decrease in factors regu-lating mitochondrial biogenesis in the aged wild-type animalsprompted us to analyze whether this decline causes accumulation ofdamaged cellular structures and/or change in degradative processes.We observed increased oxidized nucleic acids in the aged wild-typeas evidenced by immunodetection of 8-OH-guanosine in skeletalmuscle sections in contrast to aged MCK-PGC-1� animals, whichshowed only weak staining (Fig. 3A). When analyzing the anti-oxidant response, we observed that young and aged MCK-PGC-1�animals expressed higher levels of superoxide dismutase 2 (SOD2)

Fig. 1. Increased PGC-1� expression in skeletal muscle prevents age-associatedweightgainand improvesexercise capacityduringaging. (A)Comparisonofmiceexpressing the PGC-1� transgene in skeletal muscle (MCK-PGC-1�) and wild-typelittermates (control) at different ages. (B and C) Lean and fat mass of 22-month-old wild-type and PGC-1� animals as determined by DEXA scans (n � 6 for eachgroup). *, P � 0.05. (D) Relative hindlimb mass of 22-month-old wild-type andPGC-1� animals (n � 6 for each group). *, P � 0.001. (E) Treadmill performancetest at different ages for wild-type and PGC-1� animals (n � 9 for each group). *,P � 0.001. In this and all subsequent figures, error bars represent SD.

Fig. 2. Increased PGC-1� expression in skeletal muscle preserves OXPHOSfunction on organelle basis and increases overall OXPHOS capacity. (A and B)Relative COX activity in skeletal muscle mitochondria (A) and muscle homoge-nates (B) at different ages (n � 9 for each mouse group). *, P � 0.05, **, P � 0.001.(CandD)WesternblotofNDUFA9,flavoprotein(Fp),COXI,ATPase�,VDAC1,and�-tubulin in skeletal muscle mitochondria (C) and muscle homogenates (D) of 3-and 22-month-old mice. (E) Histology of biceps femoris muscle from MCK-PGC-1�

andwild-typecontrolmiceat3and22months showingsuccinatedehydrogenase(SDH), COX, and combined COX/SDH activity staining. (F and G) Levels of serumlactate and skeletal muscle ATP before and after exercise (n � 3 for each mousegroup). *, P � 0.05, **, P � 0.001.

20406 � www.pnas.org�cgi�doi�10.1073�pnas.0911570106 Wenz et al.

See Retraction Published December 19, 2016

in their skeletal muscle mitochondria compared to wild-type con-trols at all ages (Fig. 3B) and showed a 2.5- to 3-fold increase incatalase activity compared to wild-type samples (Fig. 3C). In agedwild-type samples, a decrease in the SOD2 levels and catalaseactivity compared to the young wild-type samples is evident (Fig. 3B and C), in agreement with the decrease in mitochondrial proteinsobserved before. While there is also a slight decrease in SOD2 andcatalase in aged vs. young MCK-PGC-1� mice, the levels are stilllargely preserved compared to the aged wild-type samples andpresumably could prevent the oxidative damage. Decreased oxi-dative protein modification in the aged MCK-PGC-1� skeletalmuscle homogenates compared to the aged wild-type could also beseen as increased protein carbonylation (Fig. S3). Activation ofPGC-1� was previously shown to increase the anti-oxidant responsein young animals and thus prevent the oxidative damage associatedwith muscle damage (11).

MCK-PGC-1� Mice Have Preserved Muscle Integrity and NMJ. Overall,the transgenic MCK-PGC-1� expression had a beneficialeffect on muscle integrity. In the aged wild-type animals,serum creatine kinase levels were significantly increased, whilethe levels in aged MCK-PGC-1� animals were within the rangeobserved in the young mice samples (Fig. S4A). As evidentfrom the hematoxylin and eosin (H&E) staining of musclesections from young and aged wild-type and MCK-PGC-1�animals, aged wild-type animals showed increased number of

fibers with centered nuclei, indicating regenerating fibers (Fig.3D and Fig. S4B). The aged wild-type animals also showedgreatly increased inf lammatory markers TNF� and IL-6 inskeletal muscle compared to their age-matched MCK-PGC-1�littermates (Fig. 3E) indicating severe tissue damage. RT-PCRexperiments revealed, that already at 3 months of age, MCK-PGC-1� had slightly lowered levels of the cytokines (Fig. S4C).As a response to the proinf lammatory stimuli, NF�B isphosphorylated and translocated from cytoplasm to the nu-cleus to initiate transcriptional response (17). In agreement,the levels of the phosphorylated p65 NF�B subunit weresignificantly increased in the muscle of wild-type animals,whereas only very low levels of p65 NF�B could be detectedin MCK-PGC-1� animals at all ages (Fig. 3F). Interestingly, weobserved that aged MCK-PGC-1� mice had lower serumTNF� and IL-6 levels compared to controls (Fig. S4D). Wealso observed that skeletal muscle from aged wild-type miceshowed increased collagen deposition indicative of fibrosis asa result of the muscle dysfunction. In contrast, in muscle ofMCK-PGC-1� mice, less collagen deposition was observedsuggesting that the elevated PGC-1� levels attenuate age-related muscle fibrosis (Fig. S5).

Because PGC-1� is involved in NMJ remodeling (8), weanalyzed whether its expression in skeletal muscle inf luencesNMJ organization and acetylcholinesterase (AChE) expres-sion during aging. In both wild-type and MCK-PGC-1� mice,

Fig. 3. Increased PGC-1� expression during aging resultsin increased anti-oxidant response, lowered inflammatorymarkers, and preserved muscular structure. (A) Immunohis-tochemistry of mice biceps femoris using anti-8-OH-guanosine antibody to detect oxidative damage to nucleicacids. (B) Western blot of SOD2, HSP70, and VDAC1 inskeletalmusclemitochondria. (C)Catalaseactivity inmusclemitochondria (n � 9 for each group). *, P � 0.001. (D)Quantification of regenerating fibers in biceps femorisbased on the number of fibers with centered nuclei (Fig.S3C) (n � 6 for each group). *, P � 0.05. (E) Level ofinflammatory markers in skeletal muscle (n � 9 for eachgroup). *, P � 0.01. (F) Western blot of p65-NF�B and�-tubulin in skeletal muscles homogenates of 3- and 22-month-old mice. (G) NMJs stained with Alexa-555 �-bun-garotoxin to label AChR and Alexa-488 Fasciculin2 to labelAChE to visualize the NMJ. (H) Sucrose gradient profiles ofAChE activity show the different oligomeric forms of AChEexpressed in young and older animals. G1, G2, and G4 arethe globular monomeric, dimeric, and tetrameric AChEforms, respectively.A8andA12indicatethepositionsoftheasymmetric collagen tailed synaptic forms.

Wenz et al. PNAS � December 1, 2009 � vol. 106 � no. 48 � 20407

MED

ICA

LSC

IEN

CES

See Retraction Published December 19, 2016

we observed an increase in segmentation in NMJs related toaging (18) (Fig. 3G). However, the area comprising the NMJwas significantly larger by �50% (Fig. S6), with a parallelincrease in levels of AChE and acetylcholine receptors(AChR). The density of AChR at the NMJs was particularlyelevated (Fig. S6). In addition, there were also significantchanges in the pattern of AChE oligomeric forms expressed(Fig. 3H). Although there was little or no change in theexpression of the monomeric (G1) and dimeric (G2) newly-synthesized endoplasmic reticulum resident forms of the en-zyme, there was a dramatic age-related decrease in the asym-metric A8 and A12 AChE forms consisting of two to threetetramers of catalytic subunits, which are the major oneslocalized to the NMJ (19). The pattern of AChE formsexpressed in MCK-PGC-1� mice more closely resembled thepattern of enzyme expressed in young mice muscles, indicatingthat the increased expression of PGC-1� is muscle maintainedsynaptic proteins, and hence function, in a younger state.

Increased PGC-1� Expression During Aging Attenuates DegradativeProcesses Associated with Muscle Atrophy. Degradative processessuch as apoptosis and autophagy have been implicated inmuscle atrophy and sarcopenia (4). PGC-1� has been sug-gested to have anti-apoptotic effects (20) and has been iden-tified as a negative regulator of autophagy in skeletal muscle(7) Here we show that increased PGC-1� levels in skeletalmuscle are also protective against degradative processes dur-ing aging. When analyzing the frequency of apoptotic events,we saw a clear increase in active caspase 3 in aged wild-typemuscle both by immunostaining (Fig. 4A) and by Western blotanalysis (Fig. S7B), while no significant increase in activecaspase 3 was observed in aging MCK-PGC-1� muscle (Fig. 4Aand Fig. S7B). We further analyzed DNA fragmentation inmuscle at different ages for both wild-type and MCK-PGC-1�mice. We observed a significant increase in this apoptoticindex in wild-type skeletal muscle starting at 12 months of age(Fig. 4B). While the levels of DNA fragmentation also in-creased in MCK-PGC-1� animals, this increase was �50%lower than in the wild-type muscles (Fig. 4B). This protective

effect was also evident in the preserved Bcl2:Bax ratios inskeletal muscle of MCK-PGC-1� animals. In contrast, theBcl2:Bax ratio declined during aging in the wild-type musclethereby favoring apoptotic events (Fig. 4C and quantificationin Fig. S7A). We also observed increases in the levels of Bax,and to a lesser extent Bcl2, in both aging wild-type and PGC-1�animals.

We next used RT-PCR experiments to analyze the expres-sion of genes associated with muscle atrophy. We foundincreased expression of ubiquitin ligases and autophagy re-lated genes in aging wild-type muscle, while expression in theMCK-PGC-1� samples remained unchanged or only slightlyincreased (Fig. S7C). Therefore, we further analyzed bothdegradative processes during aging. We observed a largeincrease in the 20S proteasome subunit in aged wild-typemuscles compared to aged-matched MCK-PGC-1� and youngwild-type muscles (Fig. 4D). We also observed a minor in-crease in the levels of 20S proteasome subunit in the agedMCK-PGC-1� samples compared to the young MCK-PGC-1�ones (Fig. 4D). We also observed less ubiquitinylation insamples from aged MCK-PGC-1� mice compared to agedwild-type controls (Fig. S7E). We then analyzed alterations inautophagy during skeletal muscle aging by studying the mi-crotubule-associated protein light chain 3 (LC3). The cytosolicform, called LC3-I, is further converted to an autophagosome-associating form, LC3-II. Therefore, the protein level ofLC3-II is often used as a measure to determine autophagicactivity (21). Western blot analysis of the LC3-I and LC3-IIisoforms revealed, that the LC3-II:LC3-I ratio increases sig-nificantly during aging in wild-type muscle, whereas no sig-nificant increase was observed in the MCK-PGC-1� samples(Fig. 4E, quantification in Fig. S7D) emphasizing the regula-tory role of PGC-1� in autophagy. Notably, the FoxO proteins,which control autophagy in skeletal muscle, have been shownto be activated by reactive oxygen species (ROS) and TNF�(22). Hence, the increase anti-oxidant defense and decreasedTNF� levels in the aging MCK-PGC-1� mice might alsocontribute to prevent dysregulated autophagy. Our resultsindicate that during muscle aging, PGC-1� reduces autophagy,presumably similarly as in muscle wasting diseases (7), andprotects muscle from proteolysis and apoptosis.

Muscle-Specific PGC-1� Expression Prevents Age-Associated InsulinResistance and Improves Muscular Insulin Signaling. We next ad-dressed whether the preserved muscle function in aging MCK-PGC-1� mice had an effect on overall metabolism. Loss of themetabolic quality in the aging muscle is a major risk factor forthe development of insulin resistance and diabetes duringaging (1). We found that MCK-PGC-1� animals had increasedglucose tolerance and increased insulin sensitivity comparedto age-matched wild-type animals (Fig. 5 A and B). In addition,we found that aged MCK-PGC-1� animals had lower fastingblood glucose levels (Fig. S8A). This difference in insulinsensitivity is clearly induced by the aging process, since nodifferences between MCK-PGC-1� and wild-type animalswere detected at 3 months of age when fed a regular diet (Fig.S8 B and C). Thus, elevated muscle-specific PGC-1� expres-sion seems to be protective for age-associated insulin resis-tance. Interestingly, young MCK-PGC-1� transgenic micewere more prone to fat-induced insulin resistance due todecreased insulin-stimulated muscle glucose uptake (23).These findings suggest that PGC-1� has different effects indiet-induced versus age-associated insulin resistance. We alsoobserved that aged MCK-PGC-1� animals had unchangedserum triglyceride levels, whereas aged wild-type animals hadsignificantly increased triglyceride levels indicative of theimbalanced metabolism (Fig. 5C). Clearly the preserved mus-cle function and muscle mass are major contributors to the

Fig. 4. Increased PGC-1� levels in aging muscle prevent degradative processes(A) Immunohistochemistry of biceps femoris using anti-active caspase 3 antibodyto detect apoptosis. (B) Apoptotic index in skeletal muscle homogenates ofwild-type and MCK-PGC-1� of different age-groups based on nucleosome frag-mentation (n � 6 for each group). *, P � 0.05, **, P � 0.01, ***, P � 0.001. (C)Western blot of Bax and Bcl-2 in skeletal muscle homogenates. (D) Western blotof the 20S subunit of the proteasome and tubulin in skeletal muscle homoge-nates. (E) Western blot of LC3-I and LC3-II in skeletal muscle homogenates.

20408 � www.pnas.org�cgi�doi�10.1073�pnas.0911570106 Wenz et al.

See Retraction Published December 19, 2016

increased insulin sensitivity in the aged MCK-PGC-1� mice.We further assessed whether increased expression of PGC-1�during aging also affected insulin signaling. We analyzedinsulin signaling by assaying phosphorylation of two compo-nents of the insulin signaling pathway, Akt and mTOR. Wefound that levels of phospho-Akt (pAkt) and phospho-mTOR(pmTOR) did not significantly decrease during aging in theMCK-PGC-1� mice (Fig. 5D). In contrast, pAkt and pmTORlevels were already decreased in the young wild-type animalscompared to the age-matched MCK-PGC-1� mice anddropped even further during aging (Fig. 5D). These findingsindicate increased and robust insulin signaling in the MCK-PGC-1� mice during aging, whereas the insulin signaling isattenuated in the older wild-type animals. Akt and mTOR arekey players in regulating muscle integrity; Akt, along withPGC-1� is a negative regulator of the FoxO proteins thatcontrol protein degradation, whereas mTOR is a modulator ofprotein synthesis in muscle (24). Hence, the improved insulinsignaling plus the regulatory function of PGC-1� on proteindegradation in the aged MCK-PGC-1� mice prevents theexcessive protein degradation observed in the wild-type ani-mals, where negative regulation of the FoxO proteins isreduced, presumably contributing to the age-associated musclewasting. Improved insulin sensitivity and improve insulinaction is suggested to be a key mechanism in the life-prolonging effect of caloric restriction (25). Intriguingly, whilecaloric restriction is a systemic approach, we see similar effectsin the MCK-PGC-1� mice. We further observed that, in bothyoung and aged MCK-PGC-1� animals, vascularization of themuscle was greatly increased compared to the wild-typeanimals, as evident in the activity stain for the endothelialvascular structure (Fig. S9). The increased blood supply con-tributes to the improved muscular function (26) as well asimproved muscle metabolism as previously reported (27).

DiscussionMitochondrial dysfunction has been implicated in many age-related diseases (14). In particular, OXPHOS function de-clines with age in multiple tissues (28). Hence, mitochondriaseem to be a prime target for anti-aging interventions (14). Wecould previously show that elevated PGC-1� expression in

skeletal muscle enhanced OXPHOS function in a mousemodel of mitochondrial myopathy, delaying the onset of themyopathy and markedly prolonging lifespan (10). This findingprompted us to investigate if elevated PGC-1� expression canalso compensate for the age-associated decline of OXPHOSfunction in muscle. Intriguingly, we found that in MCK-PGC-1� animals Sirt1 expression was increased compared towild-type controls suggesting a positive feedback mechanism.Sirt1 is known to activate PGC-1� by deacetylation (29), andwhile the detailed mechanisms of this feedback regulationremain unclear, our findings suggest that Sirt1/PGC-1� forma regulatory axis to control mitochondrial function duringaging. This positive feedback may also contribute to some ofthe anti-aging effects observed.

We found that increased PGC-1� expression buffered againstthe decline in OXPHOS function during aging thereby preserv-ing metabolic fitness and exercise capacity. Unexpectedly, wefound that PGC-1� expression not only increases mitochondrialbiogenesis but also maintains OXPHOS function at the mito-chondrial level during aging. Hence, the observed overall en-hancement in OXPHOS capacity in the MCK- PGC-1� mice isthe sum of an increased mitochondrial mass and increasedmitochondrial function.

Sarcopenia is associated with increased apoptosis, autoph-agy, and proteolysis (4). Elevated PGC-1� expression inmuscle during aging decreased these degradative processes,which presumably preserved muscle integrity and muscle mass.In addition, we observed that MCK-PGC-1� mice had ‘‘func-tionally younger’’ NMJ, which together with the enhancedATP generating capacity and muscle integrity preserved mus-cle function during aging. MCK-PGC-1� mice also showed lessmuscle fibrosis than wild-type mice during aging. Fibrosis issupposedly driven by the repeated bouts of muscle-fiberdegeneration and ensuing inf lammation, such as in Duchennemuscular dystrophy (30). During aging, there is increasedcollagen deposition and fibrosis (31). It seems likely, thatelevated PGC-1� expression in muscle might protect from theage-associated fibrosis by maintaining muscle integrity andpreventing an inf lammatory response. In agreement, increasedmuscle PGC-1� expression ameliorates a dystrophin-deficientphenotype in mice (8).

We also observed that muscle from MCK-PGC-1� mice hadincreased anti-oxidant defense and less oxidative damagecompared to their wild-type littermates. Higher anti-oxidantlevels and less ROS have been implicated in longevity (14).Moreover, accumulation of damaged cellular structures suchas oxidized proteins and lipids trigger degradative processessuch as autophagy, apoptosis, and proteolysis, which, whenbecoming excessive, are detrimental for cellular survival. Inaddition, degradation of damaged structures requires de novosynthesis to replenish and maintain cellular function. Theelevated PGC-1� expression in muscle stimulates mitochon-drial biogenesis and might thus facilitate mitochondrial turn-over, so that damaged organelles do not accumulate and themitochondria remain in a ‘‘functionally younger’’ state.

Intriguingly, we found that the maintenance of muscleintegrity and metabolic function in the MCK-PGC-1� micehad systemic effects. Chronic inf lammation is a major under-lying condition of the aging process (32). Here, we could showthat mildly elevate expression of PGC-1� in muscle modulatesthis inf lammatory response not only in the muscle itself, butalso systemically. Elevated circulating inf lammatory markerscan also damage other tissues unrelated to the primary dam-aged organ (33) and are prognostic for many age-relateddiseases, such as cardiovascular disease, diabetes, and demen-tia (34–36). Hence, increased muscle-specific PGC-1� expres-sion seems to improve whole-body health by maintaining

Fig. 5. Increased PGC-1� expression in skeletal muscle during aging improvesinsulin signaling and vascularization, resulting in improved glucose and insulintolerance. (A) Glucose tolerance test in 22-month-old wild-type and MCK-PGC-1�

animals (n � 9 for each group). *, P � 0.05, **, P � 0.01. (B) Insulin tolerance testin 22-month-old animals (n � 9 for each group). *, P � 0.05, **, P � 0.01, ***, P �0.001. (C) Quantification of circulating triglyceride levels (n � 9 for each group).

*, P � 0.01. (D) Western blot of pAkt/Akt, pmTOR/mTOR, and tubulin in skeletalmuscle homogenates.

Wenz et al. PNAS � December 1, 2009 � vol. 106 � no. 48 � 20409

MED

ICA

LSC

IEN

CES

See Retraction Published December 19, 2016

muscular integrity and thus preventing a systemic inf lamma-tory response.

In addition, MCK-PGC-1� mice did not show the age-associated weight gain, loss of bone mineral density, and mostimportantly, did not develop age-related insulin resistance.These results can be attributed to the preserved metabolicprocesses and muscle function in the MCK-PGC-1� mice.PGC-1� and its target genes are down-regulated in humandiabetes (37). Our results imply that increasing PGC-1� inmuscle can compensate for this down-regulation and preventinsulin resistance in mice. Interestingly, resveratrol-inducedPGC-1� expression also protected from diet-induced diabetes(38). These findings clearly underscore the importance ofmuscle function for whole body homeostasis.

In conclusion, our results indicate that increased musclePGC-1� expression not only prevents the age-associated mus-cle wasting and dysregulated muscle metabolism underlyingsarcopenia, but has also a significant beneficial effect onwhole-body metabolism. Preservation of muscle integrity bypreventing the decline in mitochondrial function, the NMJstructure, and excessive degradative processes during agingimproves insulin sensitivity and insulin action. Additionally,this protection of muscle integrity also prevents the age-related increase of circulating cytokines, which are prognosticfor disease development and death during aging. Interestingly,these findings support the concept that the anti-aging effectsof SIRT1 are mediated, at least in part, by a metabolicremodeling that involves increased mitochondrial biogenesis(14). These robust effects clearly identify PGC-1� activationnot only as a therapeutic target for treatment of sarcopenia

and age-related muscle metabolic disorders, but also highlightsthe importance of muscle function for whole-body metabolismand health.

Materials and MethodsAnimal Husbandry. The mice expressing PGC-1� in skeletal muscle were describedin ref. 6. Mice used in the different experimental groups were littermates, andtheir background was �94% C57BL6 and 6% 129svj. The mice were kept in acondition of 12-h light/dark cycle at room temperature. They were allowed aregular diet (Rodent Chow 5010, Harlan) ad libitum.

Phenotyping. Treadmill performance and endurance tests were carried out asdescribed in ref. 10. Blood, body composition measurements, and pathologicalstainings are described in SI Text.

Mitochondrial Function and Composition. Enzyme activities and Western blotswere performed as described in ref. 10. The list of antibodies used is described inSI Text.

Data Analysis. Data obtained are represented as mean (� SD) from three to sevenmice per group, and statistical significance was determined using the Student’s ttest. A P value � 0.05 was considered significant.

Additional Experimental Procedures. The detailed additional procedures aredescribed in SI Text.

ACKNOWLEDGMENTS. This work was supported by Public Health Service GrantsNS041777, CA85700, and EY10804 (to C.T.M.); AG0005917 (to R.L.R.); andDK61562 and DK54477 (to B.M.S.); the Muscular Dystrophy Association (C.T.M.);andtheEllisonFoundation(B.M.S.).Dr.Wenzwassupportedbyafellowshipfromthe United Mitochondrial Disease Foundation.

1. Dela F, Kjaer M (2006) Resistance training, insulin sensitivity and muscle function in theelderly. Essays Biochem 42:75–88.

2. Rolland Y, et al. (2008) Sarcopenia: Its assessment, etiology, pathogenesis, consequencesand future perspectives. J Nutr Health Aging 12:433–450.

3. Marcell TJ (2003) Sarcopenia: Causes, consequences, and preventions. J Gerontol A Biol SciMed Sci 58:M911–M916.

4. Marzetti E, Anne Lees H, Eva Wohlgemuth S, Leeuwenburgh C (2009) Sarcopenia ofaging: Underlying cellular mechanisms and protection by calorie restriction. Biofactors35:28–35.

5. Taaffe DR (2006) Sarcopenia—Exercise as a treatment strategy. Aust Fam Physician35:130–134.

6. Lin J, et al. (2002) Transcriptional co-activator PGC-1 � drives the formation of slow-twitchmuscle fibres. Nature 418:797–801.

7. Sandri M, et al. (2006) PGC-1� protects skeletal muscle from atrophy by suppressingFoxO3 action and atrophy-specific gene transcription. Proc Natl Acad Sci USA103:16260–16265.

8. Handschin C, et al. (2007) PGC-1� regulates the neuromuscular junction program andameliorates Duchenne muscular dystrophy. Genes Dev 21:770–783.

9. AndersonR,ProllaT (2009)PGC-1� inagingandanti-aging interventions. BiochimBiophysActa 1790:1059–1066.

10. WenzT,DiazF, SpiegelmanBM,MoraesCT (2008)Activationof thePPAR/PGC-1� pathwayprevents a bioenergetic deficit and effectively improves a mitochondrial myopathy phe-notype. Cell Metab 8:249–256.

11. Wenz T, Diaz F, Hernandez D, Moraes CT (2009) Endurance exercise is protective for micewith mitochondrial myopathy. J Appl Physiol 106:1712–1719.

12. Calvo JA, et al. (2008) Muscle-specific expression of PPAR� coactivator-1� improves exer-cise performance and increases peak oxygen uptake. J Appl Physiol 104:1304–1312.

13. Lima RM, et al. (2009) Fat-free mass, strength, and sarcopenia are related to bone mineraldensity in older women. J Clin Densitom 12:35–41.

14. Guarente L (2008) Mitochondria—A nexus for aging, calorie restriction, and sirtuins? Cell132:171–176.

15. Desai VG, Weindruch R, Hart RW, Feuers RJ (1996) Influences of age and dietary restrictionon gastrocnemius electron transport system activities in mice. Arch Biochem Biophys333:145–151.

16. Scarpulla RC (2008) Transcriptional paradigms in mammalian mitochondrial biogenesisand function. Physiol Rev 88:611–638.

17. Traenckner EB, et al. (1995) Phosphorylation of human I � B-� on serines 32 and 36 controlsI � B-� proteolysis and NF-� B activation in response to diverse stimuli. EMBO J 14:2876–2883.

18. Courtney J, Steinbach JH (1981) Age changes in neuromuscular junction morphology andacetylcholine receptor distribution on rat skeletal muscle fibres. J Physiol 320:435–447.

19. Rotundo RL (2003) Expression and localization of acetylcholinesterase at the neuromus-cular junction. J Neurocytol 32:743–766.

20. Adhihetty PJ, Uguccioni G, Leick L, Hidalgo J, Pilegaard H, Hood DA (2009) The role ofPGC-1� on mitochondrial function and apoptotic susceptibility in muscle. Am J Physiol CellPhysiol 297:C217–C225.

21. KabeyaY,etal. (2004)LC3,GABARAPandGATE16 localizetoautophagosomalmembranedepending on form-II formation. J Cell Sci 117:2805–2812.

22. Essers MA, et al. (2004) FOXO transcription factor activation by oxidative stress mediatedby the small GTPase Ral and JNK. EMBO J 23:4802–4812.

23. Choi CS, et al. (2008) Paradoxical effects of increased expression of PGC-1� on musclemitochondrial function and insulin-stimulated muscle glucose metabolism. Proc Natl AcadSci USA 105:19926–19931.

24. Mammucari C, Schiaffino S, Sandri M (2008) Downstream of Akt: FoxO3 and mTOR in theregulation of autophagy in skeletal muscle. Autophagy 4:524–526.

25. Bonkowski MS, et al. (2009) Disruption of growth hormone receptor prevents calorierestriction from improving insulin action and longevity. PLoS One 4:e4567.

26. Arany Z, et al. (2008) HIF-independent regulation of VEGF and angiogenesis by thetranscriptional coactivator PGC-1�. Nature 451:1008–1012.

27. Sung HK, et al. (2009) COMP-angiopoietin-1 enhances skeletal muscle blood flow andinsulin sensitivity in mice. Am J Physiol Endocrinol Metab 297:E402–E409.

28. Wallace DC (2005) A mitochondrial paradigm of metabolic and degenerative diseases,aging, and cancer: A dawn for evolutionary medicine. Annu Rev Genet 39:359–407.

29. Nemoto S, Fergusson MM (2005) Finkel T SIRT1 functionally interacts with the metabolicregulator and transcriptional coactivator PGC-1�. J Biol Chem 280:16456–16460.

30. McDouall RM, Dunn MJ, Dubowitz V (1990) Nature of the mononuclear infiltrate and themechanism of muscle damage in juvenile dermatomyositis and Duchenne muscular dys-trophy. J Neurol Sci 99:199–217.

31. Goldspink G, Fernandes K, Williams PE, Wells DJ (1994) Age-related changes in collagengene expression in the muscles of mdx dystrophic and normal mice. Neuromuscul Disord4:183–191.

32. Chung HY, et al. (2009) Molecular inflammation: Underpinnings of aging and age-relateddiseases. Ageing Res Rev 8:18–30.

33. Glaros T, Larsen M, Li L (2009) Macrophages and fibroblasts during inflammation, tissuedamage and organ injury. Front Biosci 14:3988–3993.

34. Ravaglia G, et al. (2007) Blood inflammatory markers and risk of dementia: The ConseliceStudy of Brain Aging. Neurobiol Aging 28:1810–1820.

35. Armoni M, Harel C, Karnieli E (2007) Transcriptional regulation of the GLUT4 gene: FromPPAR-� and FOXO1 to FFA and inflammation. Trends Endocrinol Metab 18:100–107.

36. CesariM,etal. (2003) Inflammatorymarkersandcardiovasculardisease(TheHealth,Agingand Body Composition [Health ABC] Study). Am J Cardiol 92:522–528.

37. Mootha VK, et al. (2003) PGC-1�-responsive genes involved in oxidative phosphorylationare coordinately downregulated in human diabetes. Nat Genet 34:267–273.

38. LagougeM,etal. (2006)Resveratrol improvesmitochondrial functionandprotectsagainstmetabolic disease by activating SIRT1 and PGC-1�. Cell 127:1109–1122.

20410 � www.pnas.org�cgi�doi�10.1073�pnas.0911570106 Wenz et al.

See Retraction Published December 19, 2016