Role of PGC-1α in acute and low-grade inflammation Olesen.pdf · Summary The aim of the present...

Role of PGC-1α in acute and low-grade inflammation

FACULTY OF SCIENCE

UNIVERSITY OF COPENHAGEN

PhD thesis by Jesper Olesen

Academic supervisor: Professor Henriette Pilegaard

http://www.anyvitamins.com/anti-aging-supplement.htm

http://holistikhealth.com/blog/superfoods/resveratrol/resveratrol-2

1

Contents……………………………………………………………………………………………………….1

Acknowledgements ............................................................................................................................... 3

List of abbreviations ............................................................................................................................. 6

Summary .............................................................................................................................................. 8

Dansk resumé (danish summary) ........................................................................................................ 10

Introduction ........................................................................................................................................ 12

Aging ........................................................................................................................................................... 12

Inflammation................................................................................................................................................ 13

Low-grade inflammation ............................................................................................................................. 13

Causes of low-grade inflammation .......................................................................................................... 14

Metabolic effects of low-grade inflammation ......................................................................................... 15

Health beneficial and anti-inflammatory effects of exercise ....................................................................... 16

Resveratrol: A natural exercise mimetic? ................................................................................................... 17

The transcriptional co-activator PGC-1 ................................................................................................... 18

PGC-1α-mediated regulation in skeletal muscle ......................................................................................... 19

PGC-1α and inflammation .......................................................................................................................... 20

PGC-1 as a potential mediator of exercise-induced adaptations in skeletal muscle ................................ 21

Upstream activation of PGC-1α .................................................................................................................. 22

Ca2+

-signaling ......................................................................................................................................... 22

ROS-signaling ......................................................................................................................................... 23

β-adrenergic signaling ............................................................................................................................. 23

p38 signaling ........................................................................................................................................... 23

AMPK-signaling ...................................................................................................................................... 23

SIRT1 ...................................................................................................................................................... 24

Resveratrol-mediated activation of PGC-1α ............................................................................................ 24

Objectives of the thesis ........................................................................................................................ 25

Methods .............................................................................................................................................. 26

Primary mouse cell cultures ........................................................................................................................ 26

LPS stimulation of primary mouse myotubes ......................................................................................... 26

Mouse models .............................................................................................................................................. 27

Whole body PGC-1α KO mice ................................................................................................................ 27

Muscle specific PGC-1α KO mice .......................................................................................................... 27

Muscle specific PGC-1α over-expression mice ....................................................................................... 28

LPS as a model of acute inflammation .................................................................................................... 29

Long-term exercise training and resveratrol supplementation ................................................................ 30

Human study ................................................................................................................................................ 31

Subjects.................................................................................................................................................... 31

Experimental setup .................................................................................................................................. 31

Exercise protocol ..................................................................................................................................... 31

2

Endurance test and DXA scanning .......................................................................................................... 32

Muscle biopsies and blood samples ......................................................................................................... 32

Analyses ....................................................................................................................................................... 32

Plasma cytokines ..................................................................................................................................... 32

RNA isolation and Reverse Transcription ............................................................................................... 32

Real-time PCR ......................................................................................................................................... 33

Muscle lysate ........................................................................................................................................... 34

Protein determination and preparation of samples for western blotting .................................................. 34

SDS-PAGE and western blotting ............................................................................................................ 35

Homogenates for determination of enzyme activities and protein carbonyl content ............................... 37

Determination of protein carbonyl content .............................................................................................. 37

Statistics ................................................................................................................................................... 38

Integrated discussion .......................................................................................................................... 39

Acute inflammation ...................................................................................................................................... 39

Role of skeletal muscle PGC-1α in acute inflammation .......................................................................... 39

Chronic low-grade inflammation ................................................................................................................ 41

Age-associated low-grade inflammation ................................................................................................. 41

Role of PGC-1α in low-grade inflammation ........................................................................................... 43

Exercise training-induced adaptations in skeletal muscle of elderly men ............................................... 44

Physical activity and low-grade inflammation ........................................................................................ 44

Role of PGC-1α in exercise training-induced anti-inflammatory effects ................................................ 48

Metabolic and anti-inflammatory effects of resveratrol? ........................................................................ 49

Role of PGC-1α in resveratrol-mediated metabolic and anti-inflammatory effects ................................ 50

Conclusion .......................................................................................................................................... 53

Closing remarks and future perspectives ............................................................................................ 54

Reference list ...................................................................................................................................... 55

Appendix ............................................................................................................................................ 74

Study I and co-authorship statement Study II and co-authorship statement Study III and co-authorship statement

3

Acknowledgements

The present PhD thesis is based on studies performed at the Molecular and Integrative Physiology Section,

Department of Biology at University of Copenhagen in the period 2010-2013. Funding was provided by the

Faculty of Science, University of Copenhagen and by Centre of Inflammation and Metabolism,

Rigshospitalet, University of Copenhagen. I would like to acknowledge the help and support I have received

from several people during my time as a PhD student.

First and foremost I would like to thank Henriette Pilegaard for the inspiring, enthusiastic and ever so

present guidance and support throughout my time as a PhD student in her research group.

I would also like to thank all the present and former members of the “HP lab” for an inspirational

and enjoyable scientific environment, and especially:

o My office-mate, Rasmus S. Biensø for our daily discussions and for his co-work in Study III

o Stine Ringholm for her co-work in study II

I would also like to thank present and former colleagues at Centre of Inflammation and

Metabolism for all the inspiring scientific sessions (CIM workshops).

Further, I would like to thank Ylva Hellsten for sharing her expertise in primary cell cultures and her

collaboration on study I and III.

Karina Olsen is acknowledge for her technical assistance with the primary cell culture experiments

in study I

Lasse Gliemann is acknowledged for his comprehensive work and collaboration in study III

I would also like to thank Laurie Goodyear, who supervised my work during my stay at Joslin

Diabetes Center, and for giving me the opportunity to visit her lab and work with the talented

researchers. I also want to thank the members of the “Goody lab” for helping me with practical and

technical issues during my stay.

Finally, I owe my family, friends and not least my lovely girlfriend Line a great thank for their invaluable

support, encouragements and patience, without it would not have been possible.

4

Role of PGC-1α in acute and low-grade inflammation

By Jesper Olesen, M.sc.

This thesis is based on the present review and the following manuscripts:

I. Jesper Olesen, Signe Larsson, Ninna Iversen, Simi Yousafzai, Ylva Hellsten, Henriette

Pilegaard (2012). Skeletal muscle PGC-1α is required for Maintaining an Acute LPS-induced TNFα

Response. PlosOne 7(2): 32222

II. Jesper Olesen, Stine Ringholm, Maja M. Nielsen, Christina T. Brandt, Jesper T. Pedersen,

Jens Halling, Laurie J. Goodyear, Henriette Pilegaard. Role of PGC-1α in exercise training- and

resveratrol-induced prevention of age-associated inflammation. In press, Exp Gerontol, 2013

III. Jesper Olesen, Lasse Gliemann, Rasmus S. Biensø, Jakob Schmidt, Ylva Hellsten, Henriette

Pilegaard. Exercise training, but not resveratrol, improves metabolic and inflammatory status in

human skeletal muscle of aged men. Submitted to J. Physiol.

The three manuscripts are included in the appendix and will be referred to as study I-III in the thesis.

5

The work performed during my PhD has additionally contributed to the following manuscripts:

Rasmus S. Biensø, Lasse Gliemann, Jesper Olesen, Jakob Schmidt, Ninna Iversen, Jørgen F.

Wojtaszewski, Ylva Hellsten, Henriette Pilegaard. Exercise training, but not resveratrol improves

glucose metabolism in elderly men. In preparation.

Stine Ringholm, Jesper Olesen, Jesper T. Pedersen, Ylva Hellsten, Henriette Pilegaard. Effect

of lifelong resveratrol supplementation and exercise training on skeletal muscle oxidative capacity in

aging mice; impact of PGC-1α. In pending review Exp Gerontol.

Lasse Gliemann, Jakob Schmidt, Jesper Olesen, Rasmus S. Biensø, Sebastian Peronard, Simon

Grandjean, Stefan Mortensen, Michael Nyberg, Jens Bangsbo, Henriette Pilegaard, Ylva

Hellsten. Resveratrol Blunts the Positive Effects of Exercise Training on Cardiovascular Health in

Aged Men. J. Physiol 2013.

6

List of abbreviations

HAD; 3-hydroacyl-CoA dehydrogenase

AMPK; AMP-activated protein kinase

ATF2; activating transcription factor 2

AICAR; 5-aminoimidazole-4-carboxamide-1-beta-4-ribofuranoside

CamK; Ca2+/calmodulin-dependent protein kinase

cAMP; cyclic AMP

CD68; cluster of differentiation 68

COXI; cytochrome c oxidase I

CBP/p300; CREB binding protein

CREB; cAMP response element-binding protein

CRP; C-reactive protein

CS; citrate synthase

CXCL-1; chemokine (C-X-C motif) ligand 1

Cyt c; cytochrome c

DMEM; Dulbecco’s Modified Eagle’s Medium

DMSO; Dimethyl sulfoxide

DPBS; Dulbecco’s Phosphate Buffered Saline

EDTA; ethylenediaminetetraacetic

EMR-1; EGF-like module-containing mucin-like hormone receptor-like 1

Epac; exchange protein directly activated by cAMP

ER; endoplasmic reticulum

ERR; estrogen-related receptor

F4/80; rodent homolog to EMR-1

FFA; free fatty acid

FOX; forkhead box

GCN5; histone acetyltransferase

GPX1; glutathione peroxidase 1

HEK; human embryonic kidney

IL; interleukin

IL-1ra; interleukin receptor antagonist

IκB; inhibitor of κB

IKK; inhibitor of κB kinase

iNOS; inducible nitric oxide synthase

IRS-1; insulin receptor substrate 1

JNK; c-Jun N-terminal kinase

KO; knockout

LPS; lipopolysaccharide

MCP1; monocyte chemoattractant protein 1

MCK; creatine kinase promotor

MEF; myocyte enhancer factor

7

MKO; muscle specific knockout

NAD+; nicotine adenine dinucleotide

NF-κB; nuclear factor kappa-light-chain-enhancer of activated B cells

NRF; nuclear respiratory factor

NSAIDS; non-steroidal anti-inflammatory drugs

p38; p38 mitogen-activated protein kinase

p65; p65 subunit of NF-κB

PGC-1α; peroxisome proliferator-activated receptor-γ co-activator

PPAR; peroxisome proliferator-activated receptor

PRC; PGC-1 related co-activator

ROS; reactive oxygen species

S-AT; subcutaneous adipose tissue

SDS; sodium dodecyl sulphate

SIRT1; sirtuin 1

SkM; skeletal muscle

SOD; superoxide dismutase

SRC-1; steroid receptor co-activator 1

TG PGC-1α; transgenic PGC-1α overexpression in skeletal muscle

TLR; Toll-like receptor

TNFα; tumor necrosis factor α

TACE; TNFα converting enzyme

Tfam; mitochondrial transcription factor A

V-AT; visceral adipose tissue

V-CAM; vascular cell adhesion molecule

VEGF; vascular endothelial growth factor

WT; wild type

8

Summary

The aim of the present thesis was to examine the role of the exercise-induced transcriptional co-activator,

PGC-1α, in acute and low-grade inflammation. To investigate this, the following three hypotheses were

tested: 1) Skeletal muscle PGC-1α plays an important role in acute LPS-induced systemic inflammation as

well as in the inflammatory response in mouse skeletal muscle. 2) Long-term exercise training and/or

resveratrol supplementation prevents age-associated low-grade- and skeletal muscle inflammation in mice

with PGC-1α being required for these improvements. 3) Exercise training and/or resveratrol supplementation

reduces systemic- as well as skeletal muscle inflammation in aged human subjects.

Study I demonstrated an impaired LPS-induced plasma TNFα and skeletal muscle TNFα response in PGC-

1α muscle specific knockout mice compared with WT mice. Conversely, mice with transgenic

overexpression of PGC-1α in skeletal muscle showed a greater fold increase in plasma TNFα than WT mice,

when stimulated with LPS. Taken together, these results suggest that skeletal muscle PGC-1α is required for

a robust LPS-induced TNFα response.

Study II demonstrated that plasma TNFα and IL-6 as well as liver TNFα mRNA and protein, visceral

adipose tissue TNFα mRNA and skeletal muscle TNFα protein were all increased in old mice (15 month old)

compared with young mice (3 month old), confirming that aging is associated with systemic low-grade

inflammation and tissue inflammation in mice.

Study II and III demonstrated that exercise training reduced skeletal muscle TNFα protein content and

systemic IL-6 levels in mice and skeletal muscle TNFα mRNA content in aged human subjects. This

importance of physical activity in reducing inflammation is supported by results from our inactivity study (7

days of bed-rest) in young men, showing increased inflammation as evidenced by enhanced skeletal muscle

IL-6 mRNA and adipose tissue iNOS mRNA content. In conjunction, these results may indicate that skeletal

muscle inflammation is inversely related to the level of physical activity. However, no clear association

between the physical activity level and the level of systemic inflammation existed in aged exercise trained

human subjects or in young inactive human subjects.

Study II further demonstrated that PGC-1α was required for the exercise training-induced prevention of an

age-associated increase in skeletal muscle TNFα protein content. However, PGC-1α was not mandatory for

the exercise training-induced reductions in systemic IL-6 in mice, suggesting that additional factors

contribute to the systemic anti-inflammatory effects of exercise training.

Study II and III demonstrated that while resveratrol increased the protein content of the anti-oxidant enzyme

GPX1 and appeared to reduce oxidative stress in mouse skeletal muscle, resveratrol did not elicit any anti-

inflammatory effects neither in mice nor in human subjects. In contrast, resveratrol even impaired the

exercise training-induced reduction in protein carbonylation and TNFα mRNA in human subjects.

9

Study II demonstrated that the minor effects on GPX1 and oxidative stress observed with resveratrol in mice

were independent of PGC-1α and these findings were further supported by preliminary in vitro data showing

that the resveratrol-induced up-regulation of cyt c mRNA in primary myotubes was independent of PGC-1α.

In conclusion, skeletal muscle PGC-1α was required for a robust LPS-induced TNFα response. No anti-

inflammatory effects of resveratrol were observed in mice or human subjects. Skeletal muscle inflammation

was increased by physical inactivity in humans and reduced by exercise training in mice and human subjects.

The observed anti-inflammatory effect of exercise training in mice was partly mediated through PGC-1α.

10

Dansk resumé (danish summary)

Formålet med denne afhandling var at undersøge den trænings-inducerede transkriptionelle co-activator,

PGC-1α’s, rolle i akut og lav-grads inflammation. For at belyse dette blev følgende tre hypoteser undersøgt:

1) Ekspressionen af PGC-1α i skeletmuskulaturen spiller en afgørende rolle for et akut LPS-induceret

inflammatorisk respons systemisk og i skeletmuskulaturen. 2) Langvarig træning og/eller resveratrol tilskud

forhindrer aldersassocieret lav-grads inflammation og inflammation i skeletmuskulaturen og PGC-1α er

nødvendig for mediering af disse gavnlige effekter. 3) Træning og/eller resveratrol tilskud reducerer

systemisk inflammation og inflammation i skeletmuskulaturen i ældre mænd.

Studie I viser, at mus med muskel specifik knockout af PGC-1α har et forringet LPS-induceret plasma TNFα

respons sammenlignet med WT mus. Modsat har mus med overexpression af PGC-1α i skeletmuskulaturen

et større plasma TNFα respons end WT mus ved stimulering med LPS. Sammenholdt tyder disse resultater

på, at PGC-1α i skeletmuskulaturen er nødvendig for et robust LPS-induceret respons.

Studie II viser at plasma TNFα og IL-6 samt lever TNFα mRNA og protein, TNFα mRNA i det viscerale

fedtvæv og TNFα protein i skeletmuskulaturen alle er forøget i gamle (15 mdr.) mus i forhold til unge (3

mdr.) mus, hvilket bekræfter at øget alder er associeret med systemisk inflammation såvel som inflammation

i forskellige væv i mus.

Studie II og III viser at træning reducerer TNFα protein i skeletmuskulaturen såvel som systemisk IL-6 i mus

samt TNFα mRNA i skeletmuskulaturen hos ældre mænd. At fysisk aktivitet har stor betydning for

reducering af inflammation understøttes af resultater fra vores inaktivitets-studie (7 dages sengeleje) udført

med unge raske mænd. Dette viser øget inflammation i form af øget IL-6 mRNA indhold i

skeletmuskulaturen og øget iNOS mRNA indhold i fedtvævet efter sengelejet. Sammenholdt kunne disse

resultater hermed tyde på at der eksisterer en omvendt sammenhæng mellem inflammation i

skeletmuskulaturen og niveauet af fysisk aktivitet. Når man kigger systemisk viser studie II og sengeleje-

studiet dog ingen sammenhæng mellem inflammationsniveauet og niveauet af fysisk aktivitet.

Studie II viser desuden at PGC-1α er nødvendig for de gavnlige effekter af træning, heriblandt hindring af

den aldersassocierede stigning i TNFα protein i skeletmuskulaturen. PGC-1α er derimod ikke nødvendig for

den træningsinducerede gavnlige reducering af systemisk IL-6 i mus, hvilket indikerer at andre faktorer end

PGC-1α er involveret i medieringen af de systemiske anti-inflammatoriske effekter ved træning.

Studie II og III viser, at mens resveratrol øger protein indholdet af antioxidanten GPX1 og i tillæg

tilsyneladende reducerer niveauet af oxidativt stress i skeletmuskulaturen hos mus, har resveratrol derimod

ingen anti-inflammatoriske effekter i hverken mus eller mennesker. Modsat forhindrer resveratrol ligefrem

de gavnlige effekter ved træning såsom trænings-induceret sænkning af protein carbonylering og TNFα

mRNA niveauet i ældre mænd.

Studie II viser ydermere at de mindre effekter af reveratrol på GPX1 protein indholdet og niveauet af

oxidative stress i skeletmuskulaturen er uafhængigt af PGC-1α. Disse fund understøttes af de foreløbige in

11

vitro data, der viser at resveratrol uafhængigt af PGC-1α opregulerer cyt c mRNA indholdet i primære

muskel-cellekulturer.

Samlet kan det konkluderes, at ekspression af PGC-1α i skeletmuskulaturen er nødvendig for et robust LPS-

induceret TNFα respons. Resveratrol har ingen anti-inflammatoriske effekter i hverken mus eller mennesker.

Der eksisterer en omvendt sammenhæng mellem inflammation i skeletmuskulaturen og niveauet af fysisk

aktivitet i mus og mennesker og PGC-1α er nødvendig for den gavnlige træningsinducerede hindring af

aldersassocieret stigning i TNFα protein i skeletmuskulaturen.

12

Introduction

The prevalence of lifestyle-related diseases like type 2 diabetes, cardiovascular diseases and different cancers

has increased dramatically during the last decades (Booth et al., 2012;Pedersen, 2009). It has been estimated

that the global prevalence of type 2 diabetes alone will increase from 170 million people in 2000 to 438

million people in 2030 (Wild et al., 2004). Such an unfortunate development will increase the financial

burden to all societies and influence the quality of life for each individual affected. Type 2 diabetes,

cardiovascular disease and various cancers are all characterized as metabolic diseases and many factors are

likely involved in the initiation and/or progression of these diseases. However, a strong association exists

between metabolic diseases and lifestyle-related factors such as physical inactivity and obesity (Amati et al.,

2009;Koivisto et al., 1986;Venables & Jeukendrup, 2009). The observation that metabolic diseases often co-

exist suggests that a common factor may underlie these pathologies (Handschin & Spiegelman,

2008;Pedersen, 2009;Wellen & Hotamisligil, 2005). Chronic low-grade inflammation is a likely candidate

gathering these maladies as chronic low-grade inflammation has been associated with the majority of these

diseases (Handschin & Spiegelman, 2008;Pedersen, 2009;Wellen & Hotamisligil, 2005;Woods et al., 2012).

Although these diseases may also occur in young people, they are particularly present in elderly people

(Woods et al., 2012).

Aging

Aging is directly linked with numerous lifestyle-related diseases with several tissues and organs affected

(Masoro, 2001;Woods et al., 2012). In skeletal muscle, loss of muscle mass and strength (Brooks &

Faulkner, 1994;Doherty et al., 1993), decreased oxidative capacity (Conley et al., 2000;Short et al.,

2005;Zahn et al., 2006) and reduced endogenous anti-oxidant capacity (Chabi et al., 2008;Finkel &

Holbrook, 2000;Wei et al., 1998) are all deteriorations associated with the aging process. In 1956 “The free

radical theory of aging” was put forward by Harman, suggesting that the pathophysiological cellular changes

observed with aging were a consequence of the accumulation of free radicals, which in the course of time has

deleterious effects on macromolecules such as DNA, proteins and phospholipids (Harman 1956). Harman

later revised the theory into “The mitochondrial theory of aging” acknowledging that mitochondria are the

primary source of free radicals during aging (HARMAN, 1972). Since then, these theories have been

debated, but it is generally accepted that excessive leak of reactive oxygen species (ROS) from mitochondria

combined with an impaired anti-oxidant system are central components in the development of oxidative

stress during aging (Chabi et al., 2008;Conley et al., 2000;Sarkar & Fisher, 2006;Woods et al., 2012).

Mitochondrial dysfunction caused by ROS-induced mitochondrial DNA damages may further worsen the

excessive ROS production from the mitochondria and eventually lead to a vicious cycle involving

13

inflammatory processes resulting in systemic chronic low-grade inflammation (Sarkar & Fisher,

2006;Schreck et al., 1992;Woods et al., 2012).

Inflammation

It is important to distinguish chronic low-grade inflammation from acute inflammation as these processes are

fundamentally distinct although similar signaling pathways and inflammatory mediators are involved in both

processes (fig. 1). Acute inflammation is a conserved mechanism evolved to protect organisms from foreign

pathogens by inducing a transient increase in pro-inflammatory cytokines like tumor necrosis factor (TNF)α

and interleukin (IL)-6 as a mean to orchestrate immune cells to destroy invading microbes or toxins and

repair of damaged tissues. Classical immune cells like monocytes, macrophages and dendritic cells as well as

endothelial cells and intestinal epithelial are traditionally seen as the primary responders during acute

inflammation. However other cell types like adipocytes, hepatocytes and muscle cells also have the ability to

express and secrete cytokines (Frost et al., 2002;Hotamisligil et al., 1993;Kudo et al., 2009), which may

indicate that these otherwise metabolic tissues contributes to the production of inflammatory mediators

during acute inflammation. While acute inflammation is essential for survival for all species, chronic low-

grade inflammation is more likely a result of metabolic surplus and/or physical inactivity (Handschin &

Spiegelman, 2008;Pedersen, 2009).

Low-grade inflammation

Chronic low-grade inflammation is characterized as a condition with sustained 2-4 folds elevations in

circulating levels of pro-inflammatory cytokines like TNFα and IL-6 (Bruunsgaard & Pedersen, 2003;Woods

et al., 2012). Chronic low-grade inflammation is thought to be initiated in the adipose tissue in response to

obesity and subsequently affect other organs including skeletal muscle and the liver (Shoelson et al.,

2006;Wellen & Hotamisligil, 2005). The first link between obesity and inflammation was reported in rats

where a high caloric diet led to increased expression of TNFα in adipose tissue (Hotamisligil et al., 1993).

These findings were later supported by the observations that TNFα expression was elevated in adipose tissue

and skeletal muscle from obese individuals (Hotamisligil et al., 1995a;Kern et al., 1995;Saghizadeh et al.,

1996). Since then several reports have supported this link by showing a positive correlation between obesity

and systemic low-grade inflammation (Hammett et al., 2006;Nicklas et al., 2004;Verdaet et al., 2004;Wu et

al., 2013). Chronic low-grade inflammation is also a common manifestation of aging (Woods et al., 2012)

and increased levels of inflammatory cytokines like TNFα and IL-6 and acute phase proteins like C-reactive

protein (CRP), are often observed in aged individuals compared with young (Ballou et al., 1996;Bruunsgaard

et al., 2001;Cohen et al., 1997;Dobbs et al., 1999;Ershler et al., 1993;Paolisso et al., 1998;Wei et al., 1992).

14

However, whether the age-associated low-grade inflammation is due to aging per se or due to an altered

lifestyle is not clear.

Causes of low-grade inflammation

Multiple factors may underlie the obesity-driven inflammation. Among others, hyperlipidemia with excess

lipid uptake in to adipocytes has been demonstrated to cause endoplasmic reticulum (ER) stress, which

activates inflammatory signaling pathways like c- Jun N-terminal kinase (JNK) and nuclear factor kappa-

light-chain-enhancer of activated B cells (NF-κB) (Hung et al., 2004;Ozcan et al., 2004). Moreover, the

elevated levels of circulating free fatty acids (FFA) in obesity may also directly trigger inflammatory

processes in macrophages and adipose tissue through direct interaction with Toll-like receptors (TLRs) and

the subsequent activation of JNK and NF-κB (Nguyen et al., 2007;Shi et al., 2006). In addition, ROS have

also been suggested to trigger inflammatory processes in adipocytes in states of obesity (Lin et al.,

2005;Wellen & Hotamisligil, 2005) through activation of JNK and NF-κB (Schreck et al., 1992;Wellen &

Hotamisligil, 2005). Infiltration of macrophages into expanded and necrotic adipose tissue has also been

suggested as a considerable source of pro-inflammatory cytokines in rodent models of obesity (Weisberg et

al., 2003;Xu et al., 2003). Thus, several mechanisms may trigger inflammatory signaling pathways in

various tissues and together this may lead to chronic low-grade inflammation in states of obesity.

The shift towards low-grade inflammation as people age may in large part be ascribed to the concomitant

age-associated increase in adiposity (Schaap et al., 2012;Wu et al., 2007) and the obesity-driven

inflammation outlined above. However, age-associated elevations in ROS (HARMAN, 1956) may also

directly trigger inflammatory processes through NF-κB (Schreck et al., 1992), JNK (Ventura et al., 2003)

and through interactions with TLRs (Gill et al., 2010). This may indicate the aging independently of obesity

and increases in fat mass (Sarkar & Fisher, 2006;Wei et al., 1998;Wellen & Hotamisligil, 2005) leads to low-

grade inflammation.

NF-κB is considered as a central regulator of inflammation as NF-κB both responds to and passes on

inflammatory signals. The NF-κB gene family consists of five members RelA/p65, RelB, c-Rel, p100/p52

and p105/p50. These polypeptides form homo- or hetero dimmers, of which the p65/p50 heterodimer is the

most abundant (Hoffmann & Baltimore, 2006). The mechanism by which NF-κB is activated is complex. In

un-stimulated conditions, inhibitor of κB (IκB) binds to NF-κB and masks the nuclear localization signals of

NF-κB keeping NF-κB sequestered in an inactive state in the cytoplasm (Jacobs & Harrison, 1998). Upon

stimulation/cellular stress, IκB kinase (IKK) dependent phosphorylation of IκB and subsequent proteolytic

degradation of IκB results in NF-κB translocation and increased transcriptional activity (Carpenter &

O'Neill, 2009;Hoffmann & Baltimore, 2006).

15

Figure 1. Schematic model of inflammatory signaling pathways. These pathways are activated by extracellular mediators such as cytokines,

free fatty acids (FFAs) and lipopolysaccharide (LPS) or by intracellular stresses such as endoplasmic reticulum (ER) stress or excess reactive

oxygen species (ROS) production by mitochondria. Signals from all these mediators converge on the inflammatory signaling pathways c-Jun

N-terminal protein kinese (JNK) and inhibitor of κB kinase (IKK). When activated these pathways phosphorylate nuclear factor κB (NF-κB)

and activating protein (AP)-1, which subsequently increases the production of inflammatory cytokines like tumor necrosis factor (TNF)α and

interleukin (IL)-6, through transcriptional regulation. Excessive ROS production from the mitochondria during aging may also trigger

inflammatory signaling pathways and subsequent production of cytokines (modified from Wellen et al. (Wellen & Hotamisligil, 2005)).

Metabolic effects of low-grade inflammation

Chronic elevations in systemic TNFα and IL-6 have been reported in obese and type 2 diabetes patients

(Hotamisligil et al., 1995b;Muller et al., 2002;Pedersen et al., 2003;Pickup et al., 1997;Plomgaard et al.,

2007) and TNF has directly been shown to impair insulin signaling in skeletal muscle and adipocytes

(Hotamisligil et al., 1996;Plomgaard et al., 2005). In vitro incubations of adipocytes with TNF has shown

that TNF inhibits insulin signaling through a JNK-mediated serine phosphorylation of insulin receptor

substrate-1 (IRS-1) (Hotamisligil et al., 1996). This has further been confirmed in human studies with TNF

infusion, which in addition showed an inhibition of Akt substrate 160 in skeletal muscle downstream of IRS-

1 (Plomgaard et al., 2005). Based on these observations it has been suggested that TNFα may be a primary

cause of insulin resistance in type 2 diabetes (Hotamisligil et al., 1996;Plomgaard et al., 2005). In contrast,

recombinant infusion of IL-6 in human subjects do not seem to impair insulin signaling in skeletal muscle

(Steensberg et al., 2003b;Wolsk et al., 2010), suggesting that IL-6 is not a primary cause of metabolic

diseases but rather a consequence.

16

Independently of TNF, the induction of JNK and NF-κB by other stimuli like ER stress and fatty acids may

also impair insulin action through serine phosphorylation of IRS-1 (Aguirre et al., 2000;Gao et al.,

2004;Ozcan et al., 2004;Yuan et al., 2001). In addition, some studies have reported that pharmacological

inhibition of NF-κB in rodents and humans is associated with improved whole body insulin sensitivity (Yuan

et al., 2001;Fleischman et al., 2008;Hundal et al., 2002), although other studies in rodents have failed to

show this associations (Cai et al., 2004;Polkinghorne et al., 2008;Rohl et al., 2004). Nonetheless, elevated

muscle NF-κB activity has consistently been reported in patients with type 2 diabetes (Reyna et al.,

2008;Sriwijitkamol et al., 2006;Tantiwong et al., 2010;Yuan et al., 2001). Despite some divergence, these

observations suggest that both IKK/NF-κB and JNK signaling, in addition to induce transcription of pro-

inflammatory cytokines (Karin et al., 1997;Schreck et al., 1992;Ventura et al., 2003), may also interfere with

metabolic processes and thereby be involved in the pathogenesis of insulin resistance (Wellen &

Hotamisligil, 2005). Hence, several overlaps between metabolic and inflammatory pathways exist, which

highlights the complexity of these processes.

Nevertheless, reducing low-grade inflammation may be a promising approach to improve age-associated

metabolic complications (Pedersen, 2006;Woods et al., 2012). Several pharmacological compounds have

been developed with anti-inflammatory properties like statins and non-steroidal anti-inflammatory drugs

(NSAIDS) reviewed in (Corsonello et al., 2010). However, these approaches are very expensive and a

plethora of side effects are often associated with the use of these compounds. Instead, lifestyle interventions

such as exercise training and diet supplements may provide a long-term solution to reduce inflammation

especially during aging (Woods et al., 2012).

Health beneficial and anti-inflammatory effects of exercise

Exercise training elicits a broad range of health beneficial effects on several parameters in multiple organs

(Pedersen & Saltin, 2006). In aged subjects, exercise training has been shown to counteract loss of muscle

mass and strength (Frontera et al., 1988;Hollmann et al., 2007), which may improve overall muscle function.

In addition, exercise training increases the oxidative capacity of skeletal muscle (Gollnick et al.,

1973;Henriksson & Reitman, 1977;Holloszy, 1967). Together with an exercise training-induced

improvement in blood lipid profile and a reduction in fasting plasma glucose concentration, these adaptations

collectively improve peripheral insulin sensitivity (Dela et al., 1992;Mikines et al., 1988;Perseghin et al.,

1996) and subsequently reduce the metabolic load on several tissues including the liver, adipose tissue and

skeletal muscle. In addition, exercise training reduces fat mass, which further will improve the general blood

lipid profile and reduce the obesity-related inflammation ((Pedersen & Saltin, 2006). However, exercise

training is also suggested to have anti-inflammatory effects independently of fat loss (Gleeson et al.,

2011;Handschin & Spiegelman, 2008;Pedersen, 2006;Woods et al., 2012). Among others, the anti-

17

inflammatory effects of exercise may be ascribed to an exercise-induced increase in the release of adrenaline,

cortisol, growth hormone, and other factors with immune-regulatory effects (Harbuz et al., 2003;Ignatowski

et al., 1996;Nieman, 2003). Moreover, acute exercise and exercise training have been shown to down-

regulate the mRNA expression of several TLRs on monocytes (Gleeson et al., 2006). As both acute exercise

and IL-6 infusions have been shown to reduce a lipopolysaccharide (LPS)-induced increase in plasma TNFα

(Starkie et al., 2003), contraction-induced expression and release of IL-6 (Keller et al., 2001;Steensberg et

al., 2001) is suggested to contribute to the anti-inflammatory effects of exercise. Whereas classical pro-

inflammatory cytokines like TNFα and IL-1 are not induced by exercise, exercise evokes an increase in

circulating anti-inflammatory cytokines including IL-1 receptor antagonist (IL-1ra), IL-10 and soluble TNF-

receptor (Ostrowski et al., 1999). In line with this, IL-6 infusions in humans have been shown to stimulate

the production of IL-1ra and IL-10 (Steensberg et al., 2003a), supporting that exercise-induced IL-6 release

from contracting skeletal muscle promotes an anti-inflammatory environment (Brandt & Pedersen, 2010).

The mechanisms by which skeletal muscle contributes to the anti-inflammatory effects of exercise are not

completely understood, but emerging evidence suggests that the exercise-induced transcriptional co-activator

peroxisome proliferator-activated receptor-γ co-activator (PGC)-1α regulates the expression of inflammatory

mediators in skeletal muscle (Handschin et al., 2007b;Handschin et al., 2007a;Wenz et al., 2009), which

may contribute to the anti-inflammatory effects of exercise. This will be addressed further below in the

section “PGC-1α and inflammation”.

Taken together, these observations highlight that exercise training elicits anti-inflammatory effects by an

intricate interplay between organs and cytokines. Moreover, such anti-inflammatory effects of exercise may

contribute to the beneficial effects of exercise training during aging. However not all individuals have the

ability or desire to perform regular physical activity and alternatives to induce metabolic adaptations are

warranted. Interestingly, the natural anti-oxidant resveratrol has been proposed to exert “exercise like”

metabolic effects (Baur et al., 2006;Lagouge et al., 2006;Timmers et al., 2011;Um et al., 2010).

Resveratrol: A natural exercise mimetic?

Resveratrol (3, 5, 4’-trihydroxystilbene) is a phytoalexin that belongs to the stilbene class of compounds with

anti-oxidant properties (Olas & Wachowicz, 2005;Stojanovic et al., 2001). Resveratrol is present in various

fruits including dark grapes, where it acts as a natural antibiotic and as an inhibitor of proliferation

(Harikumar & Aggarwal, 2008). Over the last decade, resveratrol has been given tremendous attention due to

the numerous reports showing health beneficial metabolic effects that in many ways are similar to the effects

of exercise training (Baur et al., 2006;Dolinsky et al., 2012;Lagouge et al., 2006;Timmers et al., 2011;Um et

al., 2010). Specifically, resveratrol has been shown to protect rodents from diet-induced obesity, insulin

resistance and inflammation (Baur et al., 2006;Kim et al., 2011;Lagouge et al., 2006;Um et al.,

2010;Pearson et al., 2008) and resveratrol has even been shown to increase lifespan of lower species (Howitz

18

et al., 2003;Valenzano et al., 2006;Wood et al., 2004) and of mice on a high caloric diet (Baur et al., 2006).

Furthermore, resveratrol has been shown to increase the oxidative capacity of skeletal muscle (Baur et al.,

2006;Lagouge et al., 2006;Um et al., 2010) and to reduce skeletal muscle oxidative stress (Jackson et al.,

2011). In humans, the metabolic effects of resveratrol are less clear. Only few trials have been reported in the

literature and these show divergent results (Brasnyo et al., 2011;Crandall et al., 2012;Poulsen et al.,

2012;Skrobuk et al., 2012;Timmers et al., 2011;Yoshino et al., 2012). While a few studies in obese

individuals and type 2 diabetes patients have shown resveratrol-mediated improvements in overall glucose

regulation (Brasnyo et al., 2011;Crandall et al., 2012;Timmers et al., 2011) and small reductions in plasma

TNFα (Timmers et al., 2011), other studies have failed to show such improvements (Poulsen et al.,

2012;Yoshino et al., 2012). This highlights the need for additional human trials with resveratrol to fully

understand the metabolic consequences of resveratrol in humans.

The molecular mechanisms by which resveratrol exerts its metabolic effects in skeletal muscle are

controversial (Baur et al., 2006;Lagouge et al., 2006;Park et al., 2012;Price et al., 2012;Um et al., 2010).

Resveratrol was originally discovered to robustly increase the activity of the NAD+-dependent deacetylase

SIRT1 (Howitz et al., 2003) and studies have subsequently supported this observation in skeletal muscle and

liver of mice (Price et al., 2012;Baur et al., 2006;Lagouge et al., 2006). However, the ability of resveratrol to

directly activate SIRT1 has later been questioned (Borra et al., 2005;Kaeberlein et al., 2005;Park et al.,

2012;Um et al., 2010). Moreover, studies in AMPK-α1 and AMPK-α2 knockout (KO) mice have reported

that the beneficial effects of resveratrol are dependent on AMPK (Um et al., 2010). Interestingly, resveratrol

has in C2C12 cells been shown to inhibit phosphodiesterase 4 causing concomitant increases in the cytosolic

cAMP levels, subsequently activating a complex signaling cascade involving Epac1-CamkII-CamKKβ-

AMPK-SIRT1, which ultimately increases the activity of PGC-1α (Park et al., 2012).

Taken together, both the metabolic effects of resveratrol and the mechanisms by which resveratrol exerts its

effects have many similarities to exercise training. The alleged resveratrol-induced activation of PGC-1α

highlights the promising therapeutic potential of resveratrol in the prevention of lifestyle- and age-related

metabolic diseases.

The transcriptional co-activator PGC-1

Much attention has been given to PGC-1α since the discovery in 1998 (Puigserver et al., 1998), due to its

numerous metabolic effects in various tissues. PGC-1α was discovered in a yeast two-hybrid screen using

brown adipose tissue cDNA library looking for candidate proteins binding to PPARγ during cold exposure.

In this process PGC-1α was highlighted as a transcriptional co-activator important for adaptive

thermogenesis (Puigserver et al., 1998). Since then, it has been shown that PGC-1α is implicated in the

regulation of a broad range of transcription factors including other peroxisome proliferator-activated

19

receptors (PPARs) (Puigserver et al., 1998), nuclear respiratory factors (NRFs) (Wu et al., 1999), myocyte

enhancer factors (MEFs) (Handschin et al., 2003;Michael et al., 2001), estrogen related receptors (ERRs)

(Huss et al., 2002), forkhead box (FOX) (Puigserver et al., 2003) and possibly also NF-κB (Alvarez-Guardia

et al., 2010;Eisele et al., 2013). PGC-1α is primarily expressed in highly metabolic and oxidative tissues like

brown adipose tissue, skeletal muscle, heart, kidney and liver (Wu et al., 1999). Upon binding to

transcription factors, PGC-1α recruits the steroid receptor co-activator 1 (SRC-1) and CREB binding protein

(CBP/p300), which possess histone acetyltransferase activity leading to increased promoter activity by

modifying chromatin structures on DNA in promoter regions of target genes (Puigserver et al., 1999).

Moreover, PGC-1α contains a RNA binding motif in the C-terminal end and two serine/arginine rich areas

that interact with RNA polymerase II important for pre-initiation and processing of mRNA (Monsalve et al.,

2000). In addition, the C-terminal end of PGC-1α contains a thyroid hormone receptor associated protein

(TRAP) complex that enables the transcriptional machinery connected with PGC-1α to access the DNA

bound transcription factors (fig. 2). When bound to a transcription factor, PGC-1α serves as docking site for

the transcriptional machinery, which is suggested to explain the powerful co-activation capacity of PGC-1α

(Handschin & Spiegelman, 2006).

Figure 2. Structure and functional domains of the PGC-1α gene (Rodgers et al., 2008).

PGC-1β and PGC-1 related co-activator (PRC) are two structural homologs to PGC-1α and together forms

the PGC-1 family. Both PGC-1β and PRC regulates many of the same metabolic functions as PGC-1α

(Andersson & Scarpulla, 2001;Lin et al., 2002a), however the present PhD thesis will primarily focus on

PGC-1α.

PGC-1α-mediated regulation in skeletal muscle

Perhaps the best-described function of PGC-1α is its role as a master regulator of mitochondrial biogenesis in

skeletal muscle (Lin et al., 2002b;Puigserver et al., 1998;Wende et al., 2007;Wu et al., 1999). By ectopic

overexpression of PGC-1 in myotubes, PGC-1α has been shown to robustly induce the mRNA expression

of several genes involved in the oxidative phosphorylation (OXPHOS) including cytochrome c, Cytchrome c

oxidase II (COXII), COXIV and ATP synthase (Wu et al., 1999). Through the use of protein-protein binding

assays and by deletion of certain promoter regions this induction was demonstrated to be mediated via direct

20

binding of PGC-1 to NRF1 (Wu et al., 1999). These studies also revealed that the mRNA content of NRF1,

NRF2 and the activity of the mitochondrial transcription factor A (Tfam) were increased by overexpression

of PGC-1 in myotubes (Wu et al., 1999). Interestingly, both NRF1 and NRF2 regulate the expression of the

nuclear encoded Tfam, which in turn translocate to the mitochondria and binds to a D loop on mitochondrial

DNA, thereby activating replication and transcription of mitochondrial encoded genes (Clayton,

1991;Virbasius & Scarpulla, 1994). Together, these findings in myotubes highlight the ability of PGC-1 to

coordinately regulate the expression of nuclear and mitochondrial encoded genes. Importantly, studies in

mice have later supported these findings. Hence, transgenic muscle specific overexpression of PGC-1 (TG

PGC-1) and inducible overexpression in skeletal muscle increased the expression of OXPHOS genes in

skeletal muscle (Calvo et al., 2008;Lin et al., 2002b;Wende et al., 2007), whereas whole body KO and

muscle specific KO of PGC-1 (MKO) mice have reduced protein levels of OXPHOS genes (Geng et al.,

2010;Handschin et al., 2007b;Leick et al., 2008;Lin et al., 2004). Through coactivation of ERR, PGC-1

also regulates the expression of medium-chain-acyl-CoA dehydrogenase important in the β-oxidation of fatty

acids (Huss et al., 2002). Intriguingly, overexpression of PGC-1 in skeletal muscle transforms otherwise

“white” glycolytic muscle fibers in to “red” oxidative muscles. In concert with these findings, TG PGC-1

mice exhibit increased running exercise capacity compared with littermate WT mice (Calvo et al., 2008),

whereas MKO PGC-1 mice accordingly have reduced running endurance capacity (Handschin et al.,

2007a). Moreover, TG PGC-1 mice display reduced respiratory exchange ratio-values compared with WT

mice at any given exercise intensity during a running test, reflecting increased oxidation of fatty acids in TG

PGC-1 mice relative to WT (Calvo et al., 2008).

Importantly, PGC-1α also regulates anti-oxidant enzymes like superoxide dismutase (SOD)1 and SOD2

(Handschin et al., 2007b;Leick et al., 2008;Leick et al., 2010;St-Pierre et al., 2006;Wenz et al., 2009). This

parallel regulation of anti-oxidant enzymes is vital as the enhanced capacity for oxidative phosphorylation

increases the capacity for production of ROS molecules. However, the mechanism by which PGC-1α

regulates anti-oxidant enzymes is still not fully understood.

PGC-1α and inflammation

Several lines of evidence suggest that PGC-1α may also have a role in regulating inflammatory processes.

First of all, inflammatory stimuli seem to regulate the activity of PGC-1α (Alvarez-Guardia et al.,

2010;Puigserver et al., 2001). The first evidence came when Puigserver et al. (2001) in C2C12 cells showed

that a cocktail of cytokines, through a p38-mediated phosphorylation increases the activity of PGC-1α

(Puigserver et al., 2001). Conversely, studies in cardiomyocytes have shown that TNFα incubations

decreased the expression of PGC-1α (Palomer et al., 2009). The same group later showed that direct binding

of p65 to PGC-1α inhibited the activity of PGC-1 and the concomitant metabolic effects after stimulation

with TNFα (Alvarez-Guardia et al., 2010). Studies in human aortic smooth muscle cells have shown that

21

adenoviral overexpression of PGC-1α suppresses TNFα-induced NF-kB activity as well as V-CAM and

MCP1 expression (Kim et al., 2007). Together this indicates that TNF impairs PGC-1-mediated

regulation and that PGC-1 prevents TNF-induced inflammation. The latter is also supported by a recent

study in C2C12 cells, showing that adenoviral overexpression of either PGC-1α and PGC-1β differentially

repressed a pro-inflammatory cytokine response after TNFα, FFA and TLR-agonist stimulation (Eisele et al.,

2013). These observations were in part mediated through repressed p65 phosphorylation, which lowers the

transcriptional activation-potential of NF-kB (Eisele et al., 2013). Moreover, it was shown that while

overexpression of PGC-1α does not alter basal expression of inflammatory mediators, overexpression of

PGC-1β repressed the basal binding of p65 and p50 (NF-κB subunits) to DNA, thereby reducing basal

expression of TNFα and IL-6 (Eisele et al., 2013). In rodent models, overexpression of PGC-1α in rat tibialis

anterior by electroporation has been reported to reduce NF-kB activity (Brault et al., 2010) and PGC-1α

MKO mice have accordingly been shown to exhibit increased basal mRNA levels of TNFα and IL-6

(Handschin et al., 2007b). In line with this, TG PGC-1α mice have reduced age-associated increases in

TNFα and IL-6 mRNA and protein content and reduced serum TNFα and IL-6 levels in old mice compared

with age-matched WT mice (Wenz et al., 2009). Finally, human cross-sectional studies have reported an

inverse correlation between the mRNA content of TNFα and PGC-1α and between the mRNA content of IL-

6 and PGC-1α in skeletal muscles independently of BMI (Handschin et al., 2007b).

Taken together, these reports indicate that PGC-1α may exert anti-inflammatory effects in several tissues,

possibly through interactions with the NF-kB pathway. Hence, activation of PGC-1α and the concomitant

anti-inflammatory effects by regular physical activity or by diet supplements like resveratrol may serve as an

important tool to attenuate or even prevent low-grade inflammation and the associated pathologies. However

the exact mechanism behind the alleged anti-inflammatory effects of PGC-1α in the regulation of the NF-kB

pathway and other inflammatory signaling pathways is still not clarified.

PGC-1 as a potential mediator of exercise-induced adaptations in skeletal muscle

The possibility that PGC-1α serves as a mediator of exercise training-induced adaptations in skeletal muscle

arose from studies in rats (Baar et al., 2002;Terada et al., 2002) and humans (Pilegaard et al., 2003),

reporting transient increases in PGC-1α transcription and mRNA content in skeletal muscle in recovery from

a single exercise bout (Baar et al., 2002;Pilegaard et al., 2003). Interestingly, transient hypomethylation of

the PGC-1α promoter has more recently been reported after acute exercise followed by increases in the PGC-

1α mRNA transcript (Barres et al., 2012). In addition, PGC-1α mRNA and protein levels are increased in

trained subjects relative to untrained subjects (Russell et al., 2003;Short et al., 2003;Trappe et al., 2013). The

importance of PGC-1α in exercise-induced adaptations has been confirmed in different PGC-1α KO mouse

models (Chinsomboon et al., 2009;Geng et al., 2010;Leick et al., 2010). Hence, it has been shown that PGC-

1α is required for training–induced increases in mitochondrial biogenesis in skeletal muscle (Geng et al.,

22

2010) and VEGF content (Chinsomboon et al., 2009;Geng et al., 2010;Leick et al., 2010). In addition PGC-

1 has been shown to be required for an exercise training-induced prevention of age-associated reductions in

citrate synthase activity and SOD2 protein content in skeletal muscle (Leick et al., 2010). Importantly, it

should be noted that other findings have revealed that PGC-1α is not mandatory for exercise training-induced

increases in cyt c, COXI, SOD2 protein content in skeletal muscle (Geng et al., 2010;Leick et al., 2008),

suggesting that additional factors contribute to the exercise training-induced adaptations in skeletal muscle.

Moreover, studies in humans have shown that skeletal muscle PGC-1α mRNA levels decline with increasing

age (Ling et al., 2004) and short-term physical inactivity (Alibegovic et al., 2010). In addition, type 2

diabetes patients also have reduced skeletal muscle PGC-1α mRNA content potentially due to

hypermethylation of the PGC-1α promoter (Barres et al., 2009). Collectively these findings emphasize that

PGC-1 is a central mediator of exercise-induced adaptations in skeletal muscle, which may also involve

regulation of inflammatory mediators.

Upstream activation of PGC-1α

Several initiating stimuli and intracellular signaling pathways have been shown to induce transcription and/or

activation PGC-1α. These include many of the traditional signaling pathways and factors typically induced

by exercise.

Ca2+

-signaling

Exercise-induced Ca2+

signaling has been reported as a likely mechanism to increase PGC-1α content and/or

activity (Irrcher et al., 2003;Kusuhara et al., 2007;Ojuka et al., 2003;Wright et al., 2007). Hence, in vitro

muscle incubations with the calcium ionophores ionomycin and caffeine (Kusuhara et al., 2007;Ojuka et al.,

2003) and ex vivo electrical stimulation of rat extensor digitorum longus muscle (Kusuhara et al., 2007) all

increased the PGC-1α mRNA content and this induction was abolished in conditions when co-treated with

cyclosporin A (calcineurin inhibitor) and KN-62 (CamK inhibitor). Moreover, it has been shown that mice

with transgenic overexpression of the Ca2+

/calmodulin-dependent kinase (CamK)IV in skeletal muscle have

increased PGC-1α mRNA content (Wu et al., 2002), together providing strong evidence that cytosolic

calcium concentrations is involved in inducing the PGC-1α gene in contracting skeletal muscle.

Additionally, Ca2+

signaling through calcineurin has been shown to induce an oxidative phenotype in muscle

cells (Olson & Williams, 2000) and MEF2 transcription factors seem to be important for a Ca2+

/calcineurin-

induced fiber type switch (Wu et al., 2002). PGC-1α binds MEF2 transcription factors (Handschin et al.,

2003;Michael et al., 2001) and overexpression of PGC-1α increases the co-activation of MEF2c and MEF2d

in C2C12 cells (Lin et al., 2002b). Through co-activation of MEF2, PGC-1α induces the transcription of

myofibrillar proteins associated with an oxidative muscle (Lin et al., 2002b). Interestingly, MEF2

transcription factors bind to the promoter of PGC-1α and regulate its transcription, suggesting that an

23

autoregulatory loop exists, where PGC-1α regulates its own transcription through co-activation of MEF

proteins (Handschin et al., 2003).

ROS-signaling

Exercise increases ROS production in skeletal muscle (Davies et al., 1982) and ROS have been shown to

increase PGC-1α mRNA content in muscle cells (Irrcher et al., 2009;Silveira et al., 2006). Moreover, anti-

oxidants have been shown to prevent an electrical stimulation-induced increase in PGC-1α mRNA in

primary rat muscle cells (Silveira et al., 2006) as well as a H2O2-induced increase in the promoter activity of

PGC-1α in C2C12 cells (Irrcher et al., 2009). Together this indicates that exercise-induced ROS signaling

may also contribute to the exercise-induced increase in PGC-1 transcription and mRNA content.

β-adrenergic signaling

Based on reports showing that the β2-adrenergic agonist clenbuterol increases PGC-1α mRNA content in

mice (Miura et al., 2007;Chinsomboon et al., 2009) and that propranolol, a non-selective β-adrenergic

antagonist blunts an exercise-induced induction of PGC-1α mRNA (Miura et al., 2007), it is suggested that

adrenaline is also implicated in the induction of PGC-1α mRNA during exercise.

p38 signaling

Muscle contractions, cell stress and cytokines have all been shown to increase the activity of p38 in skeletal

muscle (Akimoto et al., 2005;Puigserver et al., 2001;Raingeaud et al., 1995). Moreover, p38 has in C2C12

cells been shown to phosphorylate PGC-1 on three residues (Ser265

, Thr262

and Thr298

), resulting in

stabilization and increased activity of PGC-1 (Puigserver et al., 2001). P38 also phosphorylates activating

transcription factor (ATF)2, which binds to a CREB region in the PGC-1 promoter thereby inducing the

transcription of PGC-1 (Cao et al., 2004). Thus it is likely that contraction-induced p38 signaling both

increases the activity of PGC-1 (Puigserver et al., 2001) in addition to inducing PGC-1 transcription

through ATF2 (Cao et al., 2004).

AMPK-signaling

The energy sensor AMPK, which is activated during exercise (Winder & Hardie, 1996;Wojtaszewski et al.,

2000), also seems to be involved in the induction of PGC-1α mRNA (Jorgensen et al., 2005). Studies in mice

have revealed that the AMPK activator 5-aminoimidazole-4-carboxamide-1-beta-4-ribofuranoside (AICAR)

requires the AMPK-α2 subunit to increase PGC-1α mRNA content (Jorgensen et al., 2005), whereas,

exercise-induced PGC-1α mRNA expression does not require the presence of AMPK-α2 subunit (Jorgensen

et al., 2005). In addition, AMPK phosphorylates PGC-1α on two residues (Thr177

and Ser538

) increasing the

stability and activity of PGC-1α (Jager et al., 2007). Together this shows an AMPK-mediated regulation of

PGC-1α activity and transcription. However, the role of AMPK in exercise-induced gene regulation remains

unclear.

24

SIRT1

Acetylation status is another important posttranslational modification that regulates the activity of PGC-1α.

The acetyltransferase GCN5 has been shown to acetylate PGC-1α thereby reducing the activity of PGC-1

and the concomitant regulation of gluconeogenesis in the liver (Lerin et al., 2006). Conversely, SIRT1 has

been demonstrated to increase the transcriptional activity of PGC-1α through deacetylation on several lysine

residues in HEK (human embryonic kidney) cells and in the liver of fasted mice (Rodgers et al., 2005).

Interestingly, studies in C2C12 cells have indicated that AMPK-mediated phosphorylation primes PGC-1α

for subsequent NAD+ dependent deacetylation by SIRT1 (Canto et al., 2010). This suggests the existence of

a central AMPK-SIRT1-PGC-1α axis, which coordinately regulates metabolic genes in conditions with

increased energy demand such as fasting and exercise.

Resveratrol-mediated activation of PGC-1α

As mentioned, studies in both rodents and humans have also shown that resveratrol via AMPK and/or

SIRT1-mediated signaling pathways increases the content and/or activity of PGC-1α in muscle and liver

(Baur et al., 2006;Lagouge et al., 2006;Park et al., 2012;Timmers et al., 2011), however whether PGC-1α is

required to mediate the metabolic effects of resveratrol has not been established.

Taken together, most of the traditional exercise-induced signaling pathways (β-adrenergic signaling, p38,

AMPK, ROS and Ca2+

signaling) as well as resveratrol seem to regulate the abundance and/or activity of

PGC-1α in skeletal muscle (fig. 3), highlighting PGC-1α as a potential key factor coordinating the health

beneficial metabolic and anti-inflammatory effects of exercise training and resveratrol.

Figure 3. Schematic presentation of exercise-induced factors suggested to induce gene expression and/or activity of PGC-1α

and the concomitant PGC-1α-mediated regulation of metabolic genes (Modified from Olesen et al. (Olesen et al., 2010)).

25

Objectives of the thesis The overall aim of the PhD thesis was to investigate the role of PGC-1α in acute and low-grade

inflammation. The following hypotheses have been addressed:

Skeletal muscle PGC1-α plays an important role in acute LPS-induced systemic inflammation as

well as in the inflammatory response in mouse skeletal muscle.

Long-term exercise training and/or resveratrol supplementation prevents age-associated low-grade-

and skeletal muscle inflammation in mice with PGC-1α being required for these improvements.

Exercise training and/or resveratrol supplementation reduces systemic- as well as skeletal muscle

inflammation in aged human subjects.

26

Methods

Primary mouse cell cultures

Primary mouse cell cultures were used in study I.

C57BL/6 mice were euthanized by cervical dislocation and limb skeletal muscles were quickly dissected out

and placed in 15 ml ice cold Dulbecco’s phosphate buffered saline solution (DPBS; Invitrogen, Carlsbad,

CA, USA) containing 1% glucose and 0.5% penicillin/streptomycin (Invitrogen) and placed on ice. The

muscle tissue was carefully minced, transferred to 15 ml of Dulbecco’s modified Eagle medium (DMEM;

Invitrogen) containing 1% penicillin/streptomycin and 0.2% collagenase (Worthington, Freehold, NJ, USA)

and rotated for 90 min at 37 C. After centrifugation for 15 min at 300 g and 4C, the pellet was re-

suspended in 15 ml of DMEM containing 0.2% collagenase, 0.01% DNase (Sigma, St. Louis, MO, USA),

0.25% trypsin (Invitrogen) and 1% penicillin/streptomycin and rotated again for 30 min at 37C. The cell

suspension was diluted to 30 ml with primary growth medium (PGM) containing 10% horse serum

(Invitrogen), 10% foetal bovine serum (Invitrogen), 0.02% penicillin/streptomycin and 0.1% L-glutamine

(Sigma), and centrifuged for 15 min at 300 g, 4C. The supernatant was discarded, the pellet was re-

suspended in PGM and the suspension was filtered through a 70-mm sterile filter. Then 200,000 cells were

seeded onto 12 wells dishes coated with 1% Matrigel (Becton Dickinson, Stockholm, Sweden). Cells were

cultured at 37 C in a humidified atmosphere with 8% CO2. The PGM medium was exchanged on day 2, and

on day 4, half of the PGM medium was removed and replaced with an equal volume of new PGM. On day 6

(when myocytes had started to differentiate), all of the PGM was removed and replaced with fusion medium

(DMEM, Invitrogen) containing 10% horse serum and 0.1% L-glutamine. Cells were used for experiments

on day 7 or 8, when sufficient maturation of myotubes was observed.

LPS stimulation of primary mouse myotubes

On the day of the experiment, the cells were washed in DMEM without phenol red. Half of the cell cultures

were treated with 1.0 µg LPS/ml media and the other half was treated with DMEM without phenol red as

control. LPS was dissolved in dimethyl sulfoxide (DMSO) and then diluted in DMEM without phenol red.

After 2 hours of incubation, the medium was collected and the cells were harvested in Trizol reagent

(Invitrogen). The samples were stored at -80° C until analyzed. The experiment was repeated 3 times with n

= 6 in each experiment.

27

Mouse models

Whole body PGC-1α KO mice

The whole body PGC-1α KO mouse strain was used in study I and study II.

Generation

Whole body PGC-1α KO mice were originally generated by homolog recombination (Lin et al., 2004). A

targeting plasmid containing two loxP sites flanking exons 3-5 of the PGC-1α gene was constructed and a

Cre recombinase was used to generate the whole body PGC-1α KO strain (fig. 4).

Figure 4. Generation of the whole body PGC-1α knockout strain by homolog recombination (Lin et al., 2004).

Exon 3-5 encodes a highly conserved region in the PGC-1α gene required for the interaction with nuclear

receptors (Lin et al., 2004). Whole body PGC-1α KO and littermate WT mice were obtained for study I and

II by crossbreeding of heterozygous whole body PGC-1α KO parents. Homozygous PGC-1α-/-

mice were

used as whole body PGC-1α KO mice and PGC-1α+/+

littermate WT mice as controls.

Phenotype

Whole body PGC-1α KO mice are born at the expected Mendelian ratio (Lin et al., 2004) and weigh ~15 %

less than their littermate WT mice (Leick et al., 2008;Lin et al., 2004). In the fed state no differences in

plasma glucose levels are observed between WT and whole body PGC-1α KO, however in the fasted state

they develop mild hypoglycemia and they are resistant to diet-induced obesity (Lin et al., 2004). Whole body

PGC-1α KO mice exhibit behavioral characteristics resembling Huntington’s disease with sudden

uncontrolled movements (Lin et al., 2004) and they are easy to distinguish from WT mice. Whole body

PGC-1α KO have been reported to be hyperactive (Lin et al., 2004), however we do not observe this in our

animal facility and in fact PGC-1α KO mice run voluntarily less than WT mice when offered a running

wheel, (Leick et al., 2008;Leick et al., 2010).

Muscle specific PGC-1α KO mice

The PGC-1α MKO strain was used in study I.

28

Generation

The floxed PGC-1α gene construct used for generation of the PGC-1α MKO mice is similar to the construct

used for generation of the whole body PGC-1α KO strain with two loxP sites flanking exons 3-5 of the PGC-

1α gene (Fig 4) except that the neomyocin-cassette is not present in the PGC-1α MKO mice. Homozygous

floxed PGC-1α mice were bred with transgenic myogenin-Cre+/-

mice creating PGC-1α MKO mice (PGC-

1αloxP/loxP

, myogenin-Cre+/-

) and littermate control mice (PGC-1αloxP/loxP

, myogenin-Cre-/-

). RT-real time PCR

on cDNA from skeletal muscle tissue using specific primers flanking the excised exons revealed that the

efficacy of the Cre recombinase was ~80 %, meaning that ~20 % of the PGC-1αloxP/loxP

, myogenin-Cre-/-

mice

maintained a similar PGC-1α mRNA level in skeletal muscle as WT mice. A 90 % knockdown of PGC-1α in

skeletal muscle was set as the minimum criterion for the inclusion of PGC-1α MKO mice in study I.

Phenotype

The PGC-1α MKO strain used in study I is similar to the previously reported strain by Geng et al. (Geng et

al., 2010). PGC-1α MKO mice from this strain have similar body weight as littermate control mice on chow

and on a high fat diet (unpublished observations). PGC-1α MKO mice have been reported to run equally as

much as control littermate mice when offered a running wheel (Geng et al., 2010). While blood glucose

levels in PGC-1α MKO and littermate control mice are similar in the fed state and after a 4-hour fasting

period, mild hypoglycemia seems present in PGC-1α MKO after a 16-hour fasting period compared with

littermate control mice (unpublished observations).

Muscle specific PGC-1α over-expression mice

The TG PGC-1α mouse strain was used in study I

Generation

Generation of TG PGC-1α mice has been described previously (Lin 2002, Puigserver 2001). Briefly, Mice

overexpressing PGC-1α specifically in skeletal muscle and heart was generated by cloning of a full length

PGC-1α cDNA into a expression vector containing the constitutive muscle creatine kinase (MCK) promoter.

TG PGC-1α mice were obtained by crossbreeding heterozygous TG PGC-1α mice and C57BL/6 mice, and

heterozygous TG PGC-1α and littermate WT mice were used in study I.

Phenotype

TG PGC-1α mice have similar body weight as littermate WT mice on chow and high fat diets (Choi et al.,

2008). Overexpression of PGC-1α in skeletal muscle induces a muscle phenotype-switch from white

glycolytic muscles into red oxidative muscles (Lin et al., 2002b). In accordance, TG PGC-1α exhibit

increased endurance exercise running capacity relative to littermate WT mice (Calvo et al., 2008). While TG

PGC-1α mice have normal glucose homeostasis on a regular chow diet, they are paradoxically prone to high

29

fat diet-induced insulin resistance in skeletal muscle (Choi et al., 2008). In contrast, when combining high fat

diet with exercise training, TG PGC-1α mice have improved glucose homeostasis compared with littermate

WT mice (Summermatter et al., 2013).

LPS as a model of acute inflammation

LPS was used as a model to examine the role of PGC-1α in acute inflammation in study I.

LPS is a natural component of the cell walls of gram-negative bacteria and LPS administration has often

been used to simulate acute inflammation in humans and rodents (Andreasen et al., 2008;Frost et al.,

2002;Meador et al., 2008). LPS triggers an inflammatory response through binding to TLRs and subsequent

activation of inflammatory signaling pathways including IKK/NF-κB (Zhang & Ghosh, 2001), mitogen

activated protein kinase p38 (Carpenter & O'Neill, 2009) and JNK (Hambleton et al., 1996). Activation of

these pathways leads to enhanced transcription of a vast array of genes encoding inflammatory cytokines,

including TNFα and IL-6 (Carpenter & O'Neill, 2009;Collart et al., 1990;Davis, 2000;Hambleton et al.,

1996;Karin et al., 1997;May & Ghosh, 1998;Shimizu et al., 1990). TLRs are expressed on the surface of

several cell types (Carpenter & O'Neill, 2009) and especially TLR4 has been shown to bind LPS, which

subsequently initiates a pro-inflammatory signaling cascade (Chow 1999).

Prior to initiation of study I, a pilot mouse study was conducted to establish the time course of a LPS-

induced inflammatory response systemically and in skeletal muscle with samples obtained 2h, 4h and 6h post

injection of LPS or saline. The findings revealed that the plasma levels and skeletal mRNA levels of the

inflammatory cytokines TNFα and IL-6 peaked 2 hours post injection of LPS and almost returned to baseline

levels 6 hours after the LPS administration relative to saline (fig. 5). Based on these observations the 2 h time

point was chosen for the main experiment (study I). Control mice not injected were also included in the pilot

study to examine the effect of the injection per se. Similar plasma- and skeletal muscle cytokine levels were

observed between saline injected and control mice, indicating that the injection per se did not induce

inflammation. Despite these results, we chose saline injected mice as controls for the main experiment (study

I).

30

Figure 5. Plasma TNFα (a), plasma IL-6 (b), skeletal muscle (SkM) TNFα mRNA (c), SkM IL-6 mRNA 2, 4 and 6 h after

injection of either saline or 0.8 µg lipopolysaccharide (LPS) per gram mouse as well as control (Con) mice not injected.

Values are presented as mean ± SE, n = 4.

Ten weeks old PGC-1α KO, PGC-1α MKO, TG PGC-1α and their respective littermate WT mice were given

an intraperitoneal injection of either saline as control or 0.8 mg LPS per gram mouse. All mice were

euthanized by cervical dislocation followed by decapitation to collect trunk blood 2 hours post injection.

Quadriceps muscles were quickly removed from all three mouse strains and quick-frozen in liquid nitrogen.

In addition visceral adipose tissue and liver were also removed from whole body PGC-1a KO and littermate

WT mice and quick-frozen. The samples were stored at -80° C until analyses. Each group consisted of 10

mice with equal number of male and female mice.

Long-term exercise training and resveratrol supplementation

In study II, whole body PGC-1α KO and littermate WT mice were at three months of age randomly divided

into groups consisting of: untrained mice receiving rodent chow, untrained mice receiving chow

supplemented with resveratrol, voluntary exercise trained mice having access to a running wheel receiving

chow, and voluntary exercise trained mice having access to a running wheel and receiving chow

supplemented with resveratrol. Running distance and duration was monitored by a regular cycle computer

and differences between WT and PGC-1α KO mice were daily adjusted by wheel blocking of WT mice for

shorter periods to ensure similar exercise distance between the different genotypes and interventions. The

concentration of 4 g resveratrol per kg food was chosen based on previous reports in mice (Lagouge et al.,

2006;Um et al., 2010) and corresponds to ~0.7 mg per gram mouse per day. The interventions lasted from 3

months to 15 months of age. In addition, 3 months old mice receiving chow served as young controls. Each

31

of the groups consisted of 8-10 mice. All mice were euthanized by cervical dislocation followed by

decapitation to collect trunk blood. Quadriceps muscles, perigonadal visceral adipose tissue (V-AT), inguinal

subcutaneous adipose tissue (S-AT) and liver were quickly removed, quick-frozen in liquid nitrogen and

stored at -80° C until analyses.

Human study

Subjects

Forty-three aged physically inactive, but otherwise healthy male subjects participated in study III. All

subjects were non-smokers and underwent a medical examination. None had been diagnosed with

cardiovascular disease, hypertension, renal dysfunction, insulin resistance or type 2 diabetes and all subjects

had normal ECG. Two subjects were diagnosed with hypercholesterolemia regulated by their own physician

(medication was maintained during the experimental period) whereas the other participants had normal

cholesterol levels.

Experimental setup

Study design

The study was divided into two parts, which both were 8-week randomized, double-blinded placebo

controlled trials. The subjects in the first part were assigned to either a combination of exercise training and

placebo (n = 13) or exercise training and 250 mg resveratrol·day-1

(Fluxome Inc., Stenlose, Denmark, n =

14). The subjects in the second part were assigned to either placebo (n = 7) or 250 mg resveratrol·day-1

(n =

9). The allocation was based on body mass index (BMI), blood glucose, cholesterol and maximal oxygen

consumption (table S1, suppl. information and Gliemann et al., 2013). All participants were instructed to

take one tablet each morning. Subjects noted time of consumption for each tablet and any discomfort that

might appear throughout the intervention period. Furthermore, the subjects were instructed not to change

their normal eating or drinking habits as well as their general activity level or way of living throughout the

experimental period.

Exercise protocol

The exercise training intervention consisted of supervised high intensity spinning interval training (cycle

ergometer) 2 times/week and full body circuit training (crossfit) 1 time/week. In addition, the subjects were

instructed to walk 5 km ones per week. The intensity of the exercise bouts was monitored and controlled by

heart rate monitors (Polar, Kempele, Finland).

32

Endurance test and DXA scanning

On the first experimental day, a dual-energy X-ray absorptiometry (DXA) scanning was performed in

addition to an incremental time to exhaustion one-leg knee-extensor exercise test. After acclimatizing and a

short warm-up, the test started at 6 W and gradually increased with 6 W every 5 minute until exhaustion. The

total energy output (KJ) was calculated based on duration (seconds) and workload (Watt).

Muscle biopsies and blood samples

On the second experimental day (minimum 48 hours after the first experimental day), the subjects arrived

after an overnight fast and resting blood samples were taken from an arm vein and a vastus lateralis muscle

biopsy was obtained under local anesthesia (lidocaine; AstraZeneca, Södertälje, Sweden) using the

percutaneous needle biopsy technique (Bergstrom, 1975) with suction. Muscle biopsies were quick-frozen in

liquid nitrogen and stored (-80° C) until analysis. The two experimental days were repeated in the same order

after the 8-week intervention.

Analyses

Plasma cytokines

Plasma cytokines were analyzed using an ultra-sensitive MSD multi-spot 96 well assay system pre-coated

with antibodies (MesoScaleDiscovery, Gaithersburg, Maryland, USA) according to manufacturer’s protocol.

Briefly, after blocking the wells with a blocking agent, samples and a standard were dispensed into the wells

in duplicates and incubated overnight at 4° C on an orbital shaker. After washing the wells with PBST (3x),

the samples were incubated with a detection antibody (secondary antibody) for 1 h at room temp. After

additional wash (3x with PBST), READ Buffer was applied to the wells and the MSD plates were measured

on a MSD Sector Imager 2400 plate reader. Raw data were obtained as electrochemiluminescence signal

(light) detected by photodetectors and analyzed using the Discovery Workbench 3.0 software (MSD). A

standard curve was generated for each analyte and used to determine the concentration of analytes in each

sample.

RNA isolation and Reverse Transcription

Total RNA was isolated from ~20 mg crushed mouse muscle, liver and adipose tissue in study I+II and from

~20 mg muscle tissue in study III with the guanidinium thiocyanate-phenol-chloroform method as previously

described (Chomczynski & Sacchi, 1987;Pilegaard et al., 2000) and in the cell culture experiment with the

Trizol method following the manufacture’s guidelines (Invitrogen). The final pellets were re-suspended in

DEPC treated H2O containing 0.1mM ethylenediaminetetraacetic acid (EDTA). RNA was quantified by

33

measuring the absorbance at 260nm. Purity of the RNA samples was evaluated from 260nm/280nm ratio and

all samples were above 1.8.

Superscript II RNase H− system and Oligo dT (Invitrogen, Carlsbad, CA) were used to reverse transcribe the

mRNA to cDNA as described previously (Pilegaard et al., 2000) and the samples were diluted in nuclease-

free H2O. The amount of single-stranded DNA (ssDNA) was determined in each cDNA sample by use of

OliGreen reagent (Molecular Probes, Leiden, The Netherlands), as described previously (Lundby et al.,

2005).

Real-time PCR

Real-time PCR was performed using an ABI 7900 sequence-detection system (Applied Biosystems, Foster

City, CA). Primers and TaqMan probes for amplifying gene-specific mRNA fragments were designed using

the database from ensemble.org and Primer Express (Applied Biosystems) and obtained from

TagCopenhagen. The sequences are given in table 1. Real-time PCR was performed in triplicates in a total

reaction volume of 10 µl using Universal Mastermix (Applied Biosystems). Cycle threshold (Ct) was

converted to a relative amount by use of a standard curve constructed from a serial dilution of a pooled RT

sample run together with the samples. Target gene mRNA content was normalized to single-stranded cDNA

content determined by OliGreen reagent (Molecular Probes, Leiden, The Netherlands) as previously

described (Lundby et al., 2005) or GAPDH mRNA content or β-actin mRNA content when appropriate.

34

Table 1. Primer and TagMan probe sequences used in real time PCR.

Forward Primer Reverse primer TaqMan probe

Mouse genes

TNFα 5' ATGGCCCAGACCCTCACA 3' 5' TTGCTACGACGTGGGCTACA 3' 5' TCAGATCATCTTCTCAAAATTCGAGTGACAAGC 3'

IL-6 5’ GCTTAATTACACATGTTCTCTGGGAAA 3’ 5’ CAAGTGCATCATCGTTGTTCATAC 3’ 5’ ATCAGAATTGCCATTGCACAACTCTTTTCTCAT 3’

IL-10 5' AGAGAAGCATGGCCCAGAAAT 3' 5' CAGGGGAGAAATCGATGACA 3' 5' CAGGGGAGAAATCGATGACA 3'

TACE 5' TGCAAGGCTGGGAAATGC 3’ 5' TTG CACGAGTTGTCAGTGTCAA 3' 5' GCGCAGGACTCCAGCTCCTGCT 3'

F4/80 5' GGCTGCCTCCCTGACTTTC 3' 5' TGCACTGCTTGGCATTGC 3' 5' TCCTTTTGCAGTTGAAGTTTCCATATCCTTGG 3'

TLR4 5' TCTGATCATGGCACTGTTCTTCTC 3' 5' CTGATCCATGCATTGGTAGGTAATATTA 3' 5' CAGGAAGCTTGAATCCCTGCATAGAGGTAGTTC 3'

GAPDH 5' GGAAGGGCTCATGACCACAGT 3' 5' GCCCCACGGCCATCA 3' 5' TGGATGGCCCCTCTGGAAAGCTGT 3'

PGC-1α 5' AGCCAAACCAACAACTTTATCTCTTC 3' 5' TTAAGGTTCGCTCAATAGTCTTGTTC 3' 5' AGAGTCACCAAATGACCCCAAGGGTTCC 3'

Human genes

TNFα 5' TCTGGCCCAGGCAGTCAGAT 3' 5' AGCTGCCCCTCAGCTTGA 3' 5' CAAGCCTGTAGCCCATGTTGTAGCAAACC 3'

iNOS 5' AGCGGGATGACTTTCCAAGA 3' 5' TAATGGACCCCAGGCAAGATT 3' 5' CCTGCAAGTTAAAATCCCTTTGGCCTTATG 3'

PGC-1α 5' CAAGCCAAACCAACAACTTTATCTCT 3' 5' CACACTTAAGGTGCGTTCAATAGTC 3' 5' AGTCACCAAATGACCCCAAGGGTTCC 3'

Table 1. Tumor necrosis factor (TNF)α, interleukin (IL)-6, IL-10, TNFα converting enzyme (TACE), F4/80, toll-like receptor

(TLR)4, glyceraldehyde 3–phosphate dehydrogenase (GAPDH), inducible nitric oxide synthase (iNOS) and peroxisome

proliferator-activated receptor-γ co-activator (PGC-1α).

Muscle lysate

~20-30 mg crushed mouse muscle, liver and adipose tissue in study I+II and ~10-15 mg freeze-dried human

muscle biopsies, free of connective tissue, blood and visible fat in study III were homogenized (1:20 and

1:80 respectively) in an ice-cold buffer (10% Glycerol, 20 mM Na-pyrophosphate, 150 mM NaCl, 50 mM

Hepes, 1% NP-40, 20 mM β-glycerophosphate, 10 mM NaF, 1 mM EDTA, 1 mM EGTA, 2 mM PMSF, 10

µg/ml Aprotinin, 10 µg/ml Leupeptin, 2 mM Na3VO4, 3 mM Benzamidine, pH 7.5) for 3 min using a tissue

lyser (TissueLyser II; QIAGEN, Germany) with 30 oscillations per second. Homogenates were rotated end

over end for 60 min at 4° C. Lysates were generated by centrifugation at 16,000 g for 20 min at 4° C and

collection of the protein supernatant (lysate).

Protein determination and preparation of samples for western blotting

Protein content in lysates for western blotting and protein content in homogenates for determinations of

enzyme activity were measured by the bicinchoninic acid (BCA) method (Thermo Scientific, Rockford, IL,

USA). The principle of the assay is that cysteine, cystine, tyrosine and tryptophan amino acids in the samples

react with Cu2+

in the Pierce reagent to form Cu1+

. The reduction of Cu2+

to Cu1+

is proportional to the protein

content in the samples. Cu1+

binds to BCA and forms a purple-blue complex and absorbance measured at

550nm on a multiscanner (Multiscan, Thermo Scientific, Denmark) was converted to protein concentrations

by use of a known BSA standard run together with the samples.

35

Lysates were diluted to a final concentration of 2 µg·µl-1

in sample buffer containing sodium dodecyl

sulphate (SDS) and boiled for three minutes at 96° C.

SDS-PAGE and western blotting

Protein content as well as phophorylation of proteins were measured by SDS-PAGE and western blotting

using PVDF membrane and the semi-dry transfer technique. Equal amounts of total protein were loaded for

each sample in accordance to the antibody optimization (table 2). PVDF membranes were blocked for 1 hour

and incubated with a primary antibody followed by horseradish peroxidase-conjugated secondary antibody

(Dako, Denmark) according to optimizations given in table 2. Proteins were visualized and quantified using a

Carestream Image Station and Carestream MI 5.0 SE software. Protein content and phosphorylation are

expressed as arbitrary units relative to control samples loaded on each site of each gel.

36

Table 2. Western Blot details.

Antibody Manufacturer/

catalog nr.

Protein

Load

(µg)

Predicted

MW

(kda)

Blocking

Primary

antibody

conc.

Secondary

antibody

Acetylated

lysine residues CS #9441 15 - 3% FG 1:1,000 1:5,000 anti-rabbit

AMPK-α2 Grahame Hardie 20 63 3% Milk 1:20,000 1:15,000 anti-sheep

AMPKThr172 CS #2535 20 62 3% FG 1:1,000 1:5,000 anti-rabbit

Catalase Abcam #1877 15 59 5% Milk 1:5,000 1:5,000 anti-rabbit

COXI Invitrogen

#459600 10 37 3% FG 1:3,000 1:3,000 anti-mouse

Cyt c BD Pharmigen

# 556433 10 15 3% FG 1:1,000 1:10,000 anti-mouse

GPX1 Abcam #22604 15 22 3% FG 1:1,000 1:5,000 anti-rabbit

GAPDH CS #2118 15 37 3% FG 1:1,000 1:2,000 anti-rabbit

EMR-1 Thermo sci. #

PA5-21399 15 97 3% FG 1:1,000 1: 5,000 anti-rabbit

IκB-α CS #9242 15 48 3% FG 1:1,000 1:5,000 anti-rabbit

IκB-β CS #9248 15 39 3% FG 1:1,000 1:5,000 anti-rabbit

IKK CS #2678 25 89 3% FG 1:1,000 1:5,000 anti-rabbit

IKKser176, 180 CS #2697 25 89 3% FG 1:500 1:5,000 anti-rabbit

iNOS CS #2977 25 130 3% FG 1:1,000 1:5,000 anti-rabbit

JNK CS #9252 20 46, 54 5% BSA 1:1,000 1:5,000 anti-rabbit

JNKThr183, Tyr185 CS #9251 20 46, 54 5% BSA 1:1,000 1:2,000 anti-rabbit

P38 CS #9212 15 43 3% FG 1:1,000 1:2,000 anti-rabbit

p38Thr180, Tyr182 CS #4511 15 43 3% FG 1:1,000 1:2,000 anti-rabbit

P65 CS #4764 15 65 3% FG 1:1,000 1:5,000 anti-rabbit

p65Ser536 CS #3033 15 65 3% FG 1:1,000 1:5,000 anti-rabbit

SIRT1 CS #2493 120 120 3% FG 1:1,000 1:1,000 anti-rabbit

SOD2 Millipore

#06-984 15 24 3% FG 1:1,000 1:10,000 anti-rabbit

TNFα CS #3707 25-40 Precursor,

27

3% FG or

5% BSA 1:1,000

1:5000

Anti-rabbit

Table 2. AMP-activated protein kinase (AMPK), cytochrome c oxidase (COX)I, cytochrome c (cyt c), glutathione peroxidase

(GPX)1, glyceraldehydes 3-phosphate dehydrogenase, EGF-like module-containing mucin-like hormone receptor (EMR)-1,

inhibitors of κB (IκB), inhibitors of κB kinase (IKK), inducible nitric oxide synthase (iNOS), c-Jun N-terminal kinase (JNK),

p38 mitogen activated protein kinase (p38), p65 (p65 subunit of NF-κB), sirtuin (SIRT)1, superoxide dismutase (SOD)2,

tumor necrosis factor (TNF)α, Cell Signaling (CS), molecular weight (MW), fish gel (FG), bovine serum albumin (BSA).

37

Homogenates for determination of enzyme activities and protein carbonyl content

Crushed wet weight mouse quadriceps muscle in study II and freeze-dried human muscle biopsies, free of

connective tissue, blood and visible fat in study III were homogenized (1:80 and 1:400, respectively) in 0.3

M phosphate-buffer (pH 7.7) containing 0.05% bovine serum albumin by use of a tissue lyser (3 min, 30 s−1

)

(TissueLyser II; QIAGEN). These homogenates were later used for measurements of enzyme activity and

protein carbonylation.

Citrate synthase activity

Citrate synthase (CS) catalyzes the reaction between acetyl-CoA and oxaloacetate to form citric acid (citrate)

in the TCA cycle. Maximal CS activity was determined in study II and III by the principles of Lowry et al.

(Lowry et al., 1978). Hydrolysis of the thioester of acetyl-CoA results in formation of CoA-SH. The thiol

group reacts with DTNB in the reation mixture to form a yellow product NTB, which is kinetically

determined at 405 nm (Multiscan, Thermo Scientific) at baseline and after addition of oxaloacetate. CS

activity was normalized to protein content measured in the respective samples (BCA method).

3-hydroxyacyl-CoA dehydrogenase

3-hydroxyacyl-CoA dehydrogenase (HAD) catalyzes the oxidation of β-hydroxyacyl-CoA and NAD+ to

form NADH, H+ and β-ketoacyl-CoA in the fatty acid oxidation pathway. In study III, HAD activity was

kinetically determined at 355nm/460nm (excitation/emission) by the principles of Lowry et al. (Lowry et al.,

1978). After addition of acetoacetyl-CoA, the delta emission was converted to activity and HAD activity was

normalized to protein content measured in the respective samples (BCA method).

Determination of protein carbonyl content

Carbonylation of proteins is a marker of oxidative stress. Protein carbonylation is an irreversible oxidative

modification that targets proteins for proteolytic degradation or alternatively accumulation of high molecular

weight aggregates that may interfere with various processes in the cell. Protein carbonyl content was

determined in muscle homogenates (0.3 M phosphate-buffer, described above) in study II+III using an

OxiSelectTM

ELISA-kit (Cell Biolabs, SD, USA) according to manufacturer’s protocol. Briefly, BSA

standards and protein samples were absorbed onto a 96-well plate for 2 hours at 37° C. After addition of

dinitro phenylhydrazine (DNPH), the carbonyl groups on the proteins are derivitized to DNP hydrazone.

After incubation with an anti-DNP antibody followed by a HRP conjugated secondary antibody, the

absorbance was measured at 450 nm (Multiscan, Thermo Scientific). Based on a serial diluted

oxidized/reduced BSA standard run together with the samples, the absorbance of the unknown samples was

converted to protein carbonyl concentration and normalized to protein content in the respective samples

(BCA method).

38

Statistics

Study I: Two-way analysis of variance (ANOVA) was used to test the main effect of LPS and genotype

within each mouse strain. If normality or variance of the data-set were skewed, the data were logarithmically

transformed before applying the ANOVA test. Student Newman Keuls post hoc test was used to locate

differences when applicable. The non-parametric Mann-Whitney U test was applied when the equal variance

test or the normality test failed even after logarithmically transformation. A t-test was used to test if there

was a difference in the basal level between genotypes within each mouse strain.

Study II: Two-way ANOVA was applied to test the main effects of genotype and interventions and one-way

ANOVA was used to test for differences between the interventions separately within each genotype. If

normality or variance of the data-set were skewed, the data were logarithmically transformed before applying

the ANOVA test. Student Newman Keuls post hoc test was used to locate differences when applicable. The

non-parametric Mann-Whitney U test was applied when the equal variance test or the normality test failed

even after logarithmically transformation.

Study III: Two-way repeated measures ANOVA was applied to test the main effect of resveratrol vs. placebo

and the combined resveratrol/exercise training vs. combined placebo/exercise training. If normality or

variance of the data-set were skewed, the data were logarithmically transformed before applying the

ANOVA test. If a main effect was observed, pair-wise differences were located by Student Newman Keuls

post hoc test. In addition, within-group comparisons were analyzed by students paired t-test.

All values are presented as means ± SE. A P<0.05 was considered significant and a tendency is reported for

0.05≤P<0.1.

39

Integrated discussion

In the following sections, the results obtained in study I-III will be discussed. Additional manuscripts have

been written/published on the basis of study II (Ringholm et al., 2013, in pending review, Exp Gerontol) and

study III (Gliemann et al., 2013), and results from these manuscript will be referred to in the discussion.

Furthermore, additional unpublished results and experiments have been performed (depicted in figure 6-12)

to support the results from the three manuscripts and these will also be discussed.

Acute inflammation

Role of skeletal muscle PGC-1α in acute inflammation

Study I demonstrated that PGC-1α MKO mice had a reduced LPS-induced plasma TNFα response and that

TG PGC-1α mice had a greater LPS-induced fold increase in plasma TNFα than WT mice. Moreover, the

LPS-induced skeletal muscle TNFα mRNA response was impaired in PGC-1α MKO mice, while no

differences were observed in TG PGC-1α mice relative to WT mice. These findings suggest that skeletal

muscle PGC-1α was required for a robust LPS-induced systemic and skeletal muscle TNFα mRNA response.

Moreover, an additional experiment revealed that 13 month old WT mice had an impaired LPS-induced

plasma IL-6 and CXCL-1 response compared with 10 weeks old mice, but with similar changes in plasma

TNFα as young mice (fig. 6), indicating that aging to some extent is also associated with an impaired LPS-

induced inflammatory response.

Figure 6. TNFα (a), IL-6 (b) and CXCL-1 (c) from young (Y) (10 weeks old) and 13 month old (O) wild type (WT) mice 2

hours after injection with either saline (Sal) as control or 0.8 µg lipopolysaccharide (LPS) per gram mouse. Values are

presented as mean ± SE (n = 6-10). Statistical analyses were performed using two-way ANOVA followed by a Student

Newman Keuls post hoc test. *: Significantly different from Sal, P < 0.05. #: Significantly different from Y-WT within

treatment, P < 0.05 (unpublished data).

As both aging and type 2 diabetes previously have been associated with reduced expression of PGC-1α in

skeletal muscle (Barres et al., 2009;Ling et al., 2004), the impaired LPS-induced response in PGC-1α MKO

40

mice (study I) and in 13 month old mice (fig. 6) may resemble the inability of elderly people and type 2

diabetes patients to effectively respond to LPS and infectious diseases in general (Andreasen et al.,

2010;Geerlings et al., 2000;Leibovici et al., 1991;Telzak et al., 1991;Goldstein, 2010;Miller, 1996).

To further investigate the importance of skeletal muscle PGC-1α for an acute LPS-induced TNFα response,

primary mouse muscle cells isolated from WT and whole body PGC-1α KO mice were stimulated with LPS

for 2 hours. The experiments were performed on 3 separate occasions and in line with the in vivo data, the

first two experiments showed reduced LPS-induced release of TNFα from the PGC-1α KO cells compared

with WT. However, the third similar experiment showed the opposite (fig. 7). Thus, unfortunately no

conclusion could be drawn from these experiments. The reason for these discrepancies is difficult to explain

and additional experiments are required to make final conclusions on this.

Figure 7. TNFα protein in cell culture medium from wild type (WT) and whole body PGC-1α knockout (KO) primary muscle

cells incubated for 2 hours with 1 µg LPS/ml medium. The results are from 3 independent experiments (n = 6 in each

experiment). TNFα protein was measured by ELISA. The results are shown as fold changes relative to control. Values are

presented as mean ± SE (unpublished data).

The observed reduced level of skeletal muscle TNFα converting enzyme (TACE) mRNA in PGC-1α MKO

mice (study I) may in part explain the impaired LPS-induced plasma TNFα response in these mice. As

TACE cleaves the membrane-associated precursor of TNFα into a soluble biological active form of TNFα

(Black et al., 1997;Moss et al., 1997), these findings may suggest an impaired ability of PGC-1α MKO mice

to secrete TNFα from skeletal muscle. However, it should be noted that the TACE mRNA level was not

increased in TG PGC-1α compared with WT mice (study I), suggesting that TACE is not a direct target of

PGC-1α mediated regulation. To assess the contribution of macrophages to the LPS-induced inflammatory

response, the mRNA content of the macrophage specific marker F4/80 was determined. Importantly, no

differences were observed in skeletal muscle F4/80 mRNA content before and after the LPS injection in any

of the three strains (KO, MKO and TG PGC-1α) (study I). This may indicate that the skeletal muscle fibers

were largely responsible for the observed LPS-induced expression of inflammatory cytokines in the skeletal

TNF released to medium

Experiment 1

LPS

TN

F

fo

ld c

ha

ng

e f

rom

Co

n

0

100

200

300

400

500

600

700 WT

KO


Experiment 3

LPS

TN

Ff

old

ch

an

ge

fro

m C

on

0

200

400

600

800

1000

1200

WT

KO


Experiment 2

LPS

TN

Ff

old

ch

an

ge

fro

m C

on

0

200

400

600

800

1000

WT

KO

LPS

41

muscle tissue and that the requirement of skeletal muscle PGC-1α to mediate a robust LPS-induced response

was not linked to PGC-1α-related differences in macrophage infiltration.

Interestingly previous studies examining the effect of PGC-1α on acute inflammation in vitro, showed that

high levels of PGC-1α reduced a pro-inflammatory response to cellular stressors (Alvarez-Guardia et al.,

2010;Eisele et al., 2013;Kim et al., 2007), which is somewhat opposite of the in vivo findings in study I.

Eisele et al. found that adenoviral overexpression of PGC-1α in C2C12 cells repressed a TNFα-, FFA- and

TLR agonist-mediated p65 phosphorylation and through that, the pro-inflammatory response (Eisele et al.,

2013). In line with these findings, overexpression of PGC-1α in human aortic smooth muscle cells has been

reported to reduce a TNFα-induced increase in NF-κB activity as well as in MCP-1 and VCAM expression

(Kim et al., 2007). In cardiomyocytes, it has been shown that NF-κB binds to PGC-1α and inhibits the

concomitant metabolic functions of PGC-1α upon stimulation with TNFα (Alvarez-Guardia et al.,

2010;Palomer et al., 2009). Taken together, it seems evident that PGC-1α and NF-κB interacts, and that the

outcome of this apparent dual regulation may depend on the specific experimental setting. Thus, it might

have added interesting information if additional components of the NF-κB signaling cascade including the

NF-κB DNA binding activity had been analyzed in the search for an explanation of the observed differences

in the LPS-induced skeletal muscle TNFα mRNA response in MKO mice.

Chronic low-grade inflammation

Age-associated low-grade inflammation

The observed increase in plasma TNFα and IL-6 as well as liver TNFα mRNA and protein, V-AT TNFα

mRNA and skeletal muscle TNFα protein in old mice (15 month old) compared with young mice (3 month

old) in study II, clearly emphasizes that aging is associated with systemic low-grade inflammation as well as

inflammation in various tissues in mice. This is in accordance with previous findings in both rodents and

humans (Wu et al., 2007;Wei et al., 1992). In line with this, 2-3 fold higher baseline levels of plasma TNFα

and IL-6 were observed in healthy aged subjects (65.3 ± 0.4 years) from study III than in healthy younger

subjects (26.2 ± 5.3 years) from a previous study performed in our group (Bienso et al., 2012;Ringholm et

al., 2011) (table 3).

42

Table 3. Comparison of baseline cytokine levels in young and aged subjects

Young (26.2 ± 5.3 years) Aged (65.3±0.4 years)

Plasma TNFα (pg/ml) 1.4±0.1 3.3±0.2

Plasma IL-6 (pg/ml) 0.9±0.2 2.5±0.2

Table 3. Comparison of baseline levels of plasma TNFα and IL-6 in aged subjects from study III and in young subjects from a

previous study performed by our group (Bienso et al., 2012;Ringholm et al., 2011). Values are presented as mean ± SE, with n

= 12 in the young group and n = 43 in the aged group. The plasma TNFα and IL-6 levels from the aged subjects are mean

baseline values from all subjects before the different interventions.

Importantly, precautions should be taken when interpreting results from different studies due to considerable

inter-array variations in such measurements. Notably however, the analyses were performed using similar

ultra-sensitive MSD-plates and analyzed on the same detection system (see method section), which should

reduce any inter-array variation considerably.

As the age-associated increase in low-grade inflammation observed in study II was paralleled by increased

adiposity, it is difficult to address the causality between aging, adiposity and inflammation. However, in

study II no positive correlations were observed between body fat % and plasma TNFα or plasma IL-6 or

between V-AT mass and plasma TNFα or plasma IL-6 (fig. 8). This suggests that the age-associated

inflammation in study II was not caused by adiposity per se.

Figure 8. Correlations between body fat % and plasma TNFα (a), body fat % and plasma IL-6 (b), visceral adipose tissue (V-

AT) mass and plasma TNFα (c) and between V-AT mass and plasma IL-6 (d) in 3 month old and 15 month old WT mice (raw

data from study II).

43

In addition, the F4/80 mRNA results in study II do not point towards that infiltration of macrophages into

either visceral adipose tissue, liver or skeletal muscle are responsible for the observed general age-associated

inflammation. Together, these data indicate that the age-associated increase in systemic and local

inflammation in adipose tissue, liver and skeletal muscle was independent of adiposity as well as

macrophage infiltration into these tissues. These findings suggest that these otherwise metabolic tissues

contribute to the systemic levels of circulating inflammatory mediators.

Role of PGC-1α in low-grade inflammation

The observations in study II that both young and old PGC-1α KO mice had increased plasma TNFα and IL-6

compared with age-matched WT mice demonstrate that lack of PGC-1α leads to low-grade inflammation. In

line with this, a reduced plasma TNFα level in TG PGC-1α mice as compared with WT mice was observed

in study I. These findings are in accordance with previous reports showing that PGC1-1α MKO mice have

increased basal expression of TNFα and IL-6 in skeletal muscle (Handschin et al., 2007b) and that TG PGC-

1α mice are protected from age-associated systemic as well as skeletal muscle inflammation (Wenz et al.,

2009). In line with this, overexpression of PGC-1α in tibialis anterior of rats by electroporation has been

shown to reduce NF-kB activity (Brault et al., 2010). On the other hand, the observed increase in p65

phosphorylation as well as TNFα mRNA and protein in skeletal muscle of TG PGC-1α mice observed in

study I, is in contrast to these reports. Nevertheless, a recent study with adenoviral overexpression of PGC-

1α in cultured human skeletal muscle cells supports that PGC-1α, regulates the transcription of pro-

inflammatory cytokines (Mormeneo et al., 2012). This study showed that, in addition to the common

metabolically related effects of PGC-1α on mitochondrial biogenesis and oxidative capacity, PGC-1α also

induced the expression of several inflammatory cytokines and chemokines including IL-8 (Mormeneo et al.,

2012). These latter findings combined with the observed increase in TNFα mRNA and protein in TG PGC-

1α mice in study I do not support a general anti-inflammatory role of PGC-1α. However, it should be noted

that although IL-8 previously has been associated with obesity and insulin resistance (Bruun et al., 2003), IL-

8 is also implicated in angiogenic processes (Heidemann et al., 2003). Moreover, it has also been shown that

IL-8 mRNA, like PGC-1α, is induced by contracting skeletal muscle (Akerstrom et al., 2005), which may

suggest that the ability of PGC-1α to induce IL-8 expression contributes to exercise-induced angiogenesis in

response to exercise. Interestingly, the study by Mormeneo et al. also found that overexpression of PGC-1α

increased the expression of NFKBIA (the gene that encodes IκB-α protein), which may indicate that the

PGC-1α-mediated suppression of NF-κB activity (Brault et al., 2010;Eisele et al., 2013;Kim et al., 2007) is

due to increased content of IκB-α (Mormeneo et al., 2012) and further supporting the dual regulation

between PGC-1α and NF-κB (Alvarez-Guardia et al., 2010;Eisele et al., 2013). Taken together, it seems that

PGC-1α differentially regulates the expression of several cytokines and chemokines depending on the

physiological context. Hence in some situations PGC-1α may act as an anti-inflammatory modulator by

reducing the expression of TNFα and IL-6 (Eisele et al., 2013;Handschin et al., 2007b), whereas in other

44

situations PGC-1α may induce the transcription of pro-inflammatory cytokines like IL-8 to promote

metabolic processes like angiogenesis.

Exercise training-induced adaptations in skeletal muscle of elderly men

Study III demonstrated that 8-weeks of exercise training improved the endurance capacity (assessed by a

one-leg knee-extensor exercise) and increased the enzyme activity of CS and HAD in skeletal muscle as well

as increased the protein content of cyt c and COXI in skeletal muscle, which together supports that exercise

training increases the oxidative capacity of skeletal muscle (Gollnick et al., 1973;Henriksson & Reitman,

1977;Holloszy, 1967), even in aged subjects. The concomitant increase in PGC-1α mRNA may potentially

explain these effects as PGC-1α previously in mice have been shown to be required for exercise training-

induced increases in oxidative proteins (Chinsomboon et al., 2009;Geng et al., 2010;Leick et al., 2010).

Physical activity and low-grade inflammation

Study II demonstrated that long-term exercise training prevented an age-associated increase in skeletal

muscle TNFα protein content and systemic plasma IL-6 in mice. In accordance with study II, exercise

training reduced the skeletal muscle TNFα mRNA content in aged men (study III). However no systemic

anti-inflammatory effects were observed with exercise training in these subjects. Together this suggests that

exercise training reduces skeletal muscle inflammation in mice and humans and partly systemic low-grade

inflammation in mice. While several previous studies have reported anti-inflammatory effects of exercise

training in humans (Kohut et al., 2006;Nicklas et al., 2008;Ostrowski et al., 1999;Vieira et al., 2009a), other

studies have failed demonstrating such effects (Hammett et al., 2004;Hammett et al., 2006). In a review by

Woods et al., it was suggested that the anti-inflammatory effects of exercise training are dependent on the

duration of the exercise training period (Woods et al., 2012). Most studies with interventions longer than 6

months have reported anti-inflammatory effects of exercise (Kohut et al., 2006;Nicklas et al., 2008;Vieira et

al., 2009a), whereas studies with shorter durations (<6 months) have failed to observe the same anti-

inflammatory effects of exercise training (Hammett et al., 2004;Hammett et al., 2006). In addition, the anti-

inflammatory effects of exercise are much more pronounced in rodents with metabolic dysfunction

(Kawanishi et al., 2010a;Kawanishi et al., 2010b;Kizaki et al., 2011;Vieira et al., 2009b;Vieira et al., 2009c)

than in metabolically healthy rodents (Vieira et al., 2009b). The suggestion that prolonged exercise training

(>6 months) (Woods et al., 2012) and some degree of metabolic dysfunction are required to observe

systemic anti-inflammatory effects of exercise training may explain the absent systemic anti-inflammatory

effects of exercise training in the subjects in study III considering that the subjects despite their advanced age

were metabolically healthy as evidenced by their fasting blood glucose (~5.3 mM) and insulin levels (~53

45

pM).

Although only few studies have investigated the effect of short-term inactivity on systemic and tissue

specific inflammation (Friedrichsen et al., 2012;Hojbjerre et al., 2011;Krogh-Madsen et al., 2010), the

unfavorable metabolic effects elicited by short-term physical inactivity are well described (Bienso et al.,

2012;Henriksson & Reitman, 1977;Krogh-Madsen et al., 2010;Mikines et al., 1991;Ringholm et al., 2011)

and more pronounced as compared to the beneficial metabolic effects elicited by short-term exercise training

(Henriksson & Reitman, 1977). In accordance with this, published findings from a study previously

performed by our group (Bienso et al., 2012;Ringholm et al., 2011) showed that seven days of complete

physical inactivity (bed-rest) led to whole body glucose intolerance as well as skeletal muscle insulin

resistance and metabolic dysregulation (Bienso et al., 2012;Ringholm et al., 2011) resembling a sedentary

lifestyle. To further examine the impact of the physical activity level on inflammation, I have determined

inflammatory markers in plasma samples as well as muscle biopsies and adipose tissue biopsies from this

bed-rest study. As shown in figure 9, this revealed that seven days of complete physical inactivity led to

minor increases in skeletal muscle IL-6 mRNA content as well as adipose tissue iNOS mRNA content (fig.

9). In accordance with these findings, a recent similar 9 day bed-rest study (Friedrichsen et al.,

2012;Hojbjerre et al., 2011) found that the macrophage marker cluster of differentiation (CD)68 mRNA was

increased in skeletal muscle after bed-rest relative to before (Friedrichsen et al., 2012). However, no

alterations in adipose tissue inflammation were observed with bed-rest in that study (Hojbjerre et al., 2011).

Albeit the observed inflammatory changes were rather small, these findings suggest that even short-term

physical inactivity induces some degree of local inflammation in both skeletal muscle and adipose tissue. In

the present bed-rest study, no alterations were observed in the systemic levels of TNFα and IL-6 (fig. 9),

which is in accordance with the study by Hojbjerre et al. showing no association between short-term physical

inactivity and systemic low-grade inflammation (Hojbjerre et al., 2011). Nevertheless, it is tempting to

speculate that more prolonged physical inactivity may lead to more pronounced inductions of both local and

systemic inflammation.

46

Figur 9. Plasma TNFα and IL-6 (a), TNFα protein in vastus laterailis skeletal muscle (SkM) and abdominal subcutaneous

adipose tissue (AT) (b), TNFα, IL-6 and iNOS mRNA content in skeletal muscle (c) and adipose tissue (d) in healthy young

men before and after 7 days of bed-rest. Values are mean ± SE, (n = 12). Statistical analyses were performed by Students

paired T-test *: Significantly different from before bed-rest, P < 0.05. (*): Tends to be significantly different from before bed-

rest, 0.05 ≤ P < 0.1 (unpublished data).

Taken together, the results obtained in mice (study II) showing that exercise training reduced skeletal muscle

TNFα protein and systemic IL-6 levels, the results in elderly men (study III) showing that exercise training

reduced skeletal muscle TNFα mRNA and conversely that physical inactivity (bed-rest) increases skeletal

muscle IL-6 mRNA and adipose issue iNOS mRNA supports that the level of physical activity to some

extent is associated with the level of inflammation.

The link between physical activity and inflammation may be explained by several underlying mechanism.

Although outside the scope of the present PhD thesis, humoral factors including adrenaline, cortisol and

growth hormone with immune-regulatory effects (Harbuz et al., 2003;Ignatowski et al., 1996;Nieman, 2003)

may have contributed to these effects.

Another potential mechanism being infiltrated macrophages into various tissues. To examine the degree of

infiltrated macrophages into skeletal muscle in mice and human subjects in study II and III, respectively, and

additionally into adipose tissue and liver tissue in mice from study II, expression of the macrophage specific

47

human marker EGF-like module-containing mucin-like hormone receptor-like 1 (EMR1) and the rodent

homolog, F4/80, were analyzed. However, neither alterations in EMR-1 protein in elderly men (study III)

(fig. 10) n alterations in F4/80 mRNA content in mice (study II) seemed to explain any of the exercise

training-induced differences in inflammation.

EMR-1 protein content in SkM

Placebo RSV T-placebo T-RSV

EM

R-1

pro

tein

/ G

AP

DH

pro

tein

(A

U)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

pre

post

Figure 10. EGF-like module-containing mucin-like hormone receptor-like 1 (EMR1) protein content in skeletal muscle (SkM)

from placebo (n = 7), resveratrol (RSV) (n = 9, 250 mg per day), exercise trained and placebo (n = 13) and exercise trained

and RSV (n = 14, 250 mg per day) supplemented subjects pre and post 8 weeks of intervention. Values are presented as mean

± SE (unpublished data from study III).

Alternatively, alterations in the anti-oxidant enzymes may indirectly have contributed to the anti-

inflammatory effects of exercise training. Hence, exercise training increased the skeletal muscle protein

content of SOD2, GPX1 and catalase in mice (study II) and the SOD2 protein content in men (study III, data

published in (Gliemann et al., 2013)). This exercise training-induced increase in anti-oxidant enzymes was

accompanied by reductions in protein carbonylation in skeletal muscle both in mice (study II) and human

subjects (Study III), supporting that exercise training reduces oxidative stress (Cunha et al., 2012;Qi et al.,

2011). As increased ROS previously has been shown to increase transcription of pro-inflammatory cytokines

through NF-κB and JNK (Schreck et al., 1992;Wellen & Hotamisligil, 2005), this apparent reduction in

oxidative stress may contribute to the exercise training-induced prevention of an elevation in skeletal muscle

TNFα protein content observed in mice (study II) and also the exercise training-induced reduction in skeletal

muscle TNFα mRNA in elderly men (study III). In accordance, the exercise training-induced reduction in

skeletal muscle TNFα mRNA in elderly men (study III) was accompanied by a minor decrease in JNK

signaling and increased content of IκB-α and IκB-β, potentially reflecting decreased translocation of NF-κB,

which together may explain the observed exercise training-induced reduction in skeletal muscle TNFα

mRNA content. However, in mice (study II), the investigated inflammatory signaling pathways could not

explain the exercise training-induced anti-inflammatory effects observed in skeletal muscle.

Taken together, the observed exercise training-induced reductions in TNFα mRNA (study III) and TNFα

48

protein (study II) may in part be ascribed to the parallel exercise training-induced increases in anti-oxidant

enzymes and reduction in oxidative stress. In human subjects (study III), this proposal was supported by

reduced inflammatory signaling events, whereas the underlying mechanisms remain to be determined in

mice (study II).

Role of PGC-1α in exercise training-induced anti-inflammatory effects

Results from study II demonstrated that PGC-1α was required for an exercise training-induced prevention of

an age-associated increase in TNFα protein in mouse skeletal muscle. Importantly, these findings are further

supported by the observation, that exercise training rescued an age-associated decrease in PGC-1α mRNA

content in WT mice in study II (fig. 11).

Figure 11. PGC-1α mRNA content in quadriceps muscle from 3-month old (mo) untrained chow mice (UT-C), 15 month old

untrained chow mice (UT-C), 15 month old untrained resveratrol (RSV) supplemented mice (UT-R), 15 month old exercise

trained (T-C) and 15 month old exercise trained and RSV supplemented (T-R) wild type mice. Values are presented as mean

± SE. (n = 8-10). Statistical analyses were performed by one-way ANOVA followed by Student Newman Keuls post hoc test.

†: Significantly different from 3 mo UT-C mice, P < 0.05. *: Significantly different from 15-month UT-C mice, P < 0.05

(unpublished data from study II).

Moreover, the exercise training-induced decrease in skeletal muscle TNFα mRNA in elderly men (study III)

was accompanied by an exercise training-induced increase in PGC-1α mRNA, supporting that an inverse

correlation between skeletal muscle PGC-1α mRNA and TNFα mRNA exists as previously suggested

(Handschin et al., 2007b). A recent study have shown that adenoviral overexpression of PGC-1α in primary

human muscle cells increases the expression of IκB-α (Mormeneo et al., 2012), which binds NF-κB and keep

it sequestered in an inactive state in the cytosol. Based on the concomitant exercise training-induced increase

in IκB-α and IκB-β protein in the human subjects, it may be speculated that the exercise training-induced

increase in PGC-1α mRNA may have inhibited NF-κB activity by inducing the expression of IκB-α and IκB-

β, which in turn may explain the exercise training-induced decrease in TNFα mRNA in the elderly subjects.

However, PGC-1α was not mandatory for the exercise training-induced reduction in systemic IL-6 levels

observed in mice (study II) indicating that other factors including humoral factors may have contributed to

the systemic exercise training-induced anti-inflammatory effects.

49

In mice (study II), no PGC-1α-dependent differences in the abundance of IκB-α or IκB-β or p65, p38 and

JNK phosphorylation were observed to possibly explain the observed differences in TNFα protein in skeletal

muscle. However, based on a previous study showing that overexpression of PGC-1α in skeletal muscle

reduced the NF-κB activity (Brault et al., 2010), it may be speculated that the dependency of PGC-1α for the

exercise training-induced reduction in age-associated skeletal muscle TNFα protein content may relate to

PGC-1α dependent changes in the activity of NF-κB, however this remains to be determined.

No previous studies have investigated the role of PGC-1α in the potential exercise training-induced anti-

inflammatory effects. However, a previous study demonstrated that PGC-1α MKO mice have markedly

elevated skeletal muscle TNFα mRNA and systemic plasma TNFα levels after an acute exercise bout

compared with resting mice, whereas no change was observed with acute exercise in WT mice (Handschin et

al., 2007a). Together this may suggest that PGC-1α is implicated in both the acute exercise-induced

regulation of skeletal muscle TNFα mRNA in mice (Handschin et al., 2007a), as well as the long-term

exercise training-induced reduction in skeletal muscle TNFα protein in mice (study II) and potentially also

the exercise training-induced reduction in TNFα mRNA in human subjects (study III). Furthermore, it may

be speculated that the acute and chronic exercise-induced induction of PGC-1α (Baar et al., 2002;Pilegaard

et al., 2000;Pilegaard et al., 2003;Russell et al., 2003;Short et al., 2003;Terada et al., 2002;Trappe et al.,

2013) negatively regulate the activity of NF-κB thereby reducing the expression of TNFα, as previously

shown under different physiological conditions (Brault et al., 2010;Eisele et al., 2013;Kim et al., 2007).

However, the exact PGC-1α-mediated exercise-induced anti-inflammatory mechanism in skeletal muscle

needs further elucidation.

Metabolic and anti-inflammatory effects of resveratrol?

In mice (study II), resveratrol increased the GPX1 protein content in skeletal muscle and prevented an age-

associated increase in protein carbonylation, whereas no metabolic effects of resveratrol were observed in the

elderly human subjects (study III). The lack of effects on the alleged putative targets of resveratrol cyt c and

COXI (Lagouge et al., 2006;Um et al., 2010) in humans (study III) and in mice from study II (data by

(Ringholm et al., 2013)) are noteworthy. Together, these findings are modest and in contrast to previous

reports on resveratrol with marked metabolic effects in rodents (Baur et al., 2006;Dolinsky et al.,

2012;Lagouge et al., 2006;Park et al., 2012;Um et al., 2010) and in humans (Brasnyo et al., 2011;Crandall et

al., 2012;Timmers et al., 2011). As the dose used in study II (~0.7 mg resveratrol per gram mouse per day) is

the same as the dose used in previous studies with marked metabolic improvements of resveratrol (Lagouge

et al., 2006;Um et al., 2010), the discrepancies are unlikely to be dose-related. Moreover, the human study

by Timmers et al. (2011) showing beneficial effects of resveratrol used an intermediate dose (150 mg/day for

days) compared with the dose in study III (250 mg/day for 8 weeks), the dose used by Yoshino et al. (2012)

50

(75 mg/day for 12 weeks) and the dose used by Poulsen et al. (2012) (500 mg/day for 4 weeks) showing no

metabolic effects in humans. The more or less absent effects of resveratrol may rather be related to the

metabolic state of the 15 month old untrained control mice (study II) and the aged subjects (study III).

Neither the mice in study II nor the human subjects in study III showed any metabolic complications as

assessed by the fasting glucose (~5.3 mM) and insulin levels (~53 pM) in the human subjects (study III) and

by a glucose tolerance test in the mice in study II (data by Ringholm et al., 2013). In accordance, the

majority of the previous studies showing marked metabolic effects of resveratrol have examined rodents with

metabolic dysfunction (Baur et al., 2006;Lagouge et al., 2006;Um et al., 2010), obese subjects, subjects with

impaired glucose tolerance or type 2 diabetes patients (Brasnyo et al., 2011;Crandall et al., 2012;Timmers et

al., 2011). Furthermore, a previous study in metabolically healthy rodents (Menzies et al., 2013) and a study

in healthy post-menopausal women (Yoshino et al., 2012) have in contrast failed to demonstrate resveratrol-

mediated metabolic improvements. Together this may emphasize that a certain degree of metabolic

dysfunction is a prerequisite to observe health beneficial effects of resveratrol.

Role of PGC-1α in resveratrol-mediated metabolic and anti-inflammatory effects

Several reports have suggested that resveratrol acts via an AMPK-SIRT1-PGC-1α axis, whereby a

transcriptional program central for the suggested beneficial effects of resveratrol is induced (Baur et al.,

2006;Lagouge et al., 2006;Park et al., 2012;Timmers et al., 2011;Um et al., 2010). However, neither AMPK

phosphorylation nor SIRT1 protein content was affected by resveratrol supplementation in mice from study

II (data by Ringholm et al., 2013) or in the human subjects in study III. Despite an apparent increase in

SIRT1 activity in the resveratrol supplemented elderly men (assessed by decreased total acetylation levels in

skeletal muscle) (study III), PGC-1α mRNA content was unaffected by the resveratrol treatment both in mice

from study II (fig. 11) and in the elderly men (study III).

Previous findings have reported that resveratrol increases the content of oxidative enzymes (Lagouge et al.,

2006;Um et al., 2010). To examine the role of PGC-1α herein, a pilot study using primary mouse skeletal

muscle cells isolated from WT and whole body PGC-1α KO mice was conducted. Fully differentiated

myotubes were incubated with 200 μM resveratrol or control medium for 24 hours. In accordance with

previous studies (Lagouge et al., 2006;Um et al., 2010), it appeared that cyt c mRNA was markedly up

regulated by resveratrol. Interestingly, resveratrol also appeared to up-regulate cyt c mRNA in PGC-1α KO

cells (fig. 12a), suggesting that the metabolic effects of resveratrol are independent of PGC-1α. In addition,

the basal TNFα mRNA content seemed higher in PGC-1α KO cells than WT cells, supporting other studies

suggesting that PGC-1α negatively regulates the expression of TNFα in skeletal muscle (Handschin et al.,

2007b;Handschin et al., 2007a;Wenz et al., 2009). Moreover, incubations with resveratrol seemed to reduce

the TNFα mRNA content in WT and in particularly PGC-1α KO cells (fig. 12b), again suggesting that the

resveratrol-mediated effects were independent of PGC-1α.

51

Figure 12. Cytochrome (Cyt) c mRNA (a) and TNFα mRNA (b) in primary skeletal muscle cells isolated from wild type (WT)

and whole body PGC-1α knockout (KO) mice incubated for 24 hours with either 200 µM resveratrol (RSV) or control

medium (Con). Cyt c and TNFα mRNA content are normalized β-actin (pre-developed primers and probes, Applied

Biosystems). Values are mean ± SE. from 2 independent experiments with 4-6 replicates in total.

Although these observations are preliminary and need to be confirmed, the results support the findings from

study II, showing that PGC-1α was not required for the resveratrol-mediated prevention of the age-associated

oxidative stress. Based on these findings and previous reports, this may suggests that PGC-1α is involved in

the metabolic actions of resveratrol as previously suggested (Baur et al., 2006;Lagouge et al., 2006;Park et

al., 2012;Timmers et al., 2011) but not required for these effects.

The present findings in mice (study II) and humans (study III) do not support that resveratrol exerts anti-

inflammatory effects. In contrast, resveratrol even increased the expression of V-AT IL-6 mRNA in mice

(study II). Moreover, in the elderly men (study III) resveratrol blunted an exercise training-induced decrease

in skeletal muscle TNFα mRNA content, which was paralleled by a resveratrol-mediated inhibition of an

exercise training-induced reduction in protein carbonylation in skeletal muscle. In addition to these findings,

resveratrol blunted several of the traditional exercise training-induced improvements in cardiovascular

parameters in the elderly men in study III (data published by (Gliemann et al., 2013)). These apparent

adverse effects of resveratrol when combined with exercise training in humans (study III) are in contrast to

previous studies in rodents showing additive metabolic and cardiovascular effects of combined exercise

training and resveratrol (Dolinsky et al., 2012;Menzies et al., 2013). The discrepancies between these studies

and the results observed in mice and humans from study II and III, respectively, are difficult to explain,

because the putative signaling pathways suggested to be activated by resveratrol were largely unaffected

both in mice from study II (data by Ringholm et al., 2013) and in the human subjects in study III. Previous

studies have revealed that the bioavailability of resveratrol is relatively low due to extensive glucuronidation

and sulfation of resveratrol in the liver, intestines and kidney into various metabolites (Yu et al., 2002).

However, a recent study (Poulsen et al., 2012) using the same supplier of resveratrol as in our human study

(study III) has shown that the bioavailability of this resveratrol was similar to the bioavailability of the

resveratrol used in the human study by Timmers et al. (2011), which showed resveratrol-mediated metabolic

52

effects This indicates that the discrepancies in the resveratrol effects were not related to differences in

bioavailability of different resveratrol products.

Taken together, whereas resveratrol supplementation in mice showed minor metabolic effects independent of

PGC-1α, resveratrol did not elicit health beneficial metabolic or anti-inflammatory effects in healthy aged

subjects and in contrast, resveratrol even impaired some of the observed exercise training-induced metabolic

improvements in humans.

53

Conclusion

The present thesis demonstrates that skeletal muscle PGC-1α expression is required for a robust LPS-induced

systemic TNFα and skeletal muscle TNFα response in mice.

In mice, aging was demonstrated to be associated with increased tissue inflammation and systemic low-grade

inflammation. Long-term exercise training was demonstrated to prevent an age-associated increase in

skeletal muscle TNFα protein content in a PGC-1α dependent manner, whereas the exercise training-induced

reduction in systemic IL-6 was independent of PGC-1α. In line with the mouse studies, exercise training was

shown to decrease the skeletal muscle TNFα mRNA content in aged human subjects. In addition, complete

physical inactivity in young subjects was conversely shown to have increased skeletal muscle IL-6 mRNA

indicating that skeletal muscle inflammation is inversely related to the level of physical activity. On the

contrary, no clear association between the physical activity level and the systemic inflammation was found in

neither of the human studies.

In mice, resveratrol was demonstrated to increase GPX1 protein content and to reduce protein carbonylation

levels independently of PGC-1α. However, resveratrol treatment did not elicit any anti-inflammatory effects

in these mice. In healthy aged subjects, resveratrol did not induce any metabolic or anti-inflammatory health

beneficial effects. In contrast, resveratrol actually impaired some of the observed exercise training-induced

improvements in human subjects.

Taken together the findings of the present PhD thesis demonstrate that resveratrol does not possess anti-

inflammatory effects in neither mice nor humans. Skeletal muscle inflammation is inversely related to level

of physical activity in mice and human subjects with PGC-1α being involved in the exercise training-induced

prevention of age-associated increases in skeletal muscle TNFα protein content. Furthermore, in acute LPS-

induced inflammation, PGC-1α is highly involved in regulation of skeletal muscle TNFα mRNA expression

and plasma TNFα, emphasizing its role in regulating inflammatory processes.

54

Closing remarks and future perspectives

Although outside the scope of the present thesis, an interesting notion from study II was that the age-

associated inflammation in mice was not accompanied by an impaired glucose tolerance (data in Ringholm et

al., 2013), which may indicate that low-grade inflammation is not directly linked with an impaired glucose

tolerance under the present conditions. In line with this, the previously reported bed rest-induced insulin

resistance (Bienso et al., 2012;Ringholm et al., 2011) was not accompanied with increased systemic

inflammation, which suggests that inactivity-driven insulin resistance precedes systemic low-grade

inflammation at least in the present bed-rest study with young healthy individuals. Together this may suggest

that systemic inflammation is not the primary cause of insulin resistance, but rather a secondary event in the

pathogenesis of insulin resistance and type 2 diabetes.

In study I-III and in the bed-rest study, it would be interesting to measure the NF-κB DNA binding activity

to have a more direct measure of this signaling pathway in various conditions. However, experiences from

our collaborators have showed that not all assays measuring the NF-κB activity are reliable and a lot of work

to optimize these assays is required.

In recent years it has become evident that several splice-variants of PGC-1α exists (Ruas et al.,

2012;Tadaishi et al., 2011;Zhang et al., 2009). The present thesis has not distinguished between the different

splice variants as all these variants to my knowledge contain exon 3-5 (the excised part in the PGC-1α KO

and PGC-1α MKO mice). In the future, it would be interesting to elucidate whether the various PGC-1α

splice-variants differentially regulate the inflammatory pathways in different physiological situations such as

acute inflammation vs. acute exercise vs. exercise training.

In study I-III and in the bed-rest study, determination of PGC-1α protein abundance and/or activity would

have been valuable measures to increase the interpretation of results. However, in our hands, this is not an

easy task as we several times have tried to do this with different antibodies (Calbiochem, Cell Signaling,

Santa Cruz and with a custom made antibody) without success. Each time we observe multiple unspecific

bands and none of these are ablated in whole body PGC-1α KO tissue. The reason for this is difficult to

explain, but may relate to the short half-life (~ 20-30 min) of PGC-1α (Puigserver et al., 2001;Trausch-Azar

et al., 2010) due to rapid degradation (Adamovich et al., 2013;Puigserver et al., 2001). In the future, stable

methods measure PGC-1α protein content and activity are thus warranted.

55

Reference list

Adamovich Y, Shlomai A, Tsvetkov P, Umansky KB, Reuven N, Estall JL, Spiegelman BM, & Shaul Y (2013). The

protein level of PGC-1alpha, a key metabolic regulator, is controlled by NADH-NQO1. Mol Cell Biol 33, 2603-2613.

Aguirre V, Uchida T, Yenush L, Davis R, & White MF (2000). The c-Jun NH(2)-terminal kinase promotes insulin

resistance during association with insulin receptor substrate-1 and phosphorylation of Ser(307). J Biol Chem 275, 9047-

9054.

Akerstrom T, Steensberg A, Keller P, Keller C, Penkowa M, & Pedersen BK (2005). Exercise induces interleukin-8

expression in human skeletal muscle. J Physiol 563, 507-516.

Akimoto T, Pohnert SC, Li P, Zhang M, Gumbs C, Rosenberg PB, Williams RS, & Yan Z (2005). Exercise stimulates

Pgc-1alpha transcription in skeletal muscle through activation of the p38 MAPK pathway. J Biol Chem 280, 19587-

19593.

Alibegovic AC, Sonne MP, Hojbjerre L, Bork-Jensen J, Jacobsen S, Nilsson E, Faerch K, Hiscock N, Mortensen B,

Friedrichsen M, Stallknecht B, Dela F, & Vaag A (2010). Insulin resistance induced by physical inactivity is associated

with multiple transcriptional changes in skeletal muscle in young men. Am J Physiol Endocrinol Metab 299, E752-

E763.

Alvarez-Guardia D, Palomer X, Coll T, Davidson MM, Chan TO, Feldman AM, Laguna JC, & Vazquez-Carrera M

(2010). The p65 subunit of NF-kappaB binds to PGC-1alpha, linking inflammation and metabolic disturbances in

cardiac cells. Cardiovasc Res 87, 449-458.

Amati F, Dube JJ, Coen PM, Stefanovic-Racic M, Toledo FG, & Goodpaster BH (2009). Physical inactivity and obesity

underlie the insulin resistance of aging. Diabetes Care 32, 1547-1549.

Andersson U & Scarpulla RC (2001). Pgc-1-related coactivator, a novel, serum-inducible coactivator of nuclear

respiratory factor 1-dependent transcription in mammalian cells. Mol Cell Biol 21, 3738-3749.

Andreasen AS, Krabbe KS, Krogh-Madsen R, Taudorf S, Pedersen BK, & Moller K (2008). Human endotoxemia as a

model of systemic inflammation. Curr Med Chem 15, 1697-1705.

Andreasen AS, Pedersen-Skovsgaard T, Berg RM, Svendsen KD, Feldt-Rasmussen B, Pedersen BK, & Moller K

(2010). Type 2 diabetes mellitus is associated with impaired cytokine response and adhesion molecule expression in

human endotoxemia. Intensive Care Med 36, 1548-1555.

Baar K, Wende AR, Jones TE, Marison M, Nolte LA, Chen M, Kelly DP, & Holloszy JO (2002). Adaptations of

skeletal muscle to exercise: rapid increase in the transcriptional coactivator PGC-1. FASEB J 16, 1879-1886.

Ballou SP, Lozanski FB, Hodder S, Rzewnicki DL, Mion LC, Sipe JD, Ford AB, & Kushner I (1996). Quantitative and

qualitative alterations of acute-phase proteins in healthy elderly persons. Age Ageing 25, 224-230.

56

Barres R, Osler ME, Yan J, Rune A, Fritz T, Caidahl K, Krook A, & Zierath JR (2009). Non-CpG methylation of the

PGC-1alpha promoter through DNMT3B controls mitochondrial density. Cell Metab 10, 189-198.

Barres R, Yan J, Egan B, Treebak JT, Rasmussen M, Fritz T, Caidahl K, Krook A, O'Gorman DJ, & Zierath JR (2012).

Acute exercise remodels promoter methylation in human skeletal muscle. Cell Metab 15, 405-411.

Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C, Kalra A, Prabhu VV, Allard JS, Lopez-Lluch G, Lewis K,

Pistell PJ, Poosala S, Becker KG, Boss O, Gwinn D, Wang M, Ramaswamy S, Fishbein KW, Spencer RG, Lakatta EG,

Le CD, Shaw RJ, Navas P, Puigserver P, Ingram DK, de CR, & Sinclair DA (2006). Resveratrol improves health and

survival of mice on a high-calorie diet. Nature 444, 337-342.

Bergstrom J (1975). Percutaneous needle biopsy of skeletal muscle in physiological and clinical research. Scand J Clin

Lab Invest 35, 609-616.

Bienso RS, Ringholm S, Kiilerich K, Aachmann-Andersen NJ, Krogh-Madsen R, Guerra B, Plomgaard P, van HG,

Treebak JT, Saltin B, Lundby C, Calbet JA, Pilegaard H, & Wojtaszewski JF (2012). GLUT4 and glycogen synthase

are key players in bed rest-induced insulin resistance. Diabetes 61, 1090-1099.

Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, Wolfson MF, Castner BJ, Stocking KL, Reddy P, Srinivasan

S, Nelson N, Boiani N, Schooley KA, Gerhart M, Davis R, Fitzner JN, Johnson RS, Paxton RJ, March CJ, & Cerretti

DP (1997). A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature 385, 729-733.

Booth FW, Roberts CK, & Laye MJ (2012). Lack of exercise is a major cause of chronic diseases. Compr Physiol 2,

1143-1211.

Borra MT, Smith BC, & Denu JM (2005). Mechanism of human SIRT1 activation by resveratrol. J Biol Chem 280,

17187-17195.

Brandt C & Pedersen BK (2010). The role of exercise-induced myokines in muscle homeostasis and the defense against

chronic diseases. J Biomed Biotechnol 2010, 520258.

Brasnyo P, Molnar GA, Mohas M, Marko L, Laczy B, Cseh J, Mikolas E, Szijarto IA, Merei A, Halmai R, Meszaros

LG, Sumegi B, & Wittmann I (2011). Resveratrol improves insulin sensitivity, reduces oxidative stress and activates the

Akt pathway in type 2 diabetic patients. Br J Nutr 106, 383-389.

Brault JJ, Jespersen JG, & Goldberg AL (2010). Peroxisome proliferator-activated receptor gamma coactivator 1alpha

or 1beta overexpression inhibits muscle protein degradation, induction of ubiquitin ligases, and disuse atrophy. J Biol

Chem 285, 19460-19471.

Brooks SV & Faulkner JA (1994). Skeletal muscle weakness in old age: underlying mechanisms. Med Sci Sports Exerc

26, 432-439.

Bruun JM, Verdich C, Toubro S, Astrup A, & Richelsen B (2003). Association between measures of insulin sensitivity

and circulating levels of interleukin-8, interleukin-6 and tumor necrosis factor-alpha. Effect of weight loss in obese

men. Eur J Endocrinol 148, 535-542.

57

Bruunsgaard H & Pedersen BK (2003). Age-related inflammatory cytokines and disease. Immunol Allergy Clin North

Am 23, 15-39.

Bruunsgaard H, Pedersen M, & Pedersen BK (2001). Aging and proinflammatory cytokines. Curr Opin Hematol 8,

131-136.

Cai D, Frantz JD, Tawa NE, Jr., Melendez PA, Oh BC, Lidov HG, Hasselgren PO, Frontera WR, Lee J, Glass DJ, &

Shoelson SE (2004). IKKbeta/NF-kappaB activation causes severe muscle wasting in mice. Cell 119, 285-298.

Calvo JA, Daniels TG, Wang X, Paul A, Lin J, Spiegelman BM, Stevenson SC, & Rangwala SM (2008). Muscle-

specific expression of PPARgamma coactivator-1alpha improves exercise performance and increases peak oxygen

uptake. J Appl Physiol 104, 1304-1312.

Canto C, Jiang LQ, Deshmukh AS, Mataki C, Coste A, Lagouge M, Zierath JR, & Auwerx J (2010). Interdependence of

AMPK and SIRT1 for metabolic adaptation to fasting and exercise in skeletal muscle. Cell Metab 11, 213-219.

Cao W, Daniel KW, Robidoux J, Puigserver P, Medvedev AV, Bai X, Floering LM, Spiegelman BM, & Collins S

(2004). p38 mitogen-activated protein kinase is the central regulator of cyclic AMP-dependent transcription of the

brown fat uncoupling protein 1 gene. Mol Cell Biol 24, 3057-3067.

Carpenter S & O'Neill LA (2009). Recent insights into the structure of Toll-like receptors and post-translational

modifications of their associated signalling proteins. Biochem J 422, 1-10.

Chabi B, Ljubicic V, Menzies KJ, Huang JH, Saleem A, & Hood DA (2008). Mitochondrial function and apoptotic

susceptibility in aging skeletal muscle. Aging Cell 7, 2-12.

Chinsomboon J, Ruas J, Gupta RK, Thom R, Shoag J, Rowe GC, Sawada N, Raghuram S, & Arany Z (2009). The

transcriptional coactivator PGC-1alpha mediates exercise-induced angiogenesis in skeletal muscle. Proc Natl Acad Sci

U S A 106, 21401-21406.

Choi CS, Befroy DE, Codella R, Kim S, Reznick RM, Hwang YJ, Liu ZX, Lee HY, Distefano A, Samuel VT, Zhang D,

Cline GW, Handschin C, Lin J, Petersen KF, Spiegelman BM, & Shulman GI (2008). Paradoxical effects of increased

expression of PGC-1alpha on muscle mitochondrial function and insulin-stimulated muscle glucose metabolism. Proc

Natl Acad Sci U S A 105, 19926-19931.

Chomczynski P & Sacchi N (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-

chloroform extraction. Anal Biochem 162, 156-159.

Clayton DA (1991). Replication and transcription of vertebrate mitochondrial DNA. Annu Rev Cell Biol 7, 453-478.

Cohen HJ, Pieper CF, Harris T, Rao KM, & Currie MS (1997). The association of plasma IL-6 levels with functional

disability in community-dwelling elderly. J Gerontol A Biol Sci Med Sci 52, M201-M208.

58

Collart MA, Baeuerle P, & Vassalli P (1990). Regulation of tumor necrosis factor alpha transcription in macrophages:

involvement of four kappa B-like motifs and of constitutive and inducible forms of NF-kappa B. Mol Cell Biol 10,

1498-1506.

Conley KE, Jubrias SA, & Esselman PC (2000). Oxidative capacity and ageing in human muscle. J Physiol 526 Pt 1,

203-210.

Corsonello A, Garasto S, Abbatecola AM, Rose G, Passarino G, Mazzei B, Pranno L, Guffanti EE, Bustacchini S, &

Lattanzio F (2010). Targeting inflammation to slow or delay functional decline: where are we? Biogerontology 11, 603-

614.

Crandall JP, Oram V, Trandafirescu G, Reid M, Kishore P, Hawkins M, Cohen HW, & Barzilai N (2012). Pilot study of

resveratrol in older adults with impaired glucose tolerance. J Gerontol A Biol Sci Med Sci 67, 1307-1312.

Cunha TF, Bacurau AV, Moreira JB, Paixao NA, Campos JC, Ferreira JC, Leal ML, Negrao CE, Moriscot AS, Wisloff

U, & Brum PC (2012). Exercise training prevents oxidative stress and ubiquitin-proteasome system overactivity and

reverse skeletal muscle atrophy in heart failure. PLoS One 7, e41701.

Davies KJ, Quintanilha AT, Brooks GA, & Packer L (1982). Free radicals and tissue damage produced by exercise.

Biochem Biophys Res Commun 107, 1198-1205.

Davis RJ (2000). Signal transduction by the JNK group of MAP kinases. Cell 103, 239-252.

Dela F, Mikines KJ, Von LM, Secher NH, & Galbo H (1992). Effect of training on insulin-mediated glucose uptake in

human muscle. Am J Physiol 263, E1134-E1143.

Dobbs RJ, Charlett A, Purkiss AG, Dobbs SM, Weller C, & Peterson DW (1999). Association of circulating TNF-alpha

and IL-6 with ageing and parkinsonism. Acta Neurol Scand 100, 34-41.

Doherty TJ, Vandervoort AA, Taylor AW, & Brown WF (1993). Effects of motor unit losses on strength in older men

and women. J Appl Physiol 74, 868-874.

Dolinsky VW, Jones KE, Sidhu RS, Haykowsky M, Czubryt MP, Gordon T, & Dyck JR (2012). Improvements in

skeletal muscle strength and cardiac function induced by resveratrol during exercise training contribute to enhanced

exercise performance in rats. J Physiol 590, 2783-2799.

Eisele PS, Salatino S, Sobek J, Hottiger MO, & Handschin C (2013). The Peroxisome Proliferator-activated Receptor

gamma Coactivator 1alpha/beta (PGC-1) Coactivators Repress the Transcriptional Activity of NF-kappaB in Skeletal

Muscle Cells. J Biol Chem 288, 2246-2260.

Ershler WB, Sun WH, Binkley N, Gravenstein S, Volk MJ, Kamoske G, Klopp RG, Roecker EB, Daynes RA, &

Weindruch R (1993). Interleukin-6 and aging: blood levels and mononuclear cell production increase with advancing

age and in vitro production is modifiable by dietary restriction. Lymphokine Cytokine Res 12, 225-230.

59

Finkel T & Holbrook NJ (2000). Oxidants, oxidative stress and the biology of ageing. Nature 408, 239-247.

Fleischman A, Shoelson SE, Bernier R, & Goldfine AB (2008). Salsalate improves glycemia and inflammatory

parameters in obese young adults. Diabetes Care 31, 289-294.

Friedrichsen M, Ribel-Madsen R, Mortensen B, Hansen CN, Alibegovic AC, Hojbjerre L, Sonne MP, Wojtaszewski JF,

Stallknecht B, Dela F, & Vaag A (2012). Muscle inflammatory signaling in response to 9 days of physical inactivity in

young men with low compared with normal birth weight. Eur J Endocrinol 167, 829-838.

Frontera WR, Meredith CN, O'Reilly KP, Knuttgen HG, & Evans WJ (1988). Strength conditioning in older men:

skeletal muscle hypertrophy and improved function. J Appl Physiol 64, 1038-1044.

Frost RA, Nystrom GJ, & Lang CH (2002). Lipopolysaccharide regulates proinflammatory cytokine expression in

mouse myoblasts and skeletal muscle. Am J Physiol Regul Integr Comp Physiol 283, R698-R709.

Gao Z, Zhang X, Zuberi A, Hwang D, Quon MJ, Lefevre M, & Ye J (2004). Inhibition of insulin sensitivity by free

fatty acids requires activation of multiple serine kinases in 3T3-L1 adipocytes. Mol Endocrinol 18, 2024-2034.

Geerlings SE, Brouwer EC, Van Kessel KC, Gaastra W, Stolk RP, & Hoepelman AI (2000). Cytokine secretion is

impaired in women with diabetes mellitus. Eur J Clin Invest 30, 995-1001.

Geng T, Li P, Okutsu M, Yin X, Kwek J, Zhang M, & Yan Z (2010). PGC-1alpha plays a functional role in exercise-

induced mitochondrial biogenesis and angiogenesis but not fiber-type transformation in mouse skeletal muscle. Am J

Physiol Cell Physiol 298, C572-C579.

Gill R, Tsung A, & Billiar T (2010). Linking oxidative stress to inflammation: Toll-like receptors. Free Radic Biol Med

48, 1121-1132.

Gleeson M, Bishop NC, Stensel DJ, Lindley MR, Mastana SS, & Nimmo MA (2011). The anti-inflammatory effects of

exercise: mechanisms and implications for the prevention and treatment of disease. Nat Rev Immunol 11, 607-615.

Gleeson M, McFarlin B, & Flynn M (2006). Exercise and Toll-like receptors. Exerc Immunol Rev 12, 34-53.

Gliemann L, Schmidt JF, Olesen J, Bienso RS, Peronard SL, Grandjean SU, Mortensen SP, Nyberg M, Bangsbo J,

Pilegaard H, & Hellsten Y (2013). Resveratrol Blunts the Positive Effects of Exercise Training on Cardiovascular

Health in Aged Men. J Physiol.

Goldstein DR (2010). Aging, imbalanced inflammation and viral infection. Virulence 1, 295-298.

Gollnick PD, Armstrong RB, Saltin B, Saubert CW, Sembrowich WL, & Shepherd RE (1973). Effect of training on

enzyme activity and fiber composition of human skeletal muscle. J Appl Physiol 34, 107-111.

60

Hambleton J, Weinstein SL, Lem L, & DeFranco AL (1996). Activation of c-Jun N-terminal kinase in bacterial

lipopolysaccharide-stimulated macrophages. Proc Natl Acad Sci U S A 93, 2774-2778.

Hammett CJ, Oxenham HC, Baldi JC, Doughty RN, Ameratunga R, French JK, White HD, & Stewart RA (2004).

Effect of six months' exercise training on C-reactive protein levels in healthy elderly subjects. J Am Coll Cardiol 44,

2411-2413.

Hammett CJ, Prapavessis H, Baldi JC, Varo N, Schoenbeck U, Ameratunga R, French JK, White HD, & Stewart RA

(2006). Effects of exercise training on 5 inflammatory markers associated with cardiovascular risk. Am Heart J 151,

367.

Handschin C, Chin S, Li P, Liu F, Maratos-Flier E, LeBrasseur NK, Yan Z, & Spiegelman BM (2007a). Skeletal muscle

fiber-type switching, exercise intolerance, and myopathy in PGC-1alpha muscle-specific knock-out animals. J Biol

Chem 282, 30014-30021.

Handschin C, Choi CS, Chin S, Kim S, Kawamori D, Kurpad AJ, Neubauer N, Hu J, Mootha VK, Kim YB, Kulkarni

RN, Shulman GI, & Spiegelman BM (2007b). Abnormal glucose homeostasis in skeletal muscle-specific PGC-1alpha

knockout mice reveals skeletal muscle-pancreatic beta cell crosstalk. J Clin Invest 117, 3463-3474.

Handschin C, Rhee J, Lin J, Tarr PT, & Spiegelman BM (2003). An autoregulatory loop controls peroxisome

proliferator-activated receptor gamma coactivator 1alpha expression in muscle. Proc Natl Acad Sci U S A 100, 7111-

7116.

Handschin C & Spiegelman BM (2006). Peroxisome proliferator-activated receptor gamma coactivator 1 coactivators,

energy homeostasis, and metabolism. Endocr Rev 27, 728-735.

Handschin C & Spiegelman BM (2008). The role of exercise and PGC1alpha in inflammation and chronic disease.

Nature 454, 463-469.

Harbuz MS, Chover-Gonzalez AJ, & Jessop DS (2003). Hypothalamo-pituitary-adrenal axis and chronic immune

activation. Ann N Y Acad Sci 992, 99-106.

Harikumar KB & Aggarwal BB (2008). Resveratrol: a multitargeted agent for age-associated chronic diseases. Cell

Cycle 7, 1020-1035.

HARMAN D (1956). Aging: a theory based on free radical and radiation chemistry. J Gerontol 11, 298-300.

HARMAN D (1972). The biologic clock: the mitochondria? J Am Geriatr Soc 20, 145-147.

Heidemann J, Ogawa H, Dwinell MB, Rafiee P, Maaser C, Gockel HR, Otterson MF, Ota DM, Lugering N, Domschke

W, & Binion DG (2003). Angiogenic effects of interleukin 8 (CXCL8) in human intestinal microvascular endothelial

cells are mediated by CXCR2. J Biol Chem 278, 8508-8515.

61

Henriksson J & Reitman JS (1977). Time course of changes in human skeletal muscle succinate dehydrogenase and

cytochrome oxidase activities and maximal oxygen uptake with physical activity and inactivity. Acta Physiol Scand 99,

91-97.

Hoffmann A & Baltimore D (2006). Circuitry of nuclear factor kappaB signaling. Immunol Rev 210, 171-186.

Hojbjerre L, Sonne MP, Alibegovic AC, Nielsen NB, Dela F, Vaag A, Bruun JM, & Stallknecht B (2011). Impact of

physical inactivity on adipose tissue low-grade inflammation in first-degree relatives of type 2 diabetic patients.

Diabetes Care 34, 2265-2272.

Hollmann W, Struder HK, Tagarakis CV, & King G (2007). Physical activity and the elderly. Eur J Cardiovasc Prev

Rehabil 14, 730-739.

Holloszy JO (1967). Biochemical adaptations in muscle. Effects of exercise on mitochondrial oxygen uptake and

respiratory enzyme activity in skeletal muscle. J Biol Chem 242, 2278-2282.

Hotamisligil GS, Arner P, Caro JF, Atkinson RL, & Spiegelman BM (1995a). Increased adipose tissue expression of

tumor necrosis factor-alpha in human obesity and insulin resistance. J Clin Invest 95, 2409-2415.

Hotamisligil GS, Arner P, Caro JF, Atkinson RL, & Spiegelman BM (1995b). Increased adipose tissue expression of

tumor necrosis factor-alpha in human obesity and insulin resistance. J Clin Invest 95, 2409-2415.

Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, & Spiegelman BM (1996). IRS-1-mediated inhibition of

insulin receptor tyrosine kinase activity in TNF-alpha- and obesity-induced insulin resistance. Science 271, 665-668.

Hotamisligil GS, Shargill NS, & Spiegelman BM (1993). Adipose expression of tumor necrosis factor-alpha: direct role

in obesity-linked insulin resistance. Science 259, 87-91.

Howitz KT, Bitterman KJ, Cohen HY, Lamming DW, Lavu S, Wood JG, Zipkin RE, Chung P, Kisielewski A, Zhang

LL, Scherer B, & Sinclair DA (2003). Small molecule activators of sirtuins extend Saccharomyces cerevisiae lifespan.

Nature 425, 191-196.

Hundal RS, Petersen KF, Mayerson AB, Randhawa PS, Inzucchi S, Shoelson SE, & Shulman GI (2002). Mechanism by

which high-dose aspirin improves glucose metabolism in type 2 diabetes. J Clin Invest 109, 1321-1326.

Hung JH, Su IJ, Lei HY, Wang HC, Lin WC, Chang WT, Huang W, Chang WC, Chang YS, Chen CC, & Lai MD

(2004). Endoplasmic reticulum stress stimulates the expression of cyclooxygenase-2 through activation of NF-kappaB

and pp38 mitogen-activated protein kinase. J Biol Chem 279, 46384-46392.

Huss JM, Kopp RP, & Kelly DP (2002). Peroxisome proliferator-activated receptor coactivator-1alpha (PGC-1alpha)

coactivates the cardiac-enriched nuclear receptors estrogen-related receptor-alpha and -gamma. Identification of novel

leucine-rich interaction motif within PGC-1alpha. J Biol Chem 277, 40265-40274.

62

Ignatowski TA, Gallant S, & Spengler RN (1996). Temporal regulation by adrenergic receptor stimulation of

macrophage (M phi)-derived tumor necrosis factor (TNF) production post-LPS challenge. J Neuroimmunol 65, 107-

117.

Irrcher I, Adhihetty PJ, Sheehan T, Joseph AM, & Hood DA (2003). PPARgamma coactivator-1alpha expression

during thyroid hormone- and contractile activity-induced mitochondrial adaptations. Am J Physiol Cell Physiol 284,

C1669-C1677.

Irrcher I, Ljubicic V, & Hood DA (2009). Interactions between ROS and AMP kinase activity in the regulation of PGC-

1alpha transcription in skeletal muscle cells. Am J Physiol Cell Physiol 296, C116-C123.

Jackson JR, Ryan MJ, & Alway SE (2011). Long-term supplementation with resveratrol alleviates oxidative stress but

does not attenuate sarcopenia in aged mice. J Gerontol A Biol Sci Med Sci 66, 751-764.

Jacobs MD & Harrison SC (1998). Structure of an IkappaBalpha/NF-kappaB complex. Cell 95, 749-758.

Jager S, Handschin C, St-Pierre J, & Spiegelman BM (2007). AMP-activated protein kinase (AMPK) action in skeletal

muscle via direct phosphorylation of PGC-1alpha. Proc Natl Acad Sci U S A 104, 12017-12022.

Jorgensen SB, Wojtaszewski JF, Viollet B, Andreelli F, Birk JB, Hellsten Y, Schjerling P, Vaulont S, Neufer PD,

Richter EA, & Pilegaard H (2005). Effects of alpha-AMPK knockout on exercise-induced gene activation in mouse

skeletal muscle. FASEB J 19, 1146-1148.

Kaeberlein M, McDonagh T, Heltweg B, Hixon J, Westman EA, Caldwell SD, Napper A, Curtis R, DiStefano PS,

Fields S, Bedalov A, & Kennedy BK (2005). Substrate-specific activation of sirtuins by resveratrol. J Biol Chem 280,

17038-17045.

Karin M, Liu Z, & Zandi E (1997). AP-1 function and regulation. Curr Opin Cell Biol 9, 240-246.

Kawanishi N, Yano H, Yokogawa Y, & Suzuki K (2010a). Exercise training inhibits inflammation in adipose tissue via

both suppression of macrophage infiltration and acceleration of phenotypic switching from M1 to M2 macrophages in

high-fat-diet-induced obese mice. Exerc Immunol Rev 16, 105-118.

Kawanishi N, Yano H, Yokogawa Y, & Suzuki K (2010b). Exercise training inhibits inflammation in adipose tissue via

both suppression of macrophage infiltration and acceleration of phenotypic switching from M1 to M2 macrophages in

high-fat-diet-induced obese mice. Exerc Immunol Rev 16, 105-118.

Keller C, Steensberg A, Pilegaard H, Osada T, Saltin B, Pedersen BK, & Neufer PD (2001). Transcriptional activation

of the IL-6 gene in human contracting skeletal muscle: influence of muscle glycogen content. FASEB J 15, 2748-2750.

Kern PA, Saghizadeh M, Ong JM, Bosch RJ, Deem R, & Simsolo RB (1995). The expression of tumor necrosis factor

in human adipose tissue. Regulation by obesity, weight loss, and relationship to lipoprotein lipase. J Clin Invest 95,

2111-2119.

63

Kim HJ, Park KG, Yoo EK, Kim YH, Kim YN, Kim HS, Kim HT, Park JY, Lee KU, Jang WG, Kim JG, Kim BW, &

Lee IK (2007). Effects of PGC-1alpha on TNF-alpha-induced MCP-1 and VCAM-1 expression and NF-kappaB

activation in human aortic smooth muscle and endothelial cells. Antioxid Redox Signal 9, 301-307.

Kim S, Jin Y, Choi Y, & Park T (2011). Resveratrol exerts anti-obesity effects via mechanisms involving down-

regulation of adipogenic and inflammatory processes in mice. Biochem Pharmacol 81, 1343-1351.

Kizaki T, Maegawa T, Sakurai T, Ogasawara JE, Ookawara T, Oh-ishi S, Izawa T, Haga S, & Ohno H (2011).

Voluntary exercise attenuates obesity-associated inflammation through ghrelin expressed in macrophages. Biochem

Biophys Res Commun 413, 454-459.

Kohut ML, McCann DA, Russell DW, Konopka DN, Cunnick JE, Franke WD, Castillo MC, Reighard AE, & Vanderah

E (2006). Aerobic exercise, but not flexibility/resistance exercise, reduces serum IL-18, CRP, and IL-6 independent of

beta-blockers, BMI, and psychosocial factors in older adults. Brain Behav Immun 20, 201-209.

Koivisto VA, Yki-Jarvinen H, & DeFronzo RA (1986). Physical training and insulin sensitivity. Diabetes Metab Rev 1,

445-481.

Krogh-Madsen R, Thyfault JP, Broholm C, Mortensen OH, Olsen RH, Mounier R, Plomgaard P, van HG, Booth FW, &

Pedersen BK (2010). A 2-wk reduction of ambulatory activity attenuates peripheral insulin sensitivity. J Appl Physiol

108, 1034-1040.

Kudo H, Takahara T, Yata Y, Kawai K, Zhang W, & Sugiyama T (2009). Lipopolysaccharide triggered TNF-alpha-

induced hepatocyte apoptosis in a murine non-alcoholic steatohepatitis model. J Hepatol 51, 168-175.

Kusuhara K, Madsen K, Jensen L, Hellsten Y, & Pilegaard H (2007). Calcium signalling in the regulation of PGC-

1alpha, PDK4 and HKII mRNA expression. Biol Chem 388, 481-488.

Lagouge M, Argmann C, Gerhart-Hines Z, Meziane H, Lerin C, Daussin F, Messadeq N, Milne J, Lambert P, Elliott P,

Geny B, Laakso M, Puigserver P, & Auwerx J (2006). Resveratrol improves mitochondrial function and protects

against metabolic disease by activating SIRT1 and PGC-1alpha. Cell 127, 1109-1122.

Leibovici L, Samra Z, Konisberger H, Kalter-Leibovici O, Pitlik SD, & Drucker M (1991). Bacteremia in adult diabetic

patients. Diabetes Care 14, 89-94.

Leick L, Lyngby SS, Wojtaszewski JF, & Pilegaard H (2010). PGC-1alpha is required for training-induced prevention

of age-associated decline in mitochondrial enzymes in mouse skeletal muscle. Exp Gerontol 45, 336-342.

Leick L, Wojtaszewski JF, Johansen ST, Kiilerich K, Comes G, Hellsten Y, Hidalgo J, & Pilegaard H (2008). PGC-

1alpha is not mandatory for exercise- and training-induced adaptive gene responses in mouse skeletal muscle. Am J

Physiol Endocrinol Metab 294, E463-E474.

Lerin C, Rodgers JT, Kalume DE, Kim SH, Pandey A, & Puigserver P (2006). GCN5 acetyltransferase complex

controls glucose metabolism through transcriptional repression of PGC-1alpha. Cell Metab 3, 429-438.

64

Lin J, Puigserver P, Donovan J, Tarr P, & Spiegelman BM (2002a). Peroxisome proliferator-activated receptor gamma

coactivator 1beta (PGC-1beta ), a novel PGC-1-related transcription coactivator associated with host cell factor. J Biol

Chem 277, 1645-1648.

Lin J, Wu H, Tarr PT, Zhang CY, Wu Z, Boss O, Michael LF, Puigserver P, Isotani E, Olson EN, Lowell BB, Bassel-

Duby R, & Spiegelman BM (2002b). Transcriptional co-activator PGC-1 alpha drives the formation of slow-twitch

muscle fibres. Nature 418, 797-801.

Lin J, Wu PH, Tarr PT, Lindenberg KS, St-Pierre J, Zhang CY, Mootha VK, Jager S, Vianna CR, Reznick RM, Cui L,

Manieri M, Donovan MX, Wu Z, Cooper MP, Fan MC, Rohas LM, Zavacki AM, Cinti S, Shulman GI, Lowell BB,

Krainc D, & Spiegelman BM (2004). Defects in adaptive energy metabolism with CNS-linked hyperactivity in PGC-

1alpha null mice. Cell 119, 121-135.

Lin Y, Berg AH, Iyengar P, Lam TK, Giacca A, Combs TP, Rajala MW, Du X, Rollman B, Li W, Hawkins M, Barzilai

N, Rhodes CJ, Fantus IG, Brownlee M, & Scherer PE (2005). The hyperglycemia-induced inflammatory response in

adipocytes: the role of reactive oxygen species. J Biol Chem 280, 4617-4626.

Ling C, Poulsen P, Carlsson E, Ridderstrale M, Almgren P, Wojtaszewski J, Beck-Nielsen H, Groop L, & Vaag A

(2004). Multiple environmental and genetic factors influence skeletal muscle PGC-1alpha and PGC-1beta gene

expression in twins. J Clin Invest 114, 1518-1526.

Lowry CV, Kimmey JS, Felder S, Chi MM, Kaiser KK, Passonneau PN, Kirk KA, & Lowry OH (1978). Enzyme

patterns in single human muscle fibers. J Biol Chem 253, 8269-8277.

Lundby C, Nordsborg N, Kusuhara K, Kristensen KM, Neufer PD, & Pilegaard H (2005). Gene expression in human

skeletal muscle: alternative normalization method and effect of repeated biopsies. Eur J Appl Physiol 95, 351-360.

Masoro EJ (2001). Physiology of aging. Int J Sport Nutr Exerc Metab 11 Suppl, S218-S222.

May MJ & Ghosh S (1998). Signal transduction through NF-kappa B. Immunol Today 19, 80-88.

Meador BM, Krzyszton CP, Johnson RW, & Huey KA (2008). Effects of IL-10 and age on IL-6, IL-1beta, and TNF-

alpha responses in mouse skeletal and cardiac muscle to an acute inflammatory insult. J Appl Physiol 104, 991-997.

Menzies KJ, Singh K, Saleem A, & Hood DA (2013). Sirtuin 1-mediated effects of exercise and resveratrol on

mitochondrial biogenesis. J Biol Chem.

Michael LF, Wu Z, Cheatham RB, Puigserver P, Adelmant G, Lehman JJ, Kelly DP, & Spiegelman BM (2001).

Restoration of insulin-sensitive glucose transporter (GLUT4) gene expression in muscle cells by the transcriptional

coactivator PGC-1. Proc Natl Acad Sci U S A 98, 3820-3825.

Mikines KJ, Richter EA, Dela F, & Galbo H (1991). Seven days of bed rest decrease insulin action on glucose uptake in

leg and whole body. J Appl Physiol 70, 1245-1254.

65

Mikines KJ, Sonne B, Farrell PA, Tronier B, & Galbo H (1988). Effect of physical exercise on sensitivity and

responsiveness to insulin in humans. Am J Physiol 254, E248-E259.

Miller RA (1996). The aging immune system: primer and prospectus. Science 273, 70-74.

Miura S, Kawanaka K, Kai Y, Tamura M, Goto M, Shiuchi T, Minokoshi Y, & Ezaki O (2007). An increase in murine

skeletal muscle peroxisome proliferator-activated receptor-gamma coactivator-1alpha (PGC-1alpha) mRNA in response

to exercise is mediated by beta-adrenergic receptor activation. Endocrinology 148, 3441-3448.

Monsalve M, Wu Z, Adelmant G, Puigserver P, Fan M, & Spiegelman BM (2000). Direct coupling of transcription and

mRNA processing through the thermogenic coactivator PGC-1. Mol Cell 6, 307-316.

Mormeneo E, Jimenez-Mallebrera C, Palomer X, De N, V, Vazquez-Carrera M, Orozco A, Nascimento A, Colomer J,

Lerin C, & Gomez-Foix AM (2012). PGC-1alpha induces mitochondrial and myokine transcriptional programs and

lipid droplet and glycogen accumulation in cultured human skeletal muscle cells. PLoS One 7, e29985.

Moss ML, Jin SL, Milla ME, Bickett DM, Burkhart W, Carter HL, Chen WJ, Clay WC, Didsbury JR, Hassler D,

Hoffman CR, Kost TA, Lambert MH, Leesnitzer MA, McCauley P, McGeehan G, Mitchell J, Moyer M, Pahel G,

Rocque W, Overton LK, Schoenen F, Seaton T, Su JL, Becherer JD, & . (1997). Cloning of a disintegrin

metalloproteinase that processes precursor tumour-necrosis factor-alpha. Nature 385, 733-736.

Muller S, Martin S, Koenig W, Hanifi-Moghaddam P, Rathmann W, Haastert B, Giani G, Illig T, Thorand B, & Kolb H

(2002). Impaired glucose tolerance is associated with increased serum concentrations of interleukin 6 and co-regulated

acute-phase proteins but not TNF-alpha or its receptors. Diabetologia 45, 805-812.

Nguyen MT, Favelyukis S, Nguyen AK, Reichart D, Scott PA, Jenn A, Liu-Bryan R, Glass CK, Neels JG, & Olefsky

JM (2007). A subpopulation of macrophages infiltrates hypertrophic adipose tissue and is activated by free fatty acids

via Toll-like receptors 2 and 4 and JNK-dependent pathways. J Biol Chem 282, 35279-35292.

Nicklas BJ, Ambrosius W, Messier SP, Miller GD, Penninx BW, Loeser RF, Palla S, Bleecker E, & Pahor M (2004).

Diet-induced weight loss, exercise, and chronic inflammation in older, obese adults: a randomized controlled clinical

trial. Am J Clin Nutr 79, 544-551.

Nicklas BJ, Hsu FC, Brinkley TJ, Church T, Goodpaster BH, Kritchevsky SB, & Pahor M (2008). Exercise training and

plasma C-reactive protein and interleukin-6 in elderly people. J Am Geriatr Soc 56, 2045-2052.

Nieman DC (2003). Current perspective on exercise immunology. Curr Sports Med Rep 2, 239-242.

Ojuka EO, Jones TE, Han DH, Chen M, & Holloszy JO (2003). Raising Ca2+ in L6 myotubes mimics effects of

exercise on mitochondrial biogenesis in muscle. FASEB J 17, 675-681.

Olas B & Wachowicz B (2005). Resveratrol, a phenolic antioxidant with effects on blood platelet functions. Platelets

16, 251-260.

66

Olesen J, Kiilerich K, & Pilegaard H (2010). PGC-1alpha-mediated adaptations in skeletal muscle. Pflugers Arch 460,

153-162.

Olson EN & Williams RS (2000). Remodeling muscles with calcineurin. Bioessays 22, 510-519.

Ostrowski K, Rohde T, Asp S, Schjerling P, & Pedersen BK (1999). Pro- and anti-inflammatory cytokine balance in

strenuous exercise in humans. J Physiol 515 ( Pt 1), 287-291.

Ozcan U, Cao Q, Yilmaz E, Lee AH, Iwakoshi NN, Ozdelen E, Tuncman G, Gorgun C, Glimcher LH, & Hotamisligil

GS (2004). Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes. Science 306, 457-461.

Palomer X, Alvarez-Guardia D, Rodriguez-Calvo R, Coll T, Laguna JC, Davidson MM, Chan TO, Feldman AM, &

Vazquez-Carrera M (2009). TNF-alpha reduces PGC-1alpha expression through NF-kappaB and p38 MAPK leading to

increased glucose oxidation in a human cardiac cell model. Cardiovasc Res 81, 703-712.

Paolisso G, Rizzo MR, Mazziotti G, Tagliamonte MR, Gambardella A, Rotondi M, Carella C, Giugliano D, Varricchio

M, & D'Onofrio F (1998). Advancing age and insulin resistance: role of plasma tumor necrosis factor-alpha. Am J

Physiol 275, E294-E299.

Park SJ, Ahmad F, Philp A, Baar K, Williams T, Luo H, Ke H, Rehmann H, Taussig R, Brown AL, Kim MK, Beaven

MA, Burgin AB, Manganiello V, & Chung JH (2012). Resveratrol ameliorates aging-related metabolic phenotypes by

inhibiting cAMP phosphodiesterases. Cell 148, 421-433.

Pearson KJ, Baur JA, Lewis KN, Peshkin L, Price NL, Labinskyy N, Swindell WR, Kamara D, Minor RK, Perez E,

Jamieson HA, Zhang Y, Dunn SR, Sharma K, Pleshko N, Woollett LA, Csiszar A, Ikeno Y, Le CD, Elliott PJ, Becker

KG, Navas P, Ingram DK, Wolf NS, Ungvari Z, Sinclair DA, & de CR (2008). Resveratrol delays age-related

deterioration and mimics transcriptional aspects of dietary restriction without extending life span. Cell Metab 8, 157-

168.

Pedersen BK (2006). The anti-inflammatory effect of exercise: its role in diabetes and cardiovascular disease control.

Essays Biochem 42, 105-117.

Pedersen BK (2009). The diseasome of physical inactivity--and the role of myokines in muscle--fat cross talk. J Physiol

587, 5559-5568.

Pedersen BK & Saltin B (2006). Evidence for prescribing exercise as therapy in chronic disease. Scand J Med Sci

Sports 16 Suppl 1, 3-63.

Pedersen M, Bruunsgaard H, Weis N, Hendel HW, Andreassen BU, Eldrup E, Dela F, & Pedersen BK (2003).

Circulating levels of TNF-alpha and IL-6-relation to truncal fat mass and muscle mass in healthy elderly individuals and

in patients with type-2 diabetes. Mech Ageing Dev 124, 495-502.

Perseghin G, Price TB, Petersen KF, Roden M, Cline GW, Gerow K, Rothman DL, & Shulman GI (1996). Increased

glucose transport-phosphorylation and muscle glycogen synthesis after exercise training in insulin-resistant subjects. N

Engl J Med 335, 1357-1362.

67

Pickup JC, Mattock MB, Chusney GD, & Burt D (1997). NIDDM as a disease of the innate immune system: association

of acute-phase reactants and interleukin-6 with metabolic syndrome X. Diabetologia 40, 1286-1292.

Pilegaard H, Ordway GA, Saltin B, & Neufer PD (2000). Transcriptional regulation of gene expression in human

skeletal muscle during recovery from exercise. Am J Physiol Endocrinol Metab 279, E806-E814.

Pilegaard H, Saltin B, & Neufer PD (2003). Exercise induces transient transcriptional activation of the PGC-1alpha

gene in human skeletal muscle. J Physiol 546, 851-858.

Plomgaard P, Bouzakri K, Krogh-Madsen R, Mittendorfer B, Zierath JR, & Pedersen BK (2005). Tumor necrosis

factor-alpha induces skeletal muscle insulin resistance in healthy human subjects via inhibition of Akt substrate 160

phosphorylation. Diabetes 54, 2939-2945.

Plomgaard P, Nielsen AR, Fischer CP, Mortensen OH, Broholm C, Penkowa M, Krogh-Madsen R, Erikstrup C,

Lindegaard B, Petersen AM, Taudorf S, & Pedersen BK (2007). Associations between insulin resistance and TNF-alpha

in plasma, skeletal muscle and adipose tissue in humans with and without type 2 diabetes. Diabetologia 50, 2562-2571.

Polkinghorne E, Lau Q, Cooney GJ, Kraegen EW, & Cleasby ME (2008). Local activation of the IkappaK-NF-kappaB

pathway in muscle does not cause insulin resistance. Am J Physiol Endocrinol Metab 294, E316-E325.

Poulsen MM, Vestergaard PF, Clasen BF, Radko Y, Christensen LP, Stodkilde-Jorgensen H, Moller N, Jessen N,

Pedersen SB, & Jorgensen JO (2012). High-Dose Resveratrol Supplementation in Obese Men: An Investigator-

Initiated, Randomized, Placebo-Controlled Clinical Trial of Substrate Metabolism, Insulin Sensitivity, and Body

Composition. Diabetes.

Price NL, Gomes AP, Ling AJ, Duarte FV, Martin-Montalvo A, North BJ, Agarwal B, Ye L, Ramadori G, Teodoro JS,

Hubbard BP, Varela AT, Davis JG, Varamini B, Hafner A, Moaddel R, Rolo AP, Coppari R, Palmeira CM, de CR,

Baur JA, & Sinclair DA (2012). SIRT1 is required for AMPK activation and the beneficial effects of resveratrol on

mitochondrial function. Cell Metab 15, 675-690.

Puigserver P, Adelmant G, Wu Z, Fan M, Xu J, O'Malley B, & Spiegelman BM (1999). Activation of PPARgamma

coactivator-1 through transcription factor docking. Science 286, 1368-1371.

Puigserver P, Rhee J, Donovan J, Walkey CJ, Yoon JC, Oriente F, Kitamura Y, Altomonte J, Dong H, Accili D, &

Spiegelman BM (2003). Insulin-regulated hepatic gluconeogenesis through FOXO1-PGC-1alpha interaction. Nature

423, 550-555.

Puigserver P, Rhee J, Lin J, Wu Z, Yoon JC, Zhang CY, Krauss S, Mootha VK, Lowell BB, & Spiegelman BM (2001).

Cytokine stimulation of energy expenditure through p38 MAP kinase activation of PPARgamma coactivator-1. Mol

Cell 8, 971-982.

Puigserver P, Wu Z, Park CW, Graves R, Wright M, & Spiegelman BM (1998). A cold-inducible coactivator of nuclear

receptors linked to adaptive thermogenesis. Cell 92, 829-839.

68

Qi Z, He J, Zhang Y, Shao Y, & Ding S (2011). Exercise training attenuates oxidative stress and decreases p53 protein

content in skeletal muscle of type 2 diabetic Goto-Kakizaki rats. Free Radic Biol Med 50, 794-800.

Raingeaud J, Gupta S, Rogers JS, Dickens M, Han J, Ulevitch RJ, & Davis RJ (1995). Pro-inflammatory cytokines and

environmental stress cause p38 mitogen-activated protein kinase activation by dual phosphorylation on tyrosine and

threonine. J Biol Chem 270, 7420-7426.

Reyna SM, Ghosh S, Tantiwong P, Meka CS, Eagan P, Jenkinson CP, Cersosimo E, DeFronzo RA, Coletta DK,

Sriwijitkamol A, & Musi N (2008). Elevated toll-like receptor 4 expression and signaling in muscle from insulin-

resistant subjects. Diabetes 57, 2595-2602.

Ringholm, Olesen J, Pedersen J, Brandt C, Halling J, Hellsten Y, Prats C, & Pilegaard H. Effect of lifelong resveratrol

supplementation and exercise training on skeletal muscle oxidative capacity in aging mice; impact of PGC-1alpha.

Exp.Gerontol. 2013.

Ref Type: Journal (Full)

Ringholm S, Bienso RS, Kiilerich K, Guadalupe-Grau A, Aachmann-Andersen NJ, Saltin B, Plomgaard P, Lundby C,

Wojtaszewski JF, Calbet JA, & Pilegaard H (2011). Bed rest reduces metabolic protein content and abolishes exercise-

induced mRNA responses in human skeletal muscle. Am J Physiol Endocrinol Metab 301, E649-E658.

Rodgers JT, Lerin C, Gerhart-Hines Z, & Puigserver P (2008). Metabolic adaptations through the PGC-1 alpha and

SIRT1 pathways. FEBS Lett 582, 46-53.

Rodgers JT, Lerin C, Haas W, Gygi SP, Spiegelman BM, & Puigserver P (2005). Nutrient control of glucose

homeostasis through a complex of PGC-1alpha and SIRT1. Nature 434, 113-118.

Rohl M, Pasparakis M, Baudler S, Baumgartl J, Gautam D, Huth M, De LR, Krone W, Rajewsky K, & Bruning JC

(2004). Conditional disruption of IkappaB kinase 2 fails to prevent obesity-induced insulin resistance. J Clin Invest 113,

474-481.

Ruas JL, White JP, Rao RR, Kleiner S, Brannan KT, Harrison BC, Greene NP, Wu J, Estall JL, Irving BA, Lanza IR,

Rasbach KA, Okutsu M, Nair KS, Yan Z, Leinwand LA, & Spiegelman BM (2012). A PGC-1alpha isoform induced by

resistance training regulates skeletal muscle hypertrophy. Cell 151, 1319-1331.

Russell AP, Feilchenfeldt J, Schreiber S, Praz M, Crettenand A, Gobelet C, Meier CA, Bell DR, Kralli A, Giacobino JP,

& Deriaz O (2003). Endurance training in humans leads to fiber type-specific increases in levels of peroxisome

proliferator-activated receptor-gamma coactivator-1 and peroxisome proliferator-activated receptor-alpha in skeletal

muscle. Diabetes 52, 2874-2881.

Saghizadeh M, Ong JM, Garvey WT, Henry RR, & Kern PA (1996). The expression of TNF alpha by human muscle.

Relationship to insulin resistance. J Clin Invest 97, 1111-1116.

Sarkar D & Fisher PB (2006). Molecular mechanisms of aging-associated inflammation. Cancer Lett 236, 13-23.

69

Schaap LA, Koster A, & Visser M (2012). Adiposity, Muscle Mass, and Muscle Strength in Relation to Functional

Decline in Older Persons. Epidemiol Rev.

Schreck R, Albermann K, & Baeuerle PA (1992). Nuclear factor kappa B: an oxidative stress-responsive transcription

factor of eukaryotic cells (a review). Free Radic Res Commun 17, 221-237.

Shi H, Kokoeva MV, Inouye K, Tzameli I, Yin H, & Flier JS (2006). TLR4 links innate immunity and fatty acid-

induced insulin resistance. J Clin Invest 116, 3015-3025.

Shimizu H, Mitomo K, Watanabe T, Okamoto S, & Yamamoto K (1990). Involvement of a NF-kappa B-like

transcription factor in the activation of the interleukin-6 gene by inflammatory lymphokines. Mol Cell Biol 10, 561-568.

Shoelson SE, Lee J, & Goldfine AB (2006). Inflammation and insulin resistance. J Clin Invest 116, 1793-1801.

Short KR, Bigelow ML, Kahl J, Singh R, Coenen-Schimke J, Raghavakaimal S, & Nair KS (2005). Decline in skeletal

muscle mitochondrial function with aging in humans. Proc Natl Acad Sci U S A 102, 5618-5623.

Short KR, Vittone JL, Bigelow ML, Proctor DN, Rizza RA, Coenen-Schimke JM, & Nair KS (2003). Impact of aerobic

exercise training on age-related changes in insulin sensitivity and muscle oxidative capacity. Diabetes 52, 1888-1896.

Silveira LR, Pilegaard H, Kusuhara K, Curi R, & Hellsten Y (2006). The contraction induced increase in gene

expression of peroxisome proliferator-activated receptor (PPAR)-gamma coactivator 1alpha (PGC-1alpha),

mitochondrial uncoupling protein 3 (UCP3) and hexokinase II (HKII) in primary rat skeletal muscle cells is dependent

on reactive oxygen species. Biochim Biophys Acta 1763, 969-976.

Skrobuk P, von KS, Semenova MM, Zitting A, & Koistinen HA (2012). Acute exposure to resveratrol inhibits AMPK

activity in human skeletal muscle cells. Diabetologia 55, 3051-3060.

Sriwijitkamol A, Christ-Roberts C, Berria R, Eagan P, Pratipanawatr T, DeFronzo RA, Mandarino LJ, & Musi N

(2006). Reduced skeletal muscle inhibitor of kappaB beta content is associated with insulin resistance in subjects with

type 2 diabetes: reversal by exercise training. Diabetes 55, 760-767.

St-Pierre J, Drori S, Uldry M, Silvaggi JM, Rhee J, Jager S, Handschin C, Zheng K, Lin J, Yang W, Simon DK, Bachoo

R, & Spiegelman BM (2006). Suppression of reactive oxygen species and neurodegeneration by the PGC-1

transcriptional coactivators. Cell 127, 397-408.

Starkie R, Ostrowski SR, Jauffred S, Febbraio M, & Pedersen BK (2003). Exercise and IL-6 infusion inhibit endotoxin-

induced TNF-alpha production in humans. FASEB J 17, 884-886.

Steensberg A, Febbraio MA, Osada T, Schjerling P, van HG, Saltin B, & Pedersen BK (2001). Interleukin-6 production

in contracting human skeletal muscle is influenced by pre-exercise muscle glycogen content. J Physiol 537, 633-639.

Steensberg A, Fischer CP, Keller C, Moller K, & Pedersen BK (2003a). IL-6 enhances plasma IL-1ra, IL-10, and

cortisol in humans. Am J Physiol Endocrinol Metab 285, E433-E437.

70

Steensberg A, Fischer CP, Sacchetti M, Keller C, Osada T, Schjerling P, van HG, Febbraio MA, & Pedersen BK

(2003b). Acute interleukin-6 administration does not impair muscle glucose uptake or whole-body glucose disposal in

healthy humans. J Physiol 548, 631-638.

Stojanovic S, Sprinz H, & Brede O (2001). Efficiency and mechanism of the antioxidant action of trans-resveratrol and

its analogues in the radical liposome oxidation. Arch Biochem Biophys 391, 79-89.

Summermatter S, Shui G, Maag D, Santos G, Wenk MR, & Handschin C (2013). PGC-1alpha improves glucose

homeostasis in skeletal muscle in an activity-dependent manner. Diabetes 62, 85-95.

Tadaishi M, Miura S, Kai Y, Kano Y, Oishi Y, & Ezaki O (2011). Skeletal muscle-specific expression of PGC-1alpha-

b, an exercise-responsive isoform, increases exercise capacity and peak oxygen uptake. PLoS One 6, e28290.

Tantiwong P, Shanmugasundaram K, Monroy A, Ghosh S, Li M, DeFronzo RA, Cersosimo E, Sriwijitkamol A, Mohan

S, & Musi N (2010). NF-kappaB activity in muscle from obese and type 2 diabetic subjects under basal and exercise-

stimulated conditions. Am J Physiol Endocrinol Metab 299, E794-E801.

Telzak EE, Greenberg MS, Budnick LD, Singh T, & Blum S (1991). Diabetes mellitus--a newly described risk factor

for infection from Salmonella enteritidis. J Infect Dis 164, 538-541.

Terada S, Goto M, Kato M, Kawanaka K, Shimokawa T, & Tabata I (2002). Effects of low-intensity prolonged exercise

on PGC-1 mRNA expression in rat epitrochlearis muscle. Biochem Biophys Res Commun 296, 350-354.

Timmers S, Konings E, Bilet L, Houtkooper RH, van de Weijer T, Goossens GH, Hoeks J, van der Krieken S, Ryu D,

Kersten S, Moonen-Kornips E, Hesselink MK, Kunz I, Schrauwen-Hinderling VB, Blaak EE, Auwerx J, & Schrauwen

P (2011). Calorie restriction-like effects of 30 days of resveratrol supplementation on energy metabolism and metabolic

profile in obese humans. Cell Metab 14, 612-622.

Trappe S, Hayes E, Galpin A, Kaminsky L, Jemiolo B, Fink W, Trappe T, Jansson A, Gustafsson T, & Tesch P (2013).

New records in aerobic power among octogenarian lifelong endurance athletes. J Appl Physiol 114, 3-10.

Trausch-Azar J, Leone TC, Kelly DP, & Schwartz AL (2010). Ubiquitin proteasome-dependent degradation of the

transcriptional coactivator PGC-1{alpha} via the N-terminal pathway. J Biol Chem 285, 40192-40200.

Um JH, Park SJ, Kang H, Yang S, Foretz M, McBurney MW, Kim MK, Viollet B, & Chung JH (2010). AMP-activated

protein kinase-deficient mice are resistant to the metabolic effects of resveratrol. Diabetes 59, 554-563.

Valenzano DR, Terzibasi E, Genade T, Cattaneo A, Domenici L, & Cellerino A (2006). Resveratrol prolongs lifespan

and retards the onset of age-related markers in a short-lived vertebrate. Curr Biol 16, 296-300.

Venables MC & Jeukendrup AE (2009). Physical inactivity and obesity: links with insulin resistance and type 2

diabetes mellitus. Diabetes Metab Res Rev 25 Suppl 1, S18-S23.

71

Ventura JJ, Kennedy NJ, Lamb JA, Flavell RA, & Davis RJ (2003). c-Jun NH(2)-terminal kinase is essential for the

regulation of AP-1 by tumor necrosis factor. Mol Cell Biol 23, 2871-2882.

Verdaet D, Dendale P, De BD, Delanghe J, Block P, & De BG (2004). Association between leisure time physical

activity and markers of chronic inflammation related to coronary heart disease. Atherosclerosis 176, 303-310.

Vieira VJ, Hu L, Valentine RJ, McAuley E, Evans EM, Baynard T, & Woods JA (2009a). Reduction in trunk fat

predicts cardiovascular exercise training-related reductions in C-reactive protein. Brain Behav Immun 23, 485-491.

Vieira VJ, Valentine RJ, Wilund KR, Antao N, Baynard T, & Woods JA (2009b). Effects of exercise and low-fat diet

on adipose tissue inflammation and metabolic complications in obese mice. Am J Physiol Endocrinol Metab 296,

E1164-E1171.

Vieira VJ, Valentine RJ, Wilund KR, & Woods JA (2009c). Effects of diet and exercise on metabolic disturbances in

high-fat diet-fed mice. Cytokine 46, 339-345.

Virbasius JV & Scarpulla RC (1994). Activation of the human mitochondrial transcription factor A gene by nuclear

respiratory factors: a potential regulatory link between nuclear and mitochondrial gene expression in organelle

biogenesis. Proc Natl Acad Sci U S A 91, 1309-1313.

Wei J, Xu H, Davies JL, & Hemmings GP (1992). Increase of plasma IL-6 concentration with age in healthy subjects.

Life Sci 51, 1953-1956.

Wei YH, Lu CY, Lee HC, Pang CY, & Ma YS (1998). Oxidative damage and mutation to mitochondrial DNA and age-

dependent decline of mitochondrial respiratory function. Ann N Y Acad Sci 854, 155-170.

Wellen KE & Hotamisligil GS (2005). Inflammation, stress, and diabetes. J Clin Invest 115, 1111-1119.

Wende AR, Schaeffer PJ, Parker GJ, Zechner C, Han DH, Chen MM, Hancock CR, Lehman JJ, Huss JM, McClain DA,

Holloszy JO, & Kelly DP (2007). A role for the transcriptional coactivator PGC-1alpha in muscle refueling. J Biol

Chem 282, 36642-36651.

Wenz T, Rossi SG, Rotundo RL, Spiegelman BM, & Moraes CT (2009). Increased muscle PGC-1alpha expression

protects from sarcopenia and metabolic disease during aging. Proc Natl Acad Sci U S A 106, 20405-20410.

Wild S, Roglic G, Green A, Sicree R, & King H (2004). Global prevalence of diabetes: estimates for the year 2000 and

projections for 2030. Diabetes Care 27, 1047-1053.

Winder WW & Hardie DG (1996). Inactivation of acetyl-CoA carboxylase and activation of AMP-activated protein

kinase in muscle during exercise. Am J Physiol 270, E299-E304.

Wojtaszewski JF, Nielsen P, Hansen BF, Richter EA, & Kiens B (2000). Isoform-specific and exercise intensity-

dependent activation of 5'-AMP-activated protein kinase in human skeletal muscle. J Physiol 528 Pt 1, 221-226.

72

Wolsk E, Mygind H, Grondahl TS, Pedersen BK, & van HG (2010). IL-6 selectively stimulates fat metabolism in

human skeletal muscle. Am J Physiol Endocrinol Metab 299, E832-E840.

Wood JG, Rogina B, Lavu S, Howitz K, Helfand SL, Tatar M, & Sinclair D (2004). Sirtuin activators mimic caloric

restriction and delay ageing in metazoans. Nature 430, 686-689.

Woods JA, Wilund KR, Martin SA, & Kistler BM (2012). Exercise, inflammation and aging. Aging Dis 3, 130-140.

Wright DC, Geiger PC, Han DH, Jones TE, & Holloszy JO (2007). Calcium induces increases in peroxisome

proliferator-activated receptor gamma coactivator-1alpha and mitochondrial biogenesis by a pathway leading to p38

mitogen-activated protein kinase activation. J Biol Chem 282, 18793-18799.

Wu D, Ren Z, Pae M, Guo W, Cui X, Merrill AH, & Meydani SN (2007). Aging up-regulates expression of

inflammatory mediators in mouse adipose tissue. J Immunol 179, 4829-4839.

Wu H, Kanatous SB, Thurmond FA, Gallardo T, Isotani E, Bassel-Duby R, & Williams RS (2002). Regulation of

mitochondrial biogenesis in skeletal muscle by CaMK. Science 296, 349-352.

Wu SH, Shu XO, Chow WH, Xiang YB, Zhang X, Cai Q, Li HL, Milne G, Wen W, Ji BT, Rothman N, Gao YT, Zheng

W, & Yang G (2013). Adiposity and fat distribution in relation to inflammation and oxidative stress in a relatively lean

population of Chinese women. Dis Markers 34, 279-293.

Wu Z, Puigserver P, Andersson U, Zhang C, Adelmant G, Mootha V, Troy A, Cinti S, Lowell B, Scarpulla RC, &

Spiegelman BM (1999). Mechanisms controlling mitochondrial biogenesis and respiration through the thermogenic

coactivator PGC-1. Cell 98, 115-124.

Yoshino J, Conte C, Fontana L, Mittendorfer B, Imai S, Schechtman KB, Gu C, Kunz I, Rossi FF, Patterson BW, &

Klein S (2012). Resveratrol supplementation does not improve metabolic function in nonobese women with normal

glucose tolerance. Cell Metab 16, 658-664.

Yu C, Shin YG, Chow A, Li Y, Kosmeder JW, Lee YS, Hirschelman WH, Pezzuto JM, Mehta RG, & Van Breemen RB

(2002). Human, rat, and mouse metabolism of resveratrol. Pharm Res 19, 1907-1914.

Yuan M, Konstantopoulos N, Lee J, Hansen L, Li ZW, Karin M, & Shoelson SE (2001). Reversal of obesity- and diet-

induced insulin resistance with salicylates or targeted disruption of Ikkbeta. Science 293, 1673-1677.

Zahn JM, Sonu R, Vogel H, Crane E, Mazan-Mamczarz K, Rabkin R, Davis RW, Becker KG, Owen AB, & Kim SK

(2006). Transcriptional profiling of aging in human muscle reveals a common aging signature. PLoS Genet 2, e115.

Zhang G & Ghosh S (2001). Toll-like receptor-mediated NF-kappaB activation: a phylogenetically conserved paradigm

in innate immunity. J Clin Invest 107, 13-19.

73

Zhang Y, Huypens P, Adamson AW, Chang JS, Henagan TM, Boudreau A, Lenard NR, Burk D, Klein J, Perwitz N,

Shin J, Fasshauer M, Kralli A, & Gettys TW (2009). Alternative mRNA splicing produces a novel biologically active

short isoform of PGC-1alpha. J Biol Chem 284, 32813-32826.

74

Appendix

Study I

Jesper Olesen, Signe Larsson, Ninna Iversen, Simi Yousafzai, Ylva Hellsten, Henriette Pilegaard

(2012). Skeletal muscle PGC-1α is required for Maintaining an Acute LPS-induced TNFα Response.

PlosOne, 7(2): 32222

Skeletal Muscle PGC-1a Is Required for Maintaining anAcute LPS-Induced TNFa ResponseJesper Olesen1*, Signe Larsson1, Ninna Iversen1, Simi Yousafzai1, Ylva Hellsten2, Henriette Pilegaard1

1 Centre of Inflammation and Metabolism and Copenhagen Muscle Research Centre, Department of Biology, University of Copenhagen, Copenhagen, Denmark,

2 Copenhagen Muscle Research Centre, Department of Sport Sciences, University of Copenhagen, Copenhagen, Denmark

Abstract

Many lifestyle-related diseases are associated with low-grade inflammation and peroxisome proliferator activated receptor ccoactivator (PGC)-1a has been suggested to be protective against low-grade inflammation. However, whether these anti-inflammatory properties affect acute inflammation is not known. The aim of the present study was therefore to investigatethe role of muscle PGC-1a in acute inflammation. Quadriceps muscles were removed from 10-week old whole body PGC-1aknockout (KO), muscle specific PGC-1a KO (MKO) and muscle-specific PGC-1a overexpression mice (TG), 2 hours after anintraperitoneal injection of either 0.8 mg LPS/g body weight or saline. Basal TNFa mRNA content was lower in skeletalmuscle of whole body PGC-1a KO mice and in accordance TG mice showed increased TNFa mRNA and protein level relativeto WT, indicating a possible PGC-1a mediated regulation of TNFa. Basal p65 phosphorylation was increased in TG micepossibly explaining the elevated TNFa expression in these mice. Systemically, TG mice had reduced basal plasma TNFalevels compared with WT suggesting a protective effect against systemic low-grade inflammation in these animals. While TGmice reached similar TNFa levels as WT and showed more marked induction in plasma TNFa than WT after LPS injection,MKO PGC-1a mice had a reduced plasma TNFa and skeletal muscle TNFa mRNA response to LPS. In conclusion, the presentfindings suggest that PGC-1a enhances basal TNFa expression in skeletal muscle and indicate that PGC-1a does not exertanti-inflammatory effects during acute inflammation. Lack of skeletal muscle PGC-1a seems however to impair the acuteTNFa response, which may reflect a phenotype more susceptible to infections as also observed in type 2 diabetes patients.

Citation: Olesen J, Larsson S, Iversen N, Yousafzai S, Hellsten Y, et al. (2012) Skeletal Muscle PGC-1a Is Required for Maintaining an Acute LPS-Induced TNFaResponse. PLoS ONE 7(2): e32222. doi:10.1371/journal.pone.0032222

Editor: Jose A. L. Calbet, University of Las Palmas de Gran Canaria, Spain

Received October 26, 2011; Accepted January 22, 2012; Published February 27, 2012

Copyright: � 2012 Olesen et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The present study was supported by a grant from the Augustinus Foundation, Denmark. The Centre of Inflammation and Metabolism (CIM) issupported by a grant from the Danish National Research Foundation (# 02-512-55). CIM is part of the UNIK Project: Food, Fitness & Pharma for Health andDisease, supported by the Danish Ministry of Science, Technology, and Innovation. CIM is a member of DD2 - the Danish Center for Strategic Research in Type 2Diabetes (the Danish Council for Strategic Research, grant no. 09-067009 and 09-075724). The Copenhagen Muscle Research Centre is supported by a grant fromthe Capital Region of Denmark. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing Interests: The authors have declared that no competing interests exist.

* E-mail: [email protected]

Introduction

The transcriptional coactivator peroxisome proliferator activated

receptor c (PGC)-1a is known to influence many aspects of

metabolism and the role of PGC-1a as a master regulator of

mitochondrial biogenesis and oxidative metabolism in skeletal

muscle (SkM) has been confirmed repeatedly over the last decade

[1–3]. Recently it has been suggested that PGC-1a also exerts anti-

inflammatory effects. Hence overexpressing PGC-1a specifically in

SkM reduces an age-associated increase in serum TNFa as well as

TNFa mRNA and protein expression in SkM [4]. In accordance

TNFa mRNA expression and serum TNFa increases dramatically

after a single exercise bout in muscle specific PGC-1a knockout

mice, but not in WT [5]. Additionally, a negative correlation has

been reported between PGC-1a and TNFa mRNA levels in SkM of

type 2 diabetes patients independent of BMI [6]. Taken together,

these observations indicate that PGC-1a either directly or through

PGC-1a mediated adaptations in SkM provides anti-inflammatory

effects that potentially contribute to preventing low-grade inflam-

mation present in many life-style related diseases [7].

Inflammatory mediators have however also been suggested to

regulate PGC-1a expression [8–10]. Previous findings in human

cardiomyocytes indicate that nuclear factor kappa light chain-

enhancer of activated B cells (NFkB) binds to and inactivates PGC-

1a and that TNFa enhances such NFkB mediated PGC-1ainhibition and the concomitant metabolic effects of PGC-1a [8]. In

addition, a recent study shows that lipopolysaccharide (LPS)

downregulates PGC-1a expression and reduces fat oxidation in

mouse cardiomyocytes and mouse heart in vivo [10]. On the other

hand, incubating C2C12 cells with TNFa has been reported to

increase PGC-1a dependent transcriptional activity, and LPS

treatment in mice has been shown to increase mitochondrial

respiration when PGC-1a is elevated in SkM [9]. This suggests that

PGC-1a itself is regulated during acute inflammation and these

somewhat contradicting observations underline, that the role of

PGC-1a in acute inflammation is not fully elucidated. Based on the

findings that PGC-1a mediated adaptations in SkM protect against

low-grade inflammation, it may be expected that SkM PGC-1aimpairs the ability to elicit a robust acute inflammatory response.

An acute inflammatory response is a conserved mechanism

highly essential for protection against invading microorganisms in

all species. The response involves the induction of both pro-

(TNFa, IL-6) and anti-inflammatory (IL-6 and IL-10) cytokines

mediated through p38 mitogen-activated protein kinases (p38) as

well as NFkB and c-Jun N-terminal kinase (JNK) signaling [11].

LPS treatment is a well-established and frequently used model to

PLoS ONE | www.plosone.org 1 February 2012 | Volume 7 | Issue 2 | e32222

induce acute inflammation in humans and rodents [12–14].

Although circulating and infiltrating immune cells are seen as the

traditional responders during acute inflammation, LPS stimulation

has been shown to evoke p38, NFkB and JNK signaling and

concomitant production of cytokines like TNFa in many tissues

including SkM [9,14,15], which likely contributes to systemic

levels during acute inflammation. This is supported by the findings

that SkM functions as an endocrine organ [16] and that C2C12

cells incubated with LPS have been reported to produce and

secrete IL-6 to the medium [15].

The aim of the present study was to investigate whether the level

of PGC-1a expression in skeletal muscle affects an acute

inflammatory response locally in skeletal muscle and systemically.

This was addressed by giving LPS injections to PGC-1a whole

body knockout mice (KO), muscle specific PGC-1a knockout mice

(MKO) and transgenic mice overexpressing PGC-1a specifically in

SkM (TG). In addition, to investigate the potential secretion of

TNFa from muscle cells, primary myotubes from mouse skeletal

muscle were treated with LPS.

Materials and Methods

Ethics StatementExperiments were approved by the ‘‘Animal Experiment

Inspectorate’’ in Denmark (permission number: 2009/561-1607)

and complied with the European convention for the protection of

vertebrate animals used for experiments and other scientific

purposes (Council of Europe, no. 123, Strasbourg, France, 1985).

MiceGeneration and phenotypes of the whole body PGC-1a KO

mice, MKO PGC-1a and TG PGC-1a mice used in the present

study have been described elsewhere [1,17,18]. Whole body PGC-

1a KO mice and littermate WT were obtained by crossbreeding of

heterozygous PGC-1a KO parents. MKO and littermate WT

were obtained by crossing a +/2 Cre, Flox/Flox parent with a

Flox/Flox parent, while TG and their littermate WT were

obtained by crossbreeding of TG and WT parents. The genotypes

of the mice were determined by PCR-based genotyping, as

previously described [19]. Mice were kept on a 12:12 hour light/

dark cycle and had access to water and standard rodent chow ad

libitum (Altromin no. 1324, Brogarden, Lynge, Denmark).

Experimental protocolEach group consisted of 10 mice with equal number of male and

female mice. Ten weeks old PGC-1a KO, MKO PGC-1a, TG

PGC-1a and their respective littermate WT mice were given an

intraperitoneal injection of either saline as control or 0.8 mg LPS

(Sigma, St Louis, MO, USA) per gram mouse, dissolved in sterile

isotonic saline. Based on a preceding pilot study (data not shown)

and previous studies [14,15,20] showing a pronounced inflamma-

tory plasma and mRNA response 2 hours after LPS treatment, all

mice in the present study were euthanized by cervical dislocation

2 hours post injection. Trunk blood was collected after decapita-

tion and quadriceps muscles were quickly removed from all three

mouse strains and frozen in liquid nitrogen. In addition visceral

adipose tissue and liver were also removed from whole body PGC-

1a KO mice. Samples were stored at 280uC until analyzed.

Primary cell culturesC57BL/6 mice were euthanized by cervical dislocation and

limb skeletal muscles were quickly dissected out and placed in

15 ml ice cold Dulbecco’s phosphate buffered saline solution

(DPBS; Invitrogen, Carlsbad, CA, USA) containing 1% glucose

and 0.5% penicillin/streptomycin (Invitrogen) and placed on ice.

Satellite cells were isolated and handled as previously described

[21]. One day prior to the experiment, the Fusion Medium (FM)

containing DMEM (Invitrogen) with 10% horse serum and 0.1%

L-glutamine was exchanged with DMEM without phenol red

(Invitrogen) containing 0.5% glucose, 0.05% L-glutamin and 0.1%

BSA (DMEM without phenol red). Cells were used for

experiments on day 8, where sufficient maturation and differen-

tiation of the myotubes were observed. On the day of the

experiment, the cells were washed in DMEM without phenol red.

Half of the cell cultures (n = 6, in 3 independent experiments) were

treated with 1.0 mg LPS/ml media and the other half was treated

with DMEM without phenol red as control. LPS was dissolved in

dimethyl sulfoxide (DMSO) and then diluted in DMEM without

phenol red. After 2 hours of incubation, the medium was collected

and the cells were harvested in Trizol reagent (Invitrogen). The

samples were stored at 280uC until analyzed.

Cell culture mediaThe TNFa protein content in media from the incubated

primary cell cultures was determined by Enzyme-linked immuno-

sorbent assay (ELISA) according to the protocol of the

manufacturer (eBioscience, San Diego, CA, USA). The absorption

at 450 nm was determined and the TNFa protein content in the

medium was calculated based on a standard curve generated from

a serial diluted standard.

Plasma cytokinesPlasma was processed from the blood samples by centrifugation

(2600 g for 15 min at 4uC). A MSD multiplex ELISA kit

(MesoScaleDiscovery, Gaithersburg, Maryland, USA) pre-coated

with antibodies against TNFa, IL-6 and IL-10 was used according

to the manufacturer’s protocol. MSD plates were measured on a

MSD Sector Imager 2400 plate reader. Raw data were measured

as electrochemiluminescence signal (light) detected by photode-

tectors and analyzed using the Discovery Workbench 3.0 software

(MSD). A standard curve was generated for each analyte and used

to determine the concentration of analytes in each sample.

RNA isolation and Reverse TranscriptionTotal RNA was isolated from ,20–25 mg crushed muscle

tissue, liver and adipose tissue with the guanidinium thiocyanate-

phenol-chloroform method as previously described [22,23] and in

the cell culture experiment with the Trizol method following the

manufacture’s guidelines (Invitrogen). The final pellets were re-

suspended in DEPC treated H2O containing 0.1 mM EDTA.

RNA was quantified by measuring the absorbance at 260 nm.

Reverse transcription (RT) was performed using the Superscript

II RNase H2 system (Invitrogen) as previously described [23] and

the cDNA samples were diluted in nuclease-free H2O.

Real-time PCRReal-time PCR was performed using an ABI 7900 sequence-

detection system (Applied Biosystems, Foster City, CA). Primers

and TaqMan probes for amplifying gene-specific mRNA frag-

ments were designed using the database from ensemble.org and

Primer Express (Applied Biosystems). All TaqMan probes were 59-

FAM and 39-TAMRA labeled, and primers and Taqman probes

were obtained from TAG Copenhagen (Copenhagen, Denmark)

except GAPDH (59-FAM and 39-TAMRA) and beta-actin (59-VIC

and 39-nonflourescence), both obtained as a pre-developed assay

reagent (Applied Biosystems). The sequences are given in Table 1.

Real-time PCR was performed in triplicates in a total reaction

Skeletal Muscle PGC-1a and Acute Inflammation


volume of 10 ml using Universal Mastermix (Applied Biosystems)

except TNFa converting enzyme (TACE), which was amplified

using SYBGreen (Applied Biosystems). Cycle threshold (Ct) was

converted to a relative amount by use of a standard curve

constructed from a serial dilution of a pooled RT sample run

together with the samples. For muscle samples, target gene mRNA

content was for each sample normalized to single-stranded cDNA

content determined by OliGreen reagent (Molecular Probes,

Leiden, The Netherlands) as previously described [24]. For liver

and cell culture samples, target gene mRNA content was

normalized to beta-actin mRNA, which was unaffected by

genotype and treatment in the respective samples and for adipose

tissue samples, target gene mRNA content was normalized to

GAPDH mRNA, which was unaffected by genotype and

treatment in these samples.

Muscle lysateCrushed ,20–25 mg quadriceps muscles were homogenized in

an ice-cold buffer (10% Glycerol, 20 mM Na-pyrophosphate,

150 mM NaCl, 50 mM Hepes, 1% NP-40, 20 mM b-glycerophos-

phate, 10 mM NaF, 1 mM EDTA, 1 mM EGTA, 2 mM PMSF,

10 mg/ml Aprotinin, 10 mg/ml Leupeptin, 2 mM Na3VO4, 3 mM

Benzamidine, pH 7.5) for 1 min using a tissuelyser (TissueLyser II;

QIAGEN, Germany) with 30 oscillations per second. Homogenates

were rotated end over end for 30 min at 4uC. The procedure with

the tissuelyser and the end over end rotation was repeated. Lysates

were generated by centrifugation at 16,000 g for 20 min at 4uC and

collection of the protein supernatant (lysate). Protein content in

lysates was measured by the bicinchoninic acid method (Thermo

Scientific, Rockford, IL, USA).

SDS-PAGE and western blottingTNFa protein as well as phosphorylation of p38Thr180, Tyr182

and p65Ser536, the active subunit of NFkB, was measured in

muscle lysates by SDS-PAGE (10% or 15% Tris-HCl gel, BioRad,

Denmark) and western blotting using PVDF membrane and semi-

dry transfer. Twentyfive mg protein lysate was loaded for TNFaand 15 mg protein lysate for p65 and p38 phosphorylation. After

the transfer, the PVDF membrane was blocked for 1 h at room

temperature (TBST+5% bovine serum albumin (BSA)) and then

incubated with primary antibody (1:1000 in TBST+5% BSA) over

night. Commercially available antibodies were used to detect

TNFa (Cell Signaling Technology #3707), p38Thr180, Tyr182 (Cell

Signaling Technology #4511) and p65Ser536 (Cell Signaling

Technology #3033). The following day, the membrane was

incubated with horseradish peroxidase-conjugated secondary

antibody (Dako, Denmark) for 1 h at room temperature (1:5000

in TBST+5% BSA). Immobilon Western (Millipore Corporation,

Table 1. Primer and Taqman probe sequences.

Forward primers Reverse primers Taqman Probes

TNFa 59 ATGGCCCAGACCCTCACA 39 59 TTGCTACGACGTGGGCTACA 39 59 TCAGATCATCTTCTCAAAATTCGAGTGACAAGC 39

IL-6 59 GCTTAATTACACATGTTCTCTGGGAAA 39 59 CAAGTGCATCATCGTTGTTCATAC 39 59 ATCAGAATTGCCATTGCACAACTCTTTTCTCAT 39

IL-10 59 AGAGAAGCATGGCCCAGAAAT 39 59 CAGGGGAGAAATCGATGACA 39 59 CAGGGGAGAAATCGATGACA 39

TACE 59 TGCAAGGCTGGGAAATGC 39 59 TTG CACGAGTTGTCAGTGTCAA 39

TLR4 59 TCTGATCATGGCACTGTTCTTCTC 39 59 CTGATCCATGCATTGGTAGGTAATATTA 39 59 CAGGAAGCTTGAATCCCTGCATAGAGGTAGTTC 39

F4/80 59 GGCTGCCTCCCTGACTTTC 39 59 TGCACTGCTTGGCATTGC 39 59 TCCTTTTGCAGTTGAAGTTTCCATATCCTTGG 39

Sequences for forward and reverse primers as well as Taqman probes used. Tumor necrosis factor (TNF)a, interleukin (IL)-6, IL-10, TNFa converting enzyme (TACE), toll-like receptor 4 (TLR4), macrophage specific marker (F4/80).doi:10.1371/journal.pone.0032222.t001

Figure 1. Plasma TNFa. Plasma tumor necrosis factor (TNF)a from whole body PGC-1a knockout (KO) (A), muscle specific PGC-1a KO (MKO) (B) andmuscle specific PGC-1a overexpression (TG) mice (C) and their respective littermate wild type (WT) mice, 2 hours after injection with either saline (Sal)as control or lipopolysaccharide (LPS). Values are presented as means 6 S.E. with n = 6–10 in each group, except WT of whole body KO strain injectedwith saline where n = 3, due to lack of blood. *: Significantly different from Sal within given genotype, p,0.05. #: Significantly different from WTwithin given treatment, p,0.05. Notice the bi-sected y-axis.doi:10.1371/journal.pone.0032222.g001



MA) was used as detection system and bands were visualized using

an Eastman Kodak Co. Image Station 2000 MM. Band intensity

was quantified using Kodak Molecular Imaging Software v. 4.0.3,

and protein content or phosphorylation was expressed as arbitrary

units relative to control samples loaded on each site of each gel.

StatisticsTwo-way ANOVA was used to test the effect of LPS and

genotype within each mouse strain. Student Newman Keuls post

hoc test was used to locate differences. A T-test was used to test if

there was any difference in the basal levels between genotypes

within each mouse strain. All values are presented as means 6

S.E. A P,0.05 was considered significant.

Results

LPS-induced TNFa responses in primary myotubesTo verify that skeletal muscle produces and secretes TNFa in

response to LPS, primary myotubes isolated from C57BL/6 mice

were incubated with LPS. LPS treatment increased (p,0.05)

TNFa mRNA ,30 fold and TNFa protein in the media increased

(p,0.05) markedly from non-detectable levels in control cells to

,400 pg/ml (data not shown).

Plasma cytokinesPlasma TNFa. TG mice overexpressing PGC-1a specifically

in skeletal muscle had reduced basal plasma TNFa level

(0.660.5 pg/ml) compared with WT (2.661.2 pg/ml), whereas

there was no difference either in whole body PGC-1a KO or in

MKO PGC-1a mice compared with WT (fig. 1). In all groups,

LPS increased (p,0.05) the plasma TNFa level. Whereas there

was no genotype difference in the LPS-induced plasma TNFaresponse between whole body PGC-1a KO mice and WT, MKO

PGC-1a mice had a reduced (p,0.05) response to LPS compared

with WT. Furthermore, as a consequence of the lowered plasma

TNFa level in TG PGC-1a mice in the basal state, a statistical

interaction was observed between TG PGC-1a mice and WT.

Thus the LPS-induced plasma TNFa response was higher (,2000

fold) in TG PGC-1a mice than in WT mice (,1000 fold).

Plasma IL-6. Basal plasma IL-6 levels were unaffected by

genotype. LPS induced a ,100–400 fold increase (p,0.05) in plasma

IL-6 in all groups with no difference between genotypes (table 2).

Plasma IL-10. Basal plasma IL-10 levels were unaffected by

genotype. LPS induced a ,20–30 fold increase (p,0.05) in plasma

IL-10 in all groups with no difference between genotypes (table 2).

Basal and LPS-induced mRNA levels in skeletal muscleTNFa mRNA. In the basal state, whole body PGC-1a KO

mice had ,60% lower (p,0.05) TNFa mRNA content in skeletal

muscle than WT and in line with this, TG mice had ,2 fold higher

(p,0.05) TNFa mRNA content than WT, whereas no significant

difference was detected between MKO and WT despite a visual

,20% reduction in MKO. LPS injections increased (p,0.05) the

TNFa mRNA level ,13–60 fold in all groups. While no genotype

differences were observed in the LPS response in the whole body

KO and TG strains, the TNFa mRNA level after LPS treatment

was in MKO mice ,50% lower (p,0.05) than in WT (fig. 2).

TACE mRNA. To further examine the regulation of TNFa, a

metalloproteinase, TACE, which cleaves the membrane-

associated precursor of TNFa to the soluble and biological

active form of TNFa, was analyzed. In general, MKO mice had

30–40% lower (p,0.05) TACE mRNA levels in SkM than WT

mice, both at basal conditions and after LPS injections. No

difference was observed in whole body PGC-1a KO or TG mice

Ta

ble

2.

Pla

sma

IL-6

and

IL-1

0.

Wh

ole

bo

dy

PG

C-1

aK

Ost

rain

PG

C-1

aM

KO

stra

inT

GP

GC

-1a

stra

in

WT

WT

KO

KO

WT

WT

MK

OM

KO

WT

WT

TG

TG

Sa

lL

PS

Sa

lL

PS

Sa

lL

PS

Sa

lL

PS

Sa

lL

PS

Sa

lL

PS

IL-6

(pg

/ml)

53

06

40

45

10

536

75

56

*2

136

42

37

24

56

10

31

9*

43

76

45

13

32

786

25

52

9*

49

06

55

13

25

876

35

83

5*

48

56

14

26

39

076

80

1*

26

56

48

63

25

36

23

46

*

IL-1

0(p

g/m

l)1

476

30

44

966

41

2*

15

26

43

31

866

67

2*

17

76

10

47

446

76

2*

20

36

26

40

576

12

08

*2

016

37

56

706

52

7*

14

26

24

33

776

51

9*

Pla

sma

inte

rle

uki

n(I

L)-6

and

IL-1

0fr

om

wh

ole

bo

dy

PG

C-1a

kno

cko

ut

(KO

),m

usc

lesp

eci

fic

PG

C-1a

KO

(MK

O)

and

mu

scle

spe

cifi

cP

GC

-1a

ove

rexp

ress

ion

(TG

)m

ice

and

the

irre

spe

ctiv

elit

term

ate

wild

typ

e(W

T)

mic

e,2

ho

urs

afte

rin

ject

ion

wit

he

ith

er

salin

e(S

al)

asco

ntr

ol

or

lipo

po

lysa

cch

arid

e(L

PS)

.Val

ue

sar

ep

rese

nte

das

me

ans

6S.

E.w

ith

n=

6–

10

ine

ach

gro

up

,exc

ep

tW

To

fw

ho

leb

od

yK

Ost

rain

inje

cte

dw

ith

salin

ew

he

ren

=3

,du

eto

lack

of

blo

od

.*:

Sig

nif

ican

tly

dif

fere

nt

fro

mSa

lw

ith

ing

ive

ng

en

oty

pe

,p

,0

.05

.d

oi:1

0.1

37

1/j

ou

rnal

.po

ne

.00

32

22

2.t

00

2



relative to WT. LPS had no effect on TACE mRNA levels in any

of the strains (table 3).

TLR4 mRNA. There were no differences in either basal

TLR4 mRNA expression or after LPS injections in any of the

strains (table 3).

IL-6 mRNA. MKO mice had ,50% lower (p,0.05) basal

SkM IL-6 mRNA level than WT, whereas no genotype differences

were observed in the basal IL-6 mRNA level in the whole body

KO or TG mouse strain. In all groups, LPS induced a 130–400

fold increase (p,0.05) in IL-6 mRNA with no genotype

differences (table 3).

IL-10 mRNA. In the basal state, whole body PGC-1a KO

mice had ,55% lower (p,0.05) SkM IL-10 mRNA level than

WT, whereas there were no significant genotype differences in the

MKO (despite a visual 57% reduction) or TG strains. LPS

increased (p,0.05) the IL-10 mRNA content ,2–6 fold in all

groups with no significant difference between genotypes (table 3).

F4/80 mNRA. There were no genotype differences in the

mRNA level of the macrophage specific marker, F4/80, in SkM of

any of the strains and LPS had no effect on the F4/80 mRNA

level, indicating no LPS-induced recruitment of macrophages to

skeletal muscle 2 hours after LPS treatment (table 3).

TNFa protein in skeletal muscleTNFa protein content. TG mice had ,9 fold higher

(p,0.05) basal SkM TNFa protein level relative to WT, while

there were no genotype differences in whole body KO and MKO

PGC-1a mouse strains at basal conditions. LPS did not alter SkM

TNFa protein content in any mouse strain, but MKO mice

treated with LPS had ,30% lower (p,0.05) TNFa protein than

WT treated with LPS (fig. 2).

Basal and LPS-induced intracellular signaling in skeletalmuscle

Phosphorylation of p65ser536. Basal p65 phosphorylation

was not affected by genotype in the whole body PGC-1a KO and

MKO strains, but notably TG mice overexpressing PGC-1a in

skeletal muscle had ,40% higher (p,0.05) basal SkM p65

phosphorylation than WT mice. This may explain the elevated

resting TNFa mRNA and protein level in these mice. LPS

increased (p,0.05) p65 phosphorylation ,30–70% in all three

mouse strains. The TG mice reached ,40% higher (p,0.05) p65

phosphorylation level after LPS treatment than WT, whereas

there were no differences between whole body PGC-1a KO or

MKO mice and their corresponding WT (fig. 3 and 4).

Figure 2. TNFa mRNA and protein content in quadriceps. Tumor necrosis factor (TNF)a mRNA and protein in quadriceps muscle from wholebody PGC-1a knockout (KO) (A,D), muscle specific PGC-1a KO (MKO) (B,E) and muscle specific PGC-1a overexpression (TG) mice (C,F) and theirrespective littermate wild type (WT) mice, 2 hours after injection with either saline (Sal) as control or lipopolysaccharide (LPS). Values are presented asmeans 6 S.E. with n = 10 in each group. *: Significantly different from Sal within given genotype, p,0.05. #: Significantly different from WT withingiven treatment, p,0.05.doi:10.1371/journal.pone.0032222.g002



Phosphorylation of p38Thr180, Tyr182. LPS induced a similar

,2 fold increase (p,0.05) in p38 phosphorylation in TG and WT

mice, whereas no significant LPS-induced p38 phosphorylation

was observed in the whole body PGC-1a KO or the MKO PGC-

1a strain. No genotype differences were observed in the LPS-

induced p38 phosphorylation in SkM in any of the strains (table 3).

Basal and LPS-induced mRNA levels in liver and adiposetissue from whole body PGC-1a KO mice

To examine whether PGC-1a regulates the response to acute

LPS-induced inflammation in other inflammatory tissues, liver and

adipose tissue mRNA was analyzed from whole body PGC-1a KO

mice and WT.

TNFa and IL-6 mRNA content. No genotype differences

were observed in TNFa and IL-6 mRNA in liver or adipose tissue

in the basal state. LPS induced a 6–89 fold increase (p,0.05) in

TNFa and IL-6 mRNA in liver and adipose tissue with no

difference between WT and KO mice (table 4).

TACE mRNA content. There was no genotype difference in

TACE mRNA level in the liver or the adipose tissueat basal

conditions and no changes with LPS were observed (table 4).IL-10 mRNA content. In the basal state, no genotype

differences were observed in IL-10 mRNA in either the liver or

adipose tissue. LPS increased (p,0.05) the IL-10 mRNA level in

the liver similarly in KO and WT mice. In the adipose tissue, IL-

10 mRNA only increased significantly (p,0.05) in WT, whereas

only visual increase (p = 0.094) were observed in PGC-1a KO

mice in response to LPS treatment (table 4).F4/80 mRNA content. No genotype differences were observed

in F4/80 mRNA content in either the liver or adipose tissue in the

basal state. F4/80 mRNA was unaffected by LPS treatment in both

liver and adipose tissue from WT and KO mice (table 4).

Discussion

The main findings of the present study are that whole body

PGC-1a KO mice had a reduced basal TNFa mRNA level and

Figure 3. Phosphorylation of p65. Phosphorylation of p65ser536 in quadriceps muscle from whole body PGC-1a knockout (KO) (A), muscle specificPGC-1a KO (MKO) (B) and muscle specific PGC-1a overexpression (TG) mice (C) and their respective littermate wild type (WT) mice, 2 hours afterinjection with either saline (Sal) as control or lipopolysaccharide (LPS). Values are presented as mean s6 S.E. with n = 10 within each group.*: Significantly different from Sal within given genotype, p,0.05. #: Significantly different from WT within given treatment, p,0.05.doi:10.1371/journal.pone.0032222.g003

Table 3. Skeletal muscle mRNA and phosphorylation levels.

Whole body PGC-1a KO strain PGC-1a MKO strain TG PGC-1a strain

WT WT KO KO WT WT MKO MKO WT WT TG TG

Sal LPS Sal LPS Sal LPS Sal LPS Sal LPS Sal LPS

TACE mRNA 160.1 1.160.1 0.960.1 1.160.2 160.2 160.2 0.660.2# 0.760.1# 160.04 1.160.1 1.160.1 1.260.1

TLR4 mRNA 160.1 160.1 160.1 1.260.1 160.2 160.2 0.660.2 0.960.2 160.1 0.960.1 0.960.1 0.860.1

IL-6 mRNA 160.3 213645* 1.760.5 231647* 160.2 230653* 0.560.1# 183670* 160.2 130625* 0.660.2 129640*

IL-10 mRNA 160.1 4.860.4* 0.460.1# 2.560.4* 160.3 2.260.5* 0.460.1 2.060.4* 160.1 2.960.3* 160.2 2.560.5*

F4/80 mRNA 160.1 1.060.1 1.160.2 0.960.2 160.3 0.960.2 0.460.2 0.860.2 160.1 1.0160.1 1.160.1 0.760.1

p38-p (AU) 160.2 1.360.2 1.260.2 1.660.1 160.3 160.2 0.460.04 1.060.1 160.1 1.560.1* 0.760.1 1.660.2*

TNFa converting enzyme (TACE), toll-like receptor 4 (TLR4), interleukin (IL)-6, IL-10, macrophage specific marker (F4/80) mRNA as well as p38Thr180, Tyr182 phosphorylation(p38-p) in quadriceps muscle from whole body PGC-1a knockout (KO), muscle specific PGC-1a KO (MKO) and muscle specific PGC-1a overexpression (TG) mice and theirrespective littermate wild type (WT) mice, 2 hours after injection with either saline (Sal) as control or lipopolysaccharide (LPS). Values are presented as means 6 S.E. withn = 10 in each group.*: Significantly different from Sal within given genotype, p,0.05.#: Significantly different from WT within given treatment, p,0.05.doi:10.1371/journal.pone.0032222.t003



that mice overexpressing PGC-1a in skeletal muscle increased the

basal TNFa mRNA and protein content in skeletal muscle,

suggesting a PGC-1a mediated regulation of TNFa expression in

skeletal muscle. However, basal plasma TNFa was reduced in mice

overexpressing PGC-1a suggesting diminished secretion of TNFaand indicating that high TNFa levels in SkM do not lead to systemic

low-grade inflammation. In addition, while high PGC-1a levels in

skeletal muscle elicited a more marked LPS-induced plasma TNFaresponse, muscle specific knockout of PGC-1a resulted in a lower

LPS-induced plasma TNFa and TNFa mRNA response, poten-

tially reflecting a phenotype more susceptible to infections.

By use of three different PGC-1a mice models, the present study

provides new insight to the role of SkM PGC-1a in basal and acute

inflammation. The present findings that LPS treatment resulted in

similar TNFa, IL-6 and IL-10 levels in skeletal muscle and plasma

in TG and WT mice provide evidence that high levels of PGC-1a in

SkM do not alter the level of inflammatory markers during acute

inflammation. Of notice is however, that although the TG PGC-1amice had a lower basal plasma TNFa level, LPS elicited a more

marked fold increase in these mice. In accordance, the reduced

LPS-induced TNFa response in MKO PGC-1a mice both

systemically and in skeletal muscle indicates that lack of PGC-1aor PGC-1a mediated metabolic adaptations impair the ability to

respond to acute inflammation. Together this suggests that high

SkM PGC-1a does not exert anti-inflammatory effects during acute

inflammation and SkM PGC-1a is required for a normal acute

inflammatory response. Interestingly the observed responses in

PGC-1a MKO and to some extent whole body PGC-1a KO mice,

resemble the previously reported attenuations in LPS-induced

plasma TNFa response in type 2 diabetes patients [25]. This may

suggest that low levels of PGC-1a in skeletal muscle, as observed in

physically inactive subjects [26] and in type 2 diabetes patients [27],

impair the inflammatory response to invading pathogens.

TLR4 is the main receptor mediating LPS-induced responses in

SkM. The observed similar levels of TLR4 mRNA independent of

genotype and treatment indicates, that the observed lower plasma

TNFa and SkM TNFa mRNA in MKO PGC-1a mice than WT

in the present study are not due to a difference at the receptor

level. The observed LPS-induced increase in phosphorylation of

the active subunit of NFkB, p65, and to some extent p38 signaling

supports that both NFkB and p38 contributed to the inflammatory

response after LPS treatment in the present study. However, lack

of genotype differences in the LPS-induced changes in signaling

indicates that neither NFkB (p65 phosphorylation) nor p38

signaling can explain the attenuated responses in plasma TNFaand TNFa mRNA in SkM observed in MKO PGC-1a mice.

Differences in NFkB or p38 signaling, earlier than 2 hour after

LPS injections, of course can not be ruled out. The more marked

fold increase in plasma TNFa in TG mice upon LPS treatment

may on the other hand be linked to the higher p65 phosphory-

lation level in TG than WT. However, the unaffected TNFaprotein content in TG mice in response to LPS is as such in

accordance with such a mechanism as more secretion could keep

the increased level constant. But the similar TNFa mRNA

response in TG and WT does not support that a difference in

NFkB signaling alone is responsible for the more marked plasma

Table 4. mRNA expression in adipose tissue and liver from whole body PGC-1a KO mice.

Adipose tissue Liver

WT WT KO KO WT WT KO KO

Sal LPS Sal LPS Sal LPS Sal LPS

TNFa mRNA 160.4 12.262.1* 1.360.4 7.661.0* 160.2 36.263.5* 1.260.2 29.362.9*

TACE mRNA 160.2 2.260.6 1.360.2 1.060.2 160.2 1.260.1 0.960.1 1.060.1

IL-6 mRNA 160.3 38.867.2* 1.761.2 22.667.1* 160.3 88.869.7* 1.660.5 63.4610.5*

IL-10 mRNA 160.2 4.060.9* 1.260.2 1.860.3 160.4 6.761.3* 1.060.2 5.060.6*

F4/80 mRNA 160.2 1.560.4 1.160.2 0.460.1 160.1 0.960.1 1.060.2 0.860.1

Tumor necrosis factor (TNF)a, TNFa converting enzyme (TACE), interleukin (IL)-6, IL-10, macrophage specific marker (F4/80) mRNA in adipose tissue and liver from wholebody PGC-1a knockout (KO) and WT mice, 2 hours after injection with either saline (Sal) as control or lipopolysaccharide (LPS). Values are presented as means 6 S.E.with n = 10 in each group.*: Significantly different from Sal within given genotype, p,0.05.doi:10.1371/journal.pone.0032222.t004

Figure 4. Representative blots. Representative blot of tumor necrosis factor (TNF)a protein, phosphorylation of p65 protein (p65-p) andphosphorylation of p38 (p38-p) in quadriceps muscle from whole body PGC-1a knockout (KO), muscle specific PGC-1a KO (MKO) and muscle specificPGC-1a overexpression (TG) mice and their respective littermate wild type (WT) mice, 2 hours after injection with either saline (Sal) as control orlipopolysaccharide (LPS).doi:10.1371/journal.pone.0032222.g004



TNFa response in TG mice. As the LPS-induced SkM IL-6 and

IL-10 mRNA and plasma levels are independent of genotype, the

regulatory mediators and signaling factors involved in eliciting the

observed TNFa genotype difference in MKO PGC-1a mice in

response to LPS injections appear to be TNFa specific.

Interestingly, the finding that MKO PGC-1a mice, both in the

basal state and in response to LPS, had reduced SkM mRNA

expression of the TNFa regulatory enzyme TACE [28,29]

indicates that these mice may have a decreased ability to cleave

TNFa into its biological active form and thereby a reduced

secretion of TNFa protein from SkM relative to WT. Such a

mechanism may at least to some extent contribute to the observed

impaired LPS-induced plasma TNFa response when PGC-1a is

lacking in SkM.

The findings that whole body PGC-1a KO mice had lowered

basal TNFa mRNA level and that TG mice had elevated basal

TNFa mRNA and protein level in skeletal muscle compared with

WT are in contrast to previous observations in 22 months old TG

PGC-1a mice [4] and young MKO PGC-1a mice [6]. The

elevated TNFa levels in skeletal muscle of TG mice in the present

study may be explained by the observed higher level of basal p65

phosphorylation in skeletal muscle of TG mice. NFkB is activated

by several different stress associated stimuli such as LPS and TNFa[30], but is also known to be a redox-sensitive transcription factor.

An increased basal p65 activity in TG mice could possibly be due

to an imbalance in ROS production and antioxidant defense,

which could lead to an altered redox state in the mitochondria of

these mice. However, based on enhanced superoxide dismutase 2

protein content in skeletal muscle of TG PGC-1a mice in a

previous [4] and the present study (data not shown) as well as

reduced protein carbonylation and oxidized nucleic acids in old

TG PGC-1a mice compared with age-matched controls [4], there

are no indications of elevated ROS in SkM muscle of TG PGC-1aoverexpression mice. Thus, increased mitochondrial ROS levels

do not seem to be the explanation for the elevated basal p65

phosphorylation and the mechanism responsible for the enhanced

p65 phosphorylation in SkM of TG mice remains to be

determined.

The finding that TG PGC-1a mice had elevated SkM TNFamRNA and protein and reduced plasma TNFa relative to WT in

the basal state, while there were no genotype differences in SkM

and plasma for IL-6 and IL-10, again indicates a TNFa specific

regulation. As TACE is required for the conversion of the TNFaprecursor to circulating TNFa, a specific regulation of TACE

could potentially have contributed to the observed PGC-1adependent difference. However, the observed similar TACE

mRNA levels in SkM of WT and TG mice indicate that the

amount of TACE is not responsible for the observed differences

between TNFa expression in SkM and systemic levels in TG mice

in the basal state, although the present study cannot elucidate

whether the activity of TACE is altered in these mice. Another

possible explanation for the discrepancy could be that the

contribution from skeletal muscle plays a minor role regarding

systemic plasma TNFa levels, because TNFa released from

skeletal muscle is diluted in the circulation. Cytokine production

is usually believed to originate from traditional immune cells like

macrophages, lymphocytes and monocytes in the blood stream

and infiltrated in various tissues. Adipose tissue and the liver are

also both known to respond to acute inflammation, which is also

supported by the present increases in TNFa, IL-6 and IL-10

mRNA in liver and adipose tissue in response to LPS. But as

skeletal muscle is the largest organ of the body and previously has

been shown to produce cytokines [14,31,32], it may be speculated

that skeletal muscle plays a significant overall role during acute

inflammation. The present observation that primary myotubes

incubated with LPS produced and secreted TNFa supports that

skeletal muscle may be an important endocrine tissue during acute

inflammation and suggests that skeletal muscle could be classified

as an immunological tissue. In addition, the findings that the

mRNA level of the macrophage specific marker F4/80 was

unchanged with LPS treatment, suggest that the inflammatory

response does not originate from newly recruited macrophages but

rather resident macrophages or myofibers as suggested from the

cell culture results. The magnitude by which SkM contributes to

the circulating cytokine level after LPS treatment is difficult to

assess and additional studies are needed to elucidate this aspect.

The present finding that overexpression of PGC-1a in skeletal

muscle decreased the systemic plasma TNFa level supports a

previous study showing that 22 months old TG PGC-1a mice

were protected against an age-associated increase in serum TNFa[4]. Because elevated systemic TNFa has been associated with

obesity [33] and type 2 diabetes [34] and TNFa is known to

induce insulin resistance in vitro [35] and in vivo [36], the lowered

plasma TNFa in TG mice would therefore be expected to have

beneficial effects. Such a potential ability of high SkM PGC-1a to

protect against systemic low-grade inflammation and at the same

time posesses the ability to elicit a robust acute inflammatory

response underlines that different molecular mechanism mediate

these processes. Still, it may be noted, that no previous studies

have reported improvements in insulin sensitivity in young TG

PGC-1a mice and in fact when fed a high-fat diet for 3 weeks, TG

PGC-1a mice become insulin resistant compared with WT

controls [37]. In addition, the basal plasma TNFa level in the

present study was relatively low (,3 pg/ml) in both TG and WT

mice and therefore, the observed genotype difference in plasma

TNFa may not necessarily result in metabolic differences at this

age.

Finally, to elucidate the potential role of PGC-1a in the acute

inflammatory response in other tissues than skeletal muscle, the

liver and adipose tissue of whole body PGC-1a KO mice were

examined. The observed similar LPS-induced TNFa, IL-6 and IL-

10 mRNA responses in liver and adipose tissue from whole body

PGC-1a KO mice and WT mice does not indicate that PGC-1a is

necessary for a full inflammatory response in these tissues.

In conclusion, PGC-1a seems to be important for basal TNFaexpression in skeletal muscle, potentially via effects on p65

phosphorylation. While mice overexpressing PGC-1a in SkM

showed more marked plasma TNFa induction and reaching

similar TNFa levels after LPS injection as WT, lack of PGC-1aand/or lack of PGC-1a mediated metabolic regulation in skeletal

muscle impair a LPS-induced TNFa response both systemically

and in skeletal muscle. This suggests that low levels of SkM PGC-

1a could lead to an inflexible phenotype with impaired ability to

cope with infections. Whether such PGC-1a associated effects are

related to direct effects of PGC-1a or are secondary effects of

PGC-1a mediated metabolic regulation remains to be determined.

Acknowledgments

We would like to thank Professor Bruce M. Spiegelman (Harvard Medical

School, Boston, Massachusetts, USA) for the kind donation of breeding

pairs to the three different PGC-1a mice strains to start breeding initially.

Karina Olsen is acknowledged for skillful technical assistance.

Author Contributions

Conceived and designed the experiments: JO HP YH. Performed the

experiments: JO HP . Analyzed the data: JO HP SL SY NI. Contributed

reagents/materials/analysis tools: HP YH. Wrote the paper: JO HP.



References

1. Lin J, Wu H, Tarr PT, Zhang CY, Wu Z, et al. (2002) Transcriptional co-activator PGC-1 alpha drives the formation of slow-twitch muscle fibres. Nature

418: 797–801. 10.1038/nature00904 [doi];nature00904 [pii].2. Wende AR, Schaeffer PJ, Parker GJ, Zechner C, Han DH, et al. (2007) A role

for the transcriptional coactivator PGC-1alpha in muscle refueling. J Biol Chem282: 36642–36651. M707006200 [pii];10.1074/jbc.M707006200 [doi].

3. Wu Z, Puigserver P, Andersson U, Zhang C, Adelmant G, et al. (1999)

Mechanisms controlling mitochondrial biogenesis and respiration through thethermogenic coactivator PGC-1. Cell 98: 115–124. S0092-8674(00)80611-X

[pii];10.1016/S0092-8674(00)80611-X [doi].4. Wenz T, Rossi SG, Rotundo RL, Spiegelman BM, Moraes CT (2009) Increased

muscle PGC-1alpha expression protects from sarcopenia and metabolic disease

during aging. Proc Natl Acad Sci U S A 106: 20405–20410. 0911570106[pii];10.1073/pnas.0911570106 [doi].

5. Handschin C, Chin S, Li P, Liu F, Maratos-Flier E, et al. (2007) Skeletal musclefiber-type switching, exercise intolerance, and myopathy in PGC-1alpha muscle-

specific knock-out animals. J Biol Chem 282: 30014–30021. M704817200

[pii];10.1074/jbc.M704817200 [doi].6. Handschin C, Choi CS, Chin S, Kim S, Kawamori D, et al. (2007) Abnormal

glucose homeostasis in skeletal muscle-specific PGC-1alpha knockout micereveals skeletal muscle-pancreatic beta cell crosstalk. J Clin Invest 117:

3463–3474. 10.1172/JCI31785 [doi].7. Handschin C, Spiegelman BM (2008) The role of exercise and PGC1alpha in

inflammation and chronic disease. Nature 454: 463–469. nature07206

[pii];10.1038/nature07206 [doi].8. Alvarez-Guardia D, Palomer X, Coll T, Davidson MM, Chan TO, et al. (2010)

The p65 subunit of NF-kappaB binds to PGC-1alpha, linking inflammation andmetabolic disturbances in cardiac cells. Cardiovasc Res 87: 449–458. cvq080

[pii];10.1093/cvr/cvq080 [doi].

9. Puigserver P, Rhee J, Lin J, Wu Z, Yoon JC, et al. (2001) Cytokine stimulation ofenergy expenditure through p38 MAP kinase activation of PPARgamma

coactivator-1. Mol Cell 8: 971–982. S1097-2765(01)00390-2 [pii].10. Schilling J, Lai L, Sambandam N, Dey CE, Leone TC, et al. (2011) Toll-like

Receptor-mediated Inflammatory Signaling Reprograms Cardiac EnergyMetabolism by Repressing Peroxisome Proliferator-activated Receptor {gam-

ma} Coactivator-1 (PGC-1) Signaling. Circ Heart Fail. CIRCHEARTFAI-

LURE.110.959833 [pii];10.1161/CIRCHEARTFAILURE.110.959833 [doi].11. Carpenter S, O’Neill LA (2009) Recent insights into the structure of Toll-like

receptors and post-translational modifications of their associated signallingproteins. Biochem J 422: 1–10. BJ20090616 [pii];10.1042/BJ20090616 [doi].

12. Andreasen AS, Krabbe KS, Krogh-Madsen R, Taudorf S, Pedersen BK, et al.

(2008) Human endotoxemia as a model of systemic inflammation. Curr MedChem 15: 1697–1705.

13. Favorite GO, Morgan HR (1942) Effects produced by the intravenous injectionin man of a toxic antigenic material derived from eberthella typhosa: Clinical,

hematological, chemical and serological studies. J Clin Invest 21: 589–599.10.1172/JCI101337 [doi].

14. Frost RA, Nystrom GJ, Lang CH (2002) Lipopolysaccharide regulates

proinflammatory cytokine expression in mouse myoblasts and skeletal muscle.Am J Physiol Regul Integr Comp Physiol 283: R698–R709. 10.1152/

ajpregu.00039.2002 [doi].15. Frost RA, Nystrom GJ, Lang CH (2003) Lipopolysaccharide and proinflamma-

tory cytokines stimulate interleukin-6 expression in C2C12 myoblasts: role of the

Jun NH2-terminal kinase. Am J Physiol Regul Integr Comp Physiol 285:R1153–R1164. 10.1152/ajpregu.00164.2003 [doi];00164.2003 [pii].

16. Brandt C, Pedersen BK (2010) The role of exercise-induced myokines in musclehomeostasis and the defense against chronic diseases. J Biomed Biotechnol 2010:

520258. 10.1155/2010/520258 [doi].17. Geng T, Li P, Okutsu M, Yin X, Kwek J, et al. (2010) PGC-1alpha plays a

functional role in exercise-induced mitochondrial biogenesis and angiogenesis

but not fiber-type transformation in mouse skeletal muscle. Am J Physiol CellPhysiol 298: C572–C579. ajpcell.00481.2009 [pii];10.1152/ajpcell.00481.2009

[doi].18. Lin J, Wu PH, Tarr PT, Lindenberg KS, St-Pierre J, et al. (2004) Defects in

adaptive energy metabolism with CNS-linked hyperactivity in PGC-1alpha null

mice. Cell 119: 121–135. 10.1016/j.cell.2004.09.013 [doi];S0092867404008864[pii].

19. Leick L, Wojtaszewski JF, Johansen ST, Kiilerich K, Comes G, et al. (2008)

PGC-1alpha is not mandatory for exercise- and training-induced adaptive gene

responses in mouse skeletal muscle. Am J Physiol Endocrinol Metab 294:E463–E474. 00666.2007 [pii];10.1152/ajpendo.00666.2007 [doi].

20. Huey KA, Meador BM (2008) Contribution of IL-6 to the Hsp72, Hsp25, and

alphaB-crystallin [corrected] responses to inflammation and exercise training in

mouse skeletal and cardiac muscle. J Appl Physiol 105: 1830–1836. 90955.2008

[pii];10.1152/japplphysiol.90955.2008 [doi].

21. Kusuhara K, Madsen K, Jensen L, Hellsten Y, Pilegaard H (2007) Calcium

signalling in the regulation of PGC-1alpha, PDK4 and HKII mRNA expression.

Biol Chem 388: 481–488. 10.1515/BC.2007.052 [doi].

22. Chomczynski P, Sacchi N (1987) Single-step method of RNA isolation by acid

guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:

156–159. 10.1006/abio.1987.9999 [doi];0003-2697(87)90021-2 [pii].

23. Pilegaard H, Ordway GA, Saltin B, Neufer PD (2000) Transcriptional regulationof gene expression in human skeletal muscle during recovery from exercise.

Am J Physiol Endocrinol Metab 279: E806–E814.

24. Lundby C, Nordsborg N, Kusuhara K, Kristensen KM, Neufer PD (2005) Gene

expression in human skeletal muscle: alternative normalization method andeffect of repeated biopsies. Eur J Appl Physiol 95: 351–360. 10.1007/s00421-

005-0022-7 [doi].

25. Andreasen AS, Pedersen-Skovsgaard T, Berg RM, Svendsen KD, Feldt-

Rasmussen B, et al. (2010) Type 2 diabetes mellitus is associated with impairedcytokine response and adhesion molecule expression in human endotoxemia.

Intensive Care Med 36: 1548–1555. 10.1007/s00134-010-1845-1 [doi].

26. Alibegovic AC, Sonne MP, Hojbjerre L, Bork-Jensen J, Jacobsen S, et al. (2010)

Insulin resistance induced by physical inactivity is associated with multipletranscriptional changes in skeletal muscle in young men. Am J Physiol

Endocrinol Metab 299: E752–E763. ajpendo.00590.2009 [pii];10.1152/

ajpendo.00590.2009 [doi].

27. Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, et al. (2003)

PGC-1alpha-responsive genes involved in oxidative phosphorylation arecoordinately downregulated in human diabetes. Nat Genet 34: 267–273.

10.1038/ng1180 [doi];ng1180 [pii].

28. Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, et al. (1997) A

metalloproteinase disintegrin that releases tumour-necrosis factor-alpha fromcells. Nature 385: 729–733. 10.1038/385729a0 [doi].

29. Moss ML, Jin SL, Milla ME, Bickett DM, Burkhart W, et al. (1997) Cloning of a

disintegrin metalloproteinase that processes precursor tumour-necrosis factor-

alpha. Nature 385: 733–736. 10.1038/385733a0 [doi].

30. Barnes PJ, Karin M (1997) Nuclear factor-kappaB: a pivotal transcription factor

in chronic inflammatory diseases. N Engl J Med 336: 1066–1071. 10.1056/

NEJM199704103361506 [doi].

31. Borge BA, Kalland KH, Olsen S, Bletsa A, Berggreen E, et al. (2009) Cytokines

are produced locally by myocytes in rat skeletal muscle during endotoxemia.

Am J Physiol Heart Circ Physiol 296: H735–H744. 01309.2008 [pii];10.1152/

ajpheart.01309.2008 [doi].

32. Pedersen BK, Febbraio MA (2008) Muscle as an endocrine organ: focus on

muscle-derived interleukin-6. Physiol Rev 88: 1379–1406. 88/4/1379

[pii];10.1152/physrev.90100.2007 [doi].

33. Hotamisligil GS, Shargill NS, Spiegelman BM (1993) Adipose expression oftumor necrosis factor-alpha: direct role in obesity-linked insulin resistance.

Science 259: 87–91.

34. Hotamisligil GS (2006) Inflammation and metabolic disorders. Nature 444:

860–867. nature05485 [pii];10.1038/nature05485 [doi].

35. Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, et al. (1996) IRS-1-

mediated inhibition of insulin receptor tyrosine kinase activity in TNF-alpha-

and obesity-induced insulin resistance. Science 271: 665–668.

36. Plomgaard P, Bouzakri K, Krogh-Madsen R, Mittendorfer B, Zierath JR, et al.(2005) Tumor necrosis factor-alpha induces skeletal muscle insulin resistance in

healthy human subjects via inhibition of Akt substrate 160 phosphorylation.

Diabetes 54: 2939–2945. 54/10/2939 [pii].

37. Choi CS, Befroy DE, Codella R, Kim S, Reznick RM, et al. (2008) Paradoxicaleffects of increased expression of PGC-1alpha on muscle mitochondrial function

and insulin-stimulated muscle glucose metabolism. Proc Natl Acad Sci U S A

105: 19926–19931. 0810339105 [pii];10.1073/pnas.0810339105 [doi].



THE FACULTY OF SCIE } . ICE ,

U N I V E R S I T Y O F C O P E N H A G E NThe FhD School of SCIENCE

3,d. Co-authorshiP statement

All papers/manuscripts with multiple authors enclosed as annexes to a PhD thesis.synopsis

should contain a co-author statement, stating the PhD student's contribution to the paper.

ffi

Skeletal mus4e PGC-11 is required for maintaining an acute LPS'induced TNFa response'

The extent of the phD student's contribution to the article is assessed on the following scale

A. has contributed to the work (0-33%o)

B. has made a substantial contribution (34'66yo)

C. : did the majority of the work independently (67'l00yo)'

tl3Revised 29 lanuarY 2073

f ] _ t E , : ; r - ' _ , J i 1 ; , i - r , i , r { - { L l . l i ^ Fv s { L t t ! ! r

U N I V E R S i T Y O F C O P E N H A G E N

T'h'e FhD S,chool of SCIENCE

4. Declaration on the individual elements Extent (A, B, C)

l. Formulation in the concept phase of the basic scientific problem on the basis oftheoretical questions which require clarification, including a sunmary of the generalquestions which it is assumed will be answerable via analyses or concreteexperiments/investigations.

c

Planning of experiments/analyses and forrrulation of investigative methodology insuch a way that the questions asked under (1) can reasonably be expected to beanswered, including choice of method and independent methodological development.

2 .

C

3. Involvement in the analysis or the concrete experiments/investigation.C

4. Presentation, interpretation and discussion of the results obtained in article form. c

ijll3i,::s

ii!:;;t. ;lril.j

Does the paper contain data material, which has also formed part of a previous degree / thesis(e.g. your masters degree)

Please indicate which degree / thesis: Master thesis

v.r' lTlNo: f

Percentage ofthe paper that

Percentage ofthe paper that

is from the PhD degree work

is from the other degree / thesis

%

%

80

20

Please indicate which specific partGs) ofthe paper that has been produced as part ofthe PhD study:

All the experiments and analyses performed on the muscle specific PGC-1a KO mice has been conducted andanalyzed during the PhD. ln addition, the cell culture experiment has been conducted and analyzed during the PhD.lntacellular signaling and inflammatory markers has been analyzed in different tissues from the two other mice strainsdwing the PhD. The manuscript has been written during the PhD.

' '6.Signatures of co-authors:

a . . , :

Date Name Signalure

ttlft,- t3Sisne Larsson -Sgl*a- VarSy*

bl7 4\Ninna Iversen tl:ltu ( A

lela-te SimiYousafzai -(/<'{17-r: Ylva Hellsten l"4t ur1/{T-12/6- ri

Henriette PileeaardmdZAG/,"ase/

zl5Revised 29 Jonuary 2073

THE FAC I J i f Y OF SC iE i r {CE ,U N I V E R S I T Y O F C O P E N H A G E N

TheFhD Schootr of SCIENCE

When completed, the form should be sent to the relevant department's PhD secretary who on your behalf willforward it to the PhD School ([email protected]).

To find your departments PhD seuetary, please see: rhttp://wwur.science.ku.dk/enelish/research/phd/student/contact/

T.Signature

By signing 16s d6srrmetrt, the PbD student hereby declares that the above information is correct.

PbD student: o^," /7ft-/z(ddmm.yyly)

I

313Revised 29 January 2013

Study II

Jesper Olesen, Stine Ringholm, Maja M. Nielsen, Christina T. Brandt, Jesper T. Pedersen, Jens

Halling, Laurie J. Goodyear, Henriette Pilegaard (2013). Role of PGC-1α in exercise training- and

resveratrol-induced prevention of age-associated inflammation. In press, Exp Gerontol

1

2

3Q1

4

56

7

89101112131415161718192021222324

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

Experimental Gerontology xxx (2013) xxx–xxx

EXG-09240; No of Pages 11

Contents lists available at SciVerse ScienceDirect

Experimental Gerontology

j ourna l homepage: www.e lsev ie r .com/ locate /expgero

Role of PGC-1α in exercise training- and resveratrol-induced preventionof age-associated inflammation

OFJesper Olesen a,⁎, Stine Ringholm a, MajaM. Nielsen a, Christina T. Brandt a, Jesper T. Pedersen a, Jens F. Halling a,

Laurie J. Goodyear b, Henriette Pilegaard a

a Centre of Inflammation and Metabolism, August Krogh Centre, August Krogh Building, Department of Biology, University of Copenhagen, Copenhagen, Denmarkb Research Division, Joslin Diabetes Center, Harvard Medical School, Boston, MA, USA

⁎ Corresponding author at: August Krogh Building,Copenhagen Ø, Denmark. Tel.: +45 24467316.

E-mail address: [email protected] (J. Olesen).

0531-5565/$ – see front matter © 2013 Published by Elsehttp://dx.doi.org/10.1016/j.exger.2013.07.015

Please cite this article as: Olesen, J., et al., RoleExp. Gerontol. (2013), http://dx.doi.org/10.1

O

a b s t r a c t
a r t i c l e i n f o
25
Article history:
26

27

28

29

30

31

32

33

34

35

36

37

TED P

R

Received 11 June 2013Received in revised form 17 July 2013Accepted 24 July 2013Available online xxxx

Section Editor: Christiaan Leeuwenburgh

Keywords:AgingLow-grade inflammationExercise trainingResveratrolPGC-1α

Background/aim: Age-related metabolic diseases are often associated with low-grade inflammation. The aim ofthe present studywas to investigate the role of the transcriptional co-activator PGC-1α in the potential beneficialeffects of exercise training and/or resveratrol in the prevention of age-associated low-grade inflammation. Toaddress this, a long-term voluntary exercise training and resveratrol supplementation study were conducted.Experimental setup: Three month old whole body PGC-1α KO and WT mice were randomly assigned to fourgroups: untrained chow-fed, untrained chow-fed supplemented with resveratrol, chow-fed voluntarily exercisetrained and chow-fed supplemented with resveratrol and voluntarily exercise trained. The intervention lasted12 months and three month old untrained chow-fed mice served as young controls.Results: Voluntary exercise training prevented an age-associated increase (p b 0.05) in systemic IL-6 and adipos-ity inWTmice. PGC-1α expression was required for a training-induced prevention of an age-associated increase(p b 0.05) in skeletal muscle TNFα protein. Independently of PGC-1α, both exercise training and resveratrolprevented an age-associated increase (p b 0.05) in skeletal muscle protein carbonylation.Conclusion: The present findings highlight that exercise training is a more effective intervention than resveratrol

38

39
Csupplementation in reducing age-associated inflammation and that PGC-1α in part is required for the exercisetraining-induced anti-inflammatory effects.
40
© 2013 Published by Elsevier Inc.
4142
E
62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

UNCO

RR1. Introduction

Aging is associated with a broad range of metabolic complica-tions including increased adiposity (Schaap et al., 2012; Wu et al.,2007), loss of muscle mass and strength (Brooks and Faulkner,1994; Doherty et al., 1993) and oxidative stress (Harman, 1956).These unfavorable complications may be caused by aging per se orlifestyle such as decreased physical activity with increasing age.

Many factors are likely involved in the initiation and/or pro-gression of age-related diseases, but reports showing that the ma-jority of lifestyle-related diseases are associated with chronic low-grade inflammation (Handschin and Spiegelman, 2008; Wellenand Hotamisligil, 2005) support the possibility that low-gradeinflammation is an important component in the pathogenesisof these diseases (Pedersen et al., 2000; Woods et al., 2012). Inaddition, low-grade inflammation is observed in elderly subjectseven in the absence of chronic diseases (Wei et al., 1992), butwhether low-grade inflammation is a cause or consequence of

80

81

82

83

84

Universitetsparken 13, 2100

vier Inc.

of PGC-1α in exercise trainin016/j.exger.2013.07.015

age-related metabolic dysfunction is still debated (Woods et al.,2012).

Chronic low-grade inflammation is defined as sustained 2–4 foldelevations in systemic levels of pro-inflammatory cytokines like tu-mor necrosis factor (TNF)α and interleukin (IL)-6 (Bruunsgaard andPedersen, 2003; Woods et al., 2012). The immune system is thoughtto contribute to low-grade inflammation (Woods et al., 2012), but theability of several tissues like liver, adipose tissue and skeletal muscle(SkM) to express and secrete cytokines (Borge et al., 2009; Frost et al.,2002; Hotamisligil et al., 1993; Pedersen and Febbraio, 2012) raisesthe possibility that these highly metabolically active tissues alsocontribute to the systemic levels of inflammatory cytokines duringlow-grade inflammation. Circulating TNFα is increased in type 2 diabe-tes patients (Hotamisligil et al., 1995; Plomgaard et al., 2007) and TNFαhas been shown to induce insulin resistance in humans (Plomgaardet al., 2005) and rodents (Hotamisligil et al., 1994), indicating thatTNFα could be involved in the pathogenesis of type 2 diabetes. Thus,suppression of systemic TNFα may be expected to prevent thedevelopment and progression of lifestyle-related diseases.

Exercise training elicits a broad range of adaptations includingincreased skeletal muscle mass (Frontera et al., 1988) as well asincreased skeletal muscle oxidative (Holloszy, 1967) and anti-oxidantcapacity (Oh-ishi et al., 1997) and previous studies indicate that

g- and resveratrol-induced prevention of age-associated inflammation,

http://dx.doi.org/10.1016/j.exger.2013.07.015

mailto:[email protected]


http://www.sciencedirect.com/science/journal/05315565


T

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

2 J. Olesen et al. / Experimental Gerontology xxx (2013) xxx–xxx

UNCO

RREC

exercise training may have anti-inflammatory effects (Starkie et al.,2003; Woods et al., 2012). Intriguingly, the natural anti-oxidantresveratrol (RSV), primarily found in the skin of dark grapes, has beenreported to exert effects almost similar to exercise training. Hence,RSV has been shown to possess anti-inflammatory effects in rodentsand humans (Olholm et al., 2010; Pearson et al., 2008) as well as toprotect rodents from high fat diet-induced obesity and insulin resis-tance (Baur et al., 2006; Lagouge et al., 2006). Moreover, RSV has beenshown to increase longevity in several lower species (Howitz et al.,2003; Wood et al., 2004). Both RSV and exercise training have beenshown to activate the energy sensors AMP-activated protein kinase(AMPK) (Baur et al., 2006; Um et al., 2010) and sirtuin (SIRT)1(Lagouge et al., 2006), which both are believed to converge on thetranscriptional co-activator peroxisome proliferator activated receptor γco-activator (PGC)-1α (Canto et al., 2010; Jager et al., 2007).

PGC-1α is known as a master regulator of mitochondrial biogenesisand anti-oxidant defense (Leick et al., 2010; Lin et al., 2002; St-Pierreet al., 2006;Wu et al., 1999). The finding that PGC-1α expression is tran-siently increased in recovery froma single exercise bout (Baar et al., 2002;Pilegaard et al., 2003) suggests that PGC-1α is a likely mediator ofexercise training-induced adaptations in oxidative and anti-oxidantproteins in SkM. Supportive of this, previous findings have highlightedthe importance of PGC-1α in exercise training-induced preventionof age-associated reductions in oxidative and anti-oxidant proteins(Leick et al., 2010). Moreover, a positive correlation between physical ac-tivity level and SkM PGC-1αmRNA content exists in humans (Alibegovicet al., 2010). Accordingly, an inverse correlation between increasing ageand SkM PGC-1α mRNA level (Ling et al., 2004) may suggest that lackof exercise training-induced induction of PGC-1αwith aging contributesto age-associated deteriorations in skeletalmuscle. Recent studies inmicealso indicate that PGC-1α has anti-inflammatory effects (Handschin et al.,2007a, 2007b; Wenz et al., 2009). However, whether the potential anti-inflammatory effects of exercise training and RSV require PGC-1α andwhether combining RSV and exercise training elicits additive effects viaPGC-1α are still unresolved.

The aim of the present study was to test the hypothesis that PGC-1αis required for the beneficial effects of long-term exercise trainingand RSV supplementation in the prevention of age-associated low-grade inflammation. To address this, metabolic and inflammatorymarkers were determined in adipose tissue, liver, skeletal muscle andblood from whole body PGC-1α knockout (KO) and littermate wildtype (WT) mice after long-term voluntary exercise training and/orRSV supplementation.

2. Methods

2.1. Mice

Generation, phenotype and genotyping of whole body PGC-1α KOmice have previously been described in detail (Leick et al., 2008; Linet al., 2004). Briefly, whole body PGC-1α KO and littermate WT micewere obtained by crossbreeding of heterozygous whole body PGC-1αKO parents. The genotype was assessed on DNA extracted from asmall piece of the tail tip by the phenol-chloroform:isoamyl method.DNA fragments were amplified by PCR using specific primers againsttheWT and the KO alleles (Lin et al., 2004) and subsequently separatedon an agarose gel. Whole body PGC-1α KO mice have a CNS-linkedneurological disorder resembling Huntingtons disease making themmore anxious with sudden movements (Lin et al., 2004). They havereduced oxidative capacity in skeletal muscle (Leick et al., 2008; Leicket al., 2010; Lin et al., 2004), but normal glucose tolerance and do notdevelop diet-induced obesity (Lin et al., 2004). PGC-1α KO mice runvoluntarily less than WT mice when offered a running wheel (Leicket al., 2008, 2010).

Mice were kept on a 12:12 hour light/dark cycle and had access towater and food ad libitum. The experiments were approved by the

Please cite this article as: Olesen, J., et al., Role of PGC-1α in exercise traininExp. Gerontol. (2013), http://dx.doi.org/10.1016/j.exger.2013.07.015

ED P

RO

OF

“Animal Experiment Inspectorate” in Denmark (#2009/561-1689) andcompliedwith the European convention for the protection of vertebrateanimals used for experiments and other scientific purposes (Councilof Europe, no. 123, Strasbourg, France, 1985).

2.2. Experimental setup

Whole body PGC-1α KO and littermate WT female mice wererandomly assigned to a young group and four intervention groups withmice housed individually. The intervention groups were divided into:untrained mice receiving rodent chow (Altromin no. 1324, Brogården,Lynge, Denmark) (UT-C), untrained mice receiving chow supplementedwith RSV (UT-R), voluntary exercise trainedmice having access to a run-ning wheel (Minimitter, Italy) (T-C), and voluntary exercise trainedmicehaving access to a runningwheel and receiving chow supplementedwithRSV (T-R). The interventions lasted from 3 months to 15 months of age,where the mice were euthanized. The young mice receiving chow andserving as young UT-C were euthanized at 3 months of age. Each of thegroups consisted of 8–10 mice. Based on the observation that mice ofthe PGC-1α KO strain (6 WT mice and 6 PGC-1α KO mice), in ouranimal facility, on average reach ~18 months of age with no apparentdifference between WT and PGC-1α KO mice, we chose to examine15 month old mice as these mice would be in their last quartile of theirlifespan.

Based on previous studies from our group (Leick et al., 2008, 2010)we expected the PGC-1α KO mice to run less voluntarily than WTmice. Running distance and duration were monitored by a regularcycle computer and differences between WT and PGC-1α KO micewere daily adjusted by wheel blocking of WT for shorter periods toensure similar exercise distance between the different genotypesand interventions. The running distance was on average 5.9 ± 1.9(WT, T-C), 6.0 ± 2.4 (KO, T-C), 6.0 ± 1.2 (WT, T-R) and 5.8 ± 1.7(KO, T-R) km per week. Pure RSV (Orchid chemicals, India) wasmixed with chow to a concentration of 4 g RSV/kg chow as previouslyused (Lagouge et al., 2006; Um et al., 2010). The concentration wassubsequently confirmed by liquid chromatography-mass spectrometry(Eurofins, Denmark). Body weight and food intake were monitoredthroughout the experiment. Running wheels were blocked 24 h beforethe animals were euthanized. All mice were euthanized by cervicaldislocation followed by decapitation to collect trunk blood. Quadricepsmuscles, perigonadal visceral adipose tissue (V-AT), inguinal subcuta-neous adipose tissue (S-AT) and liver were quickly removed, quickfrozen in liquid nitrogen and stored at −80 °C until analyses. Thesetissues were chosen based on previous reports showing markedinflammatory responses as well as adaptations to metabolic challenges(Gollisch et al., 2009; Handschin et al., 2007b; Hotamisligil et al., 1993;Wenz et al., 2009).

A separate manuscript covers oxidative adaptations in skeletalmuscle from the current experimental setup (Ringholm et al. inreview, Exp Gerontology).

2.3. Analyses

2.3.1. Echo MRI scanningBody compositionwas determined byMRI scanning (EchoMRI, Echo

Medical Systems, Houston, TX, USA). A reduced number of animalswere measured (n = 3–8) due to limited access to the Echo MRIscanner.

2.3.2. PlasmaPlasma cytokineswere analyzed using an ultra-sensitive customized

MSD multi-spot assay system pre-coated with antibodies against TNFαand IL-6 (MesoScaleDiscovery, Gaithersburg, Maryland, USA) accordingto themanufacturer's protocol. The lower limit of detection (LLOD) was1.0 pg/ml for TNFα and 4.5 pg/ml for IL-6.



T

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

t1:1

t1:2

3J. Olesen et al. / Experimental Gerontology xxx (2013) xxx–xxx

REC

2.3.3. RNA isolation and reverse transcriptionTotal RNAwas isolated from crushed quadricepsmuscle (20–25 mg),

liver (20–25 mg) and V-AT (40–45 mg) by a modified guanidiniumthiocyanate-phenol-chloroform extraction method (Chomczynski andSacchi, 1987) as previously described (Pilegaard et al., 2000), exceptthat the tissue was homogenized for 2 min at 30 s−1 in a tissue lyser(TissueLyser II; QIAGEN, Germany).

Reverse transcription (RT)was performed using Superscript II RNaseH− and Oligo dT system (Invitrogen, Carlsbad, CA, USA) as previouslydescribed (Pilegaard et al., 2000), and the cDNA samples were dilutedin nuclease-free H2O.

2.3.4. Real-time PCRReal-timePCRwas performed using anABI 7900 sequence-detection

system (Applied Biosystems, Foster City, CA, USA) as previously de-scribed (Lundby et al., 2005). Primers and Taqman probes wereobtained from TAG Copenhagen (Copenhagen, Denmark) (Table 1).Real-time PCR was performed in triplicates in a total reaction volumeof 10 μl using Universal Mastermix (Applied Biosystems). Cycle thresh-old was converted to a relative amount by use of a standard curveconstructed from a serial dilution of a pooled RT sample run togetherwith the samples. Target gene mRNA content was for each samplenormalized to single-stranded cDNA content determined by OliGreenreagent (Molecular Probes, Leiden, The Netherlands) as previouslydescribed (Lundby et al., 2005).

2.3.5. LysateCrushed quadriceps muscles (~20–25 mg), liver (~20–25 mg) and

adipose tissue (~25–35 mg) were homogenized for 2 min at 30 s−1

in a tissue lyser (TissueLyser II; QIAGEN) in an ice-cold buffer as previ-ously described (Birk and Wojtaszewski, 2006). Protein content inlysates was measured by the bicinchoninic acid method (ThermoScientific, Rockford, IL, USA).

2.3.6. SDS-PAGE and western blottingProtein content and phosphorylation of various proteins were mea-

sured in lysates by SDS-PAGE and western blotting as previously de-scribed (Birk and Wojtaszewski, 2006). Band intensity was quantifiedusing Carestream IS 4000 MM (Fisher Scientific, ThermoFisher Scientif-ic, Waltman, MA, USA) and Carestream health molecular imaging soft-ware. Protein content and phosphorylation were expressed asarbitrary units relative to control samples loaded on each site of each

UNCO

R

Table 1Tumor necrosis factor (TNF)α, interleukin (IL)-6 and F4/80 primer and TaqMan probe sequenc

Gene

TNFα

IL-6

F4/80

TACE

SOD2

Catalase

GPX1

Forward primer Reverse prim

5’ ATGGCCCAGACCCTCACA 3’

5’ GCTTAATTACACATGTTCTCTGGGAAA’ 3

5’ GGCTGCCTCCCTGACTTTC 3’

5’ TGCAAGGCTGGGAAATGC 3’

5’ GTGGTGGAGAACCCAAAGGA 3’

5’ CTGGACGTTTTACATCCAGGTCA 3’

5’ GACTGGTGGTGCGGTTTC 3’

5’ TTGCTACGACGTG

5’ CAAGTGCATCATC

5’ TGCACTGCTTGG

5’ TTGCACGAGTTGT

5’ AACCTTGGACTCC

5’ TCCTTGTGAGGCC

5’ TTGAGGGAATTC


ED P

RO

OF

gel and normalized to glyceraldehyde 3-phosphate dehydrogenase(GAPDH) protein. Commercially available antibodies were used to de-tect TNFα (#3707), nitric oxide synthase (iNOS) (#2977), nuclear factorof kappa light polypeptide gene enhancer in B-cells (NFκB) inhibitor(IκB)-α (#9242), IκB-β (#9248), p65 (#4764) and GAPDH (#2118)protein as well as p65ser536 (#3033), c-Jun N-terminal kinases (JNK)(#9252), JNKThr183, Tyr185 (#9251), p38 mitogen-activated protein ki-nase (p38) (#9212) and p38Thr180, Tyr182 (# 4511) phosphorylation allfrom Cell Signaling and superoxide dismutase (SOD)2 (#06-984,Millipore), catalase (#SC-50508) and glutathione peroxidase (GPX)1(#SC-30147) protein from Santa Cruz Biotechnology inc.

2.3.7. Protein carbonylationProtein carbonyl content was determined in quadriceps muscle

samples homogenized in phosphate-buffer using an OxiSelect™ELISA-kit (Cell Biolabs, San Diego, USA) according to themanufacturer'sprotocol. Absorbance was measured at 450 nm and an oxidized/re-duced BSA standard curve was generated to determine the concentra-tion of protein carbonyl in each sample.

2.4. Statistics

Results are presented asmeans ± S.E. Unless otherwise noted, two-way analysis of variance (ANOVA)was applied to test themain effects ofgenotype and interventions and one-way ANOVA was used to test fordifferences between the interventions separately within each genotype.If either the equal variance test or the normality test failed, the datawere logarithmically transformed before applying the ANOVA test. Stu-dent Newman Keuls post hoc test was used to locate differences whenapplicable. The non-parametric Mann–Whitney U test was appliedwhen the equal variance test or the normality test failed even after log-arithmic transformation. A p b 0.05 was considered significant and atendency is reported for 0.05 ≤ p b 0.1.

3. Results

3.1. Body weight, food intake and body composition

Bodyweightswere initially similarwithin theWTgroups andwithinthe PGC-1α KO groups (data not shown). Total body weight increased(p b 0.05) ~20% with age in WT and PGC-1α KO mice (Fig. 1a). Therewas no effect of exercise training and/or RSV on body weight in either

es used for real time PCR.

er Probe

GGCTACA 3’

GTTGTTCATAC’ 3

CATTGC 3’

CAGTGTCAA 3’

CACAGACA 3’

AAACCTT 3’

AGAATCTCTTCA 3’

5’ TCAGATCATCTTCTCAAAATTCGAGTGACAAGC 3’

5’ ATCAGAATTGCCATTGCACAACTCTTTTCTCAT’ 3

5’ TCCTTTTGCAGTTGAAGTTTCCATATCCTTGG 3’

5’ AGCAGGAGCTGGAGTCCTGCGC 3’

5’ AGTTGCTGGAGGCTATCAAGCGTGACTTT 3’

5’ AGGCAGAAACTTTCCCATTTAATCCATTTGATC 3’

5’ AATCAGTTCGGACACCAGGAGAATGGC 3’



CTED P

RO

OF

283

284

285

286

287

288

Fig. 1. Body weight (a), food intake (b), lean body mass in % (c), fat % (d), perigonadal visceral adipose tissue (V-AT) weight (e) and inguinal subcutaneous adipose tissue (S-AT) weightfrom 3 month old (mo) untrained chow mice (UT-C), 15 month old untrained chow mice (UT-C), 15 month old untrained resveratrol (RSV) supplemented mice (UT-R), 15 month oldexercise trained (T-C) and 15 month old exercise trained and RSV supplemented (T-R) whole body PGC-1α knockout (KO) and littermate wild type (WT) mice. Values are presentedas means ± S.E.; n = 8–10 except in (c) and (d) where n = 3–8. †: significantly different from 3 month old UT-C within genotype, p b 0.05. *: significantly different from 15 monthold UT-C within genotype, p b 0.05. (*): tends to be significantly different from 15 month old UT-C within genotype, 0.05 ≤ p b 0.1. #: significantly different from WT within group,p b 0.05. (#) tends to be significantly different from WT within group, 0.05 ≤ p b 0.1.


RE

WT or PGC-1α KOmice. PGC-1α KOmiceweighed ~10% less (p b 0.05)than WT mice in all groups. There was no difference in food intake(given pr. gram of mice) between any of the groups (Fig. 1b).

UNCO

R

Fig. 2. Plasma tumor necrosis factor (TNF)α (a) and plasma interleukin (IL)-6 (b) from 3 mo15 month old untrained resveratrol (RSV) supplemented mice (UT-R), 15 month old exercisbody PGC-1α knockout (KO) and littermate wild type (WT) mice. Values are presented as mlogarithmically transformed data when appropriate. †: significantly different from 3 month oldgenotype, p b 0.05. #: significantly different from WT within group, p b 0.05.


Lean bodymass (given as %) decreased (p b 0.05) 5–10%with age inall groups (Fig. 1c). There was no effect of exercise training and/or RSVon percentage lean bodymass in either WT or PGC-1α KOmice. Fifteen

nth old (mo) untrained chow mice (UT-C), 15 month old untrained chow mice (UT-C),e trained (T-C) and 15 month old exercise trained and RSV supplemented (T-R) wholeeans ± S.E.; n = 8–10. Two-way ANOVA and one-way ANOVA tests were applied onUT-C within genotype, p b 0.05. *: significantly different from 15 month old UT-C within



289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

t2:1

t2:2

t2:3

t2:4

t2:5

t2:6

t2:7


month old untrained and exercise trained PGC-1α KO mice tended tohave 3–4% higher (p = 0.053 and p = 0.051, respectively) percentagelean body mass than WT mice, while no genotype differences wereobserved in the other groups.

Whole body fat percentage as well as V-AT and S-ATmass increased(p b 0.05) 1.6–2.1 fold with age in both WT and PGC-1α KO mice(Fig. 1d, e, f). While no effect was observed on adiposity by RSV supple-mentation alone, exercise training decreased (p b 0.05) fat percentage~30% and tended to decrease (p = 0.082) V-AT mass in WT mice.Combined exercise training and RSV also decreased (p b 0.05) fatpercentage (23%) as well as V-AT (37%) and S-AT (27%) mass in WTmice. Fat percentage was ~30% lower (p b 0.05) in PGC-1α KO thanWT within 15 month old untrained and RSV supplemented mice, andS-AT mass was 30% lower (p b 0.05) in PGC-1α KO mice than WTwithin 15 month old untrained mice. In addition, PGC-1α KO micehad in all groups 25–40% less (p b 0.05) V-AT than WT mice.

338

339

340

341

342

3.2. Plasma

Selected plasma cytokines were analyzed to evaluate the generalsystemic inflammatory status associatedwith the different interventions.

343

344

345

346

347

348

349

350

351

3.2.1. Plasma TNFαTherewas anoverall 1.7–2.0 fold increase (p b 0.05) in plasmaTNFα

with age in WT and PGC-1α KO mice (Fig. 2a). No significant effectof RSV, exercise training alone or exercise training in combinationwith RSV was observed on plasma TNFα. The plasma TNFα level was1.7–2.7 fold higher (p b 0.05) in PGC-1α KO than WT in young and15 month old untrained mice.

T

352

353

354

355

356

357

358

359

360

361

REC

3.2.2. Plasma IL-6Plasma IL-6 increased 2–2.3 fold (p b 0.05)with age inWT and PGC-

1α KO mice (Fig. 2b). Exercise training alone as well as in combinationwith RSV reduced (p b 0.05) the plasma IL-6 level 40–60% in WT andPGC-1α KO mice compared with 15 month old untrained mice. InPGC-1α KO mice, RSV supplementation reduced (p b 0.05) the plasmaIL-6 level ~60% compared with 15 month old untrained mice. Inyoung and 15 month old untrained mice, the plasma IL-6 level was1.6–1.8 fold higher (p b 0.05) in PGC-1α KOmice than that inWTmice.

UNCO

R

Table 2Tumor necrosis factor (TNF)α, interleukin (IL)-6 and F4/80 mRNA in perigonadal visceral adipomice (UT-C), 15 month old untrained chow mice (UT-C), 15 month old untrained resveratrolexercise trained and RSV supplemented (T-R) whole body PGC-1α knockout (KO) and litterANOVA and one-way ANOVA tests were applied on logarithmically transformed data whesignificantly different from 3 month old UT-C within genotype, p b 0.05. (†): tends to be signidifferent from 15 month old UT-C within genotype, p b 0.05. #: significantly different from W

V-AT

Liver

SkM

Gene

TNFα

TNFα

IL-6

IL-6

F4/80

F4/80

TNFα

IL-6

F4/80

3 mo UT-C 15 mo UT-C

WT

0.5 ± 0.1

0.8 ± 0.2

0.8 ± 0.2

0.3 ± 0.03

0.2 ± 0.02

0.6 ± 0.04

1.1 ± 0.1

1.7 ± 0.1

2.4 ± 0.1 2.3 ± 0.3

0.5 ± 0.1

0.7 ± 0.1

0.5 ± 0.1

0.2 ± 0.01

0.3 ± 0.04

0.5 ± 0.2

0.6 ± 0.1

0.8 ± 0.2 2.3 ± 0.7† 0.7 ± 0.2# 0.9 ± 0

0.6 ± 0

0.5 ± 0

0.1 ± 0

0.5 ± 0

1.2 ± 0

0.7 ± 0

0.8 ± 0

0.9 ± 00.4 ± 0.1

0.7 ± 0.3

0.3 ± 0.03#

0.2 ± 0.02

0.4 ± 0.1#

1.6 ± 0.3†

1.0 ± 0.1†

0.6 ± 0.1

1.2 ± 0.3

0.6 ± 0.1†

0.2 ± 0.01

0.7 ± 0.1

1.4 ± 0.2

1.6 ± 0.3

1.0 ± 0.1(†)

1.2 ± 0.2†

KO WT KO W


ED P

RO

OF

3.3. Inflammatory mRNA markers in V-AT, liver and SkM

The TNFα and IL-6 mRNA content was determined in V-AT, liverand SkM in order to evaluate the tissue specific inflammatory status.Additionally, the mRNA content of the macrophage specific F4/80(Khazen et al., 2005) was analyzed as a marker of macrophage infiltra-tion in these tissues.

3.3.1. V-ATIn WT mice, V-AT TNFα mRNA increased (p b 0.05) 4.4 fold with

age, but mice supplemented with RSV, exercise trained as well asexercise trained combined with RSV supplementation did not differfrom young control mice in V-AT TNFα mRNA content. In PGC-1α KOmice, RSV increased (p b 0.05) V-AT TNFα mRNA 1.8 fold comparedwith 15 month old untrained mice (Table 2).

In WT mice, both RSV and exercise training increased (p b 0.05) V-AT IL-6 mRNA 1.5–1.9 fold compared with 15 month old untrainedmice. In PGC-1α KO mice, RSV increased (p b 0.05) V-AT IL-6 mRNAcontent 1.5 fold compared with 15 month old untrained mice (Table 2).

No differences were observed in V-AT F4/80 mRNA between geno-types or any of the interventions (Table 2).

3.3.2. LiverLiver TNFα mRNA increased 1.6 fold (p b 0.05) with age only in

WT mice. Liver TNFα mRNA increased (p b 0.05) in PGC-1α KO micewith RSV and combined exercise training and RSV, while no changeoccurred in WT mice. Fifteen month old PGC-1α KO mice had lower(p b 0.05) liver TNFα mRNA content than WT, whereas no differencewas observed between genotypes in the other groups (Table 2).

No differenceswere observed in liver IL-6mRNAbetween genotypesor any of the interventions (Table 2.)

No effect of age or any of the interventions was observed in F4/80mRNA content in the liver. However, 15 month old untrained PGC-1αKO mice had 47% lower (p b 0.05) F4/80 mRNA content in the liverthan WT mice (Table 2).

3.3.3. SkMTNFαmRNA increased (p b 0.05) ~2.1 fold in SkMwith age in PGC-

1α KO mice. TNFα mRNA content was 1.8 fold higher (p b 0.05) inexercise trained WT mice than 15 month old untrained WT mice,whereas no effect was observed with RSV and combined exercisetraining and RSV.

se tissue (V-AT), liver and skeletal muscle (SkM) from 3 months old (mo) untrained chow(RSV) supplemented mice (UT-R), 15 month old exercise trained (T-C) and 15 month oldmate wild type (WT) mice. Values are presented as means ± S.E.; n = 8–10. Two-wayn appropriate. Mann–Whitney U nonparametric test was applied when appropriate †:ficantly different from 3 month old UT-C within genotype, 0.05 ≤ p b 0.1. *: significantlyT within group, p b 0.05.

15 mo UT-C15 mo UT-R 15 mo UT-R

.2

.3

.1

.01

.1

.3

.2

.2

.1*

1.3 ± 0.3*

0.8 ± 0.1*

0.7 ± 0.1

1.0 ± 0.3

1.2 ± 0.1*

1.3 ± 0.2

0.5 ± 0.1

0.2 ± 0.02

0.6 ± 0.1

2.6 ± 0.6*

0.7 ± 0.1*

3.7 ± 1.1* 3.0 ± 0.4*

0.7 ± 0.1*

1.9 ± 0.2*

0.5 ± 0.1

0.2 ± 0.2

0.4 ± 0.4

0.7 ± 0.3

0.7 ± 0.2#

0.7 ± 0.2 0.9 ± 0.1 1.4 ± 0.4

0.5 ± 0.1

0.9 ± 0.3

0.5 ± 0.1*

0.2 ± 0.03

0.7 ± 0.1

2.4 ± 0.4

0.7 ± 0.1*

3.5 ± 0.4#*

0.7 ± 0.1

0.7 ± 0.2

0.5 ± 0.04

0.2 ± 0.02

0.5 ± 0.03

1.9 ± 0.3

0.9 ± 0.1*

2.0 ± 0.3*

0.5 ± 0.1*

0.2 ± 0.03

0.7 ± 0.1

1.6 ± 0.2

1.0 ± 0.1

1.6 ± 0.3#

T KO WT KO WT KO



UNCO

RRECTED P

RO

OF

Fig. 3. Tumor necrosis factor (TNF)α protein in perigonadal visceral adipose tissue (V-AT) (a), liver (b) and quadriceps muscle (c) from 3 month old (mo) untrained chow mice (UT-C),15 month old untrained chowmice (UT-C), 15 month old untrained resveratrol (RSV) supplementedmice (UT-R), 15 month old exercise trained (T-C) and 15 month old exercise trainedand RSV supplemented (T-R) whole body PGC-1α knockout (KO) and littermate wild type (WT) mice. Values are presented as means ± S.E.; n = 8–10. †: significantly different from3 month old UT-C within genotype, p b 0.05. (†): tends to be significantly different from 3 month old UT-C, 0.05 ≤ p b 0.1. *: Significantly different from 15 month old UT-C within ge-notype, p b 0.05. #: significantly different fromWTwithin group, p b 0.05. A horizontal line indicates a main effect. Representative blots are shown on each figurewith samples loaded inthe same order as depicted on the graph.

Fig. 4. Nuclear factor of kappa light polypeptide gene enhancer in B-cells (NFκB) inhibitor (IκB)-α protein (a), p65 phosphorylation (phos) (b), c-Jun N-terminal kinase (JNK) phosphor-ylation (c) andp38mitogen-activated protein kinase (p38) phosphorylation (d) in quadricepsmuscle from3 month old (mo) untrained chowmice (UT-C), 15 month old untrained chowmice (UT-C), 15 month old untrained resveratrol (RSV) supplementedmice (UT-R), 15 month old exercise trained (T-C) and 15 month old exercise trained and RSV supplemented (T-R)whole body PGC-1α knockout (KO) and littermate wild type (WT) mice. Values are presented as means ± S.E.; n = 8–10. †: significantly different from 3 month old UT-C within geno-type, p b 0.05. Representative blots are shown on each figure with samples loaded in the same order as depicted on the graph.


Please cite this article as: Olesen, J., et al., Role of PGC-1α in exercise training- and resveratrol-induced prevention of age-associated inflammation,Exp. Gerontol. (2013), http://dx.doi.org/10.1016/j.exger.2013.07.015


T

RO

OF

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

Fig. 5. Protein carbonylation (a) and iNOSprotein (b) in quadricepsmuscle from3 month old (mo)untrained chowmice (UT-C), 15 month old untrained chowmice (UT-C), 15 month olduntrained resveratrol (RSV) supplementedmice (UT-R), 15 month old exercise trained (T-C) and 15 month old exercise trained and RSV supplemented (T-R) whole body PGC-1α knock-out (KO) and littermatewild type (WT)mice. Values are presented as means ± S.E.; n = 8–10. †: significantly different from 3 month old UT-Cwithin genotype, p b 0.05. *: significantlydifferent from15 month old UT-Cwithin genotype, p b 0.05. #: significantly different fromWTwithin group, p b 0.05. Representative blots are shown on eachfigurewith samples loadedin the same order as depicted on the graph.


EC

IL-6 mRNA increased ~2.3 fold (p b 0.05) with age in PGC-1α KOmice and tended to increase 1.4 fold (p = 0.06) with age in WT mice.Exercise training combined with RSV decreased (p b 0.05) SkM IL-6mRNA content 10–37% in WT and PGC-1α KO mice, while no effectwas observed in SkM IL-6 mRNA with RSV alone or exercise trainingalone (Table 2).

SkM F4/80 mRNA content decreased (p b 0.05) 50% with age inWT mice, whereas exercise training alone (2.5 fold) or in combinationwith RSV (1.9 fold) increased (p b 0.05) the F4/80 mRNA content inSkM of both WT and KO mice compared with 15 month old untrainedmice (Table 2). In the RSV group and in the combined exercise trainedand RSV group, PGC-1α KO mice had 1.8–2.1 fold higher (p b 0.05)F4/80 mRNA level in SkM than WT mice, whereas no differences werepresent between genotypes in the other groups (Table 2).

UNCO

RR

Fig. 6. Superoxide dismustase (SOD)2 protein (a), catalase protein (b) and glutathione peroxida(UT-C), 15 month old untrained chowmice (UT-C), 15 month old untrained resveratrol (RSV) strained and RSV supplemented (T-R) whole body PGC-1α knockout (KO) and littermate wild tfrom 3 month old UT-C within genotype, p b 0.05. *: significantly different from 15 month oldUT-C within genotype, 0.05 ≤ p b 0.1. #: significantly different fromWTwithin group, p b 0.05depicted on the graph.


ED P3.4. TNFα protein in V-AT, liver and SkM

TNFα protein content was further determined in V-AT, liver andSkM. No differences were observed in V-AT TNFα protein contentbetween any of the interventions or genotypes (Fig. 3a). An overalltendency (p = 0.079) for a 1.2–1.3 fold increase in liver TNFα proteinwas observed with age in WT and PGC-1α KO mice (Fig. 3b). No effectof exercise training and/or RSV was observed in liver TNFα protein,but these groups did not differ from young mice either. SkM TNFαprotein content increased 1.4–1.5 fold (p b 0.05) with age in WT andPGC-1α KO mice. Both exercise training alone and exercise training incombination with RSV prevented this age-associated increase in TNFαprotein in WT mice, but this response was blunted in PGC-1α KO mice(Fig. 3c). PGC-1α KO mice had 1.5–2.2 fold higher (p b 0.05) SkM

se (GPX)1 protein (c) in quadricepsmuscle from 3 month old (mo) untrained chowmiceupplementedmice (UT-R), 15 month old exercise trained (T-C) and 15 month old exerciseype (WT) mice. Values are presented as means ± S.E.; n = 8–10. †: significantly differentUT-C within genotype, p b 0.05. (*): tends to be significantly different from 15 month old. Representative blots are shown on each figurewith samples loaded in the same order as



T

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512


UNCO

RREC

TNFα protein content than WT mice in RSV, exercise trained as well ascombined exercise trained and RSV groups (Fig. 3c).

3.5. Inflammatory signaling in skeletal muscle

The 3 major inflammatory signaling pathways, IKK/NfκB, JNK andp38 were analyzed in SkM in order to investigate the underlyingmech-anisms behind the observed differences in SkM TNFα protein levels.

No difference was observed in total p65, JNK and p38 proteincontent between genotypes or interventions (data not shown). InWT mice, IκB-α protein decreased 30% (p b 0.05) in SkM with age(Fig. 4a). No differences were observed in SkM IκB-α protein contentwith exercise training and/or RSV supplementation in any of thegenotypes. SkM p65 phosphorylation did not differ between the inter-ventions or the genotypes (Fig. 4b). There were no differences inJNK (Fig. 4c) or p38 (Fig. 4d) signaling in SkM in any of the interventionsor between the genotypes.

3.6. Protein carbonylation and iNOS protein in skeletal muscle

As oxidative stress has been suggested to be a stimulus for inducinginflammation, SkM protein carbonylation (amarker of oxidative stress)was determined. SkM protein carbonylation increased 1.6–2.1 fold(p b 0.05) with age in both WT and PGC-1α KO mice (Fig. 5a). RSV,exercise training and combined exercise training and RSV prevented(p b 0.05) this age-associated increase in SkM protein carbonylationin both WT and PGC-1α KO mice. Furthermore, young PGC-1α KOmice had 1.8 fold higher (p b 0.05) protein carbonylation in SkM thanyoung WT mice (Fig. 5a).

In accordance, the protein content of iNOS increased (p b 0.05) ~1.4fold with age in SkM of WT and PGC-1α KO mice (Fig. 5b). Exercisetraining and/or RSV supplementation did not change iNOS proteincontent relative to 15 month old untrained mice, but iNOS proteincontent was in these groups not different from youngmice. Interesting-ly, PGC-1αKOmicehad in all groups 1.2–1.8 fold higher (p b 0.05) iNOSprotein content than WT mice (Fig. 5b).

3.7. Anti-oxidant enzymes in skeletal muscle

The anti-oxidant enzymes SOD2, catalase and GPX1 were analyzedin SkM to indirectly examine whether the observed indications ofoxidative stress in SkM were related to ROS neutralization capacity.Catalase protein content decreased (p b 0.05) ~50% with age in SkMof WT mice (Fig. 6b). RSV increased (p b 0.05) SkM GPX1 protein 1.8fold, whereas no effect of RSVwas observed in SOD2 or catalase protein.Exercise training increased (p b 0.05) GPX1 protein 1.8 fold (Fig. 6c)and tended to increase SOD2 protein (p = 0.051) and catalase protein(p = 0.075) 1.3–2.2 fold in SkM of WT mice. In WT mice, combinedexercise training and RSV increased (p b 0.05) catalase protein andGPX1 protein 2.2–2.5 fold compared with 15 month old untrainedmice, but with no difference between the exercise trained group andthe combined exercise trained and RSVgroup. No effect of RSV, exercisetraining or combined exercise training and RSV was observed in SOD2,catalase or GPX1 protein in PGC-1α KO mice.

PGC-1α KO mice had in all groups 25–50% lower (p b 0.05) SkMSOD2 protein content than WT mice. In contrast, 15 month olduntrained PGC-1α KO mice had ~2.0 fold higher (p b 0.05) GPX1and catalase protein content than WT and GPX1 protein increased(p b 0.05) 1.6 fold with age in SkM of PGC-1α KO mice.

4. Discussion

The main findings of the present study were that long-termexercise training prevented an age-associated increase in TNFαprotein in SkM in a PGC-1α dependent manner. Independently ofPGC-1α, long-term exercise training prevented an age-associated


ED P

RO

OF

increase in systemic IL-6 levels and oxidative stress in SkM. Long-term RSV supplementation also prevented an age-associated in-crease in systemic IL-6 and oxidative stress in SkM independentlyof PGC-1α. Taken together, the beneficial effects of long-term RSVsupplementation seemed minor compared with previous reportson short-term RSV supplementation and compared with the presenteffects of long-term exercise training.

The present increase in adiposity and total body weight withaging is in accordance with previous studies (Guo et al., 1999; Wuet al., 2007). Although a genotype specific difference in body weightand adiposity between PGC-1α KO andWTmice has previously beendescribed (Lin et al., 2004), the current finding that exercise trainingas well as exercise training combined with RSV supplementationonly prevented the age-associated adiposity in WT mice is of notice.The age-associated increase in plasma TNFα and IL-6 with aging inthe present study is also in agreement with previous studies (Weiet al., 1992) and supports that increased adiposity is associatedwith systemic low-grade inflammation. The observed prevention ofan age-associated elevation in plasma IL-6 with long-term exercisetraining is in line with previous studies indicating that exercisetraining has anti-inflammatory effects (Olholm et al., 2010; Woodset al., 2012). However, it should be noted that the present age-associated low-grade inflammation and improvements with exercisetraining were not linked to altered glucose tolerance (Ringholmet al., in review Exp Gerontol), which may suggest that low-gradeinflammation precedes glucose intolerance at least in the presentsetting. A potential role of PGC-1α in the regulation of systemicinflammation has previously been suggested based on studies inboth muscle-specific PGC-1α KO mice (Handschin et al., 2007a,2007b) and muscle-specific PGC-1α overexpression mice (Olesenet al., 2012; Wenz et al., 2009). The proposal that lack of PGC-1αleads to low-grade inflammation is supported by the present find-ings that young and aged PGC-1α KO mice had increased plasmaTNFα and IL-6 compared with WT mice.

The present findings that aging was associated with increasedTNFα mRNA in V-AT and TNFα mRNA and protein in liver as wellas increased iNOS and TNFα protein in SkM are in line with previousstudies (Lumeng et al., 2011; Wu et al., 2007) and with the presentage-associated increase in plasma TNFα. Together, these findingsstrongly indicate that several tissues are inflamed with aging andpotentially contribute to the observed increase in the systemic levelsof inflammatory mediators. However, the novel finding that long-term exercise training prevented the age-associated increase inTNFα protein in SkM supports that exercise training specificallyelicits anti-inflammatory responses in SkM. Furthermore, the almostsimilar patterns of plasma TNFα and SkM TNFα protein in responseto aging and exercise training indicate that SkM can be an importantcontributor to the systemic plasma TNFα levels as previously sug-gested during acute inflammation (Borge et al., 2009; Frost et al.,2002; Olesen et al., 2012). The current demonstration that the reduc-tion in SkM TNFα protein with exercise training was completelyblunted in PGC-1α KO mice provides evidence that PGC-1α isrequired for the exercise training-induced anti-inflammatory effectsin SkM. In addition, the novel observation that PGC-1α KO mice hadhigher iNOS protein levels than WT mice adds to the previouslyreported anti-inflammatory effect of muscle PGC-1α (Handschinet al., 2007b; Wenz et al., 2009).

To delineate the potential underlying mechanisms behind theobserved TNFα protein expression pattern in SkM, several intracel-lular inflammatory signaling pathways were investigated in SkM.The observed age-associated decrease in SkM IκB-α protein is inaccordance with previous findings (Wenz et al., 2009) and may indi-rectly reflect increased translocation of NFκB to the nucleus, andthus potentially explain the observed age-associated increase inTNFα protein inWTmice. In contrast, NFκB, JNK or p38 signaling nei-ther explains the observed exercise training-induced nor genotype



T

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638Q2

639


UNCO

RREC

specific differences in SkM TNFα protein. To our knowledge noprevious studies have shown PGC-1α dependent alterations in p38or JNK signaling, but divergent results have been reported for NFκBsignaling (Alvarez-Guardia et al., 2010; Eisele et al., 2013; Olesenet al., 2012; Wenz et al., 2009). Hence, previous studies have bothreported that 24 month old muscle-specific PGC-1α overexpressionmice had decreased SkM p65 phosphorylation (Wenz et al., 2009)and that young muscle-specific PGC-1α overexpression mice hadelevated SkM p65 phosphorylation compared with WT (Olesenet al., 2012). Together with the present findings this may indicatethat the interaction between PGC-1α and NFκB depends on thespecific experimental setting.

As macrophages are considered an important source of TNFα andiNOS production, the mRNA content of the highly specific murinemacrophage marker, F4/80 (Khazen et al., 2005), was determined toevaluate whether differences in macrophage infiltration could explainthe observed differences in SkM TNFα and/or iNOS protein content.The finding that SkM F4/80 mRNA content decreased with age andincreased with long-term exercise training indicates that macrophageswithin SkM or in the surrounding capillaries were not responsible forthe observed differences in TNFα and iNOS protein in SkM. Althoughthe present data cannot exclude the possibility that infiltrated immunecells have contributed to the observed TNFα and iNOS, these observa-tions imply that SkM fibers may indeed be a considerable source ofTNFα (Borge et al., 2009; Frost et al., 2002; Olesen et al., 2012) as wellas iNOS production. Taken together, the observed signaling events inSkM in concert with the F4/80 mRNA levels are inadequate to explainthe present TNFα and iNOS protein expression patterns. Althoughspeculative this may indicate that the observed TNFα differences arenot due to changes in synthesis, but rather degradation/turnover.However, future studies are needed to fully understand the underlyingmechanism.

The observed age-associated increase in protein carbonylation inSkM is in accordance with previous studies and supports an associationbetween adiposity, inflammation and oxidative stress (Berg andScherer, 2005; Hotamisligil et al., 1995). In addition, the increase inprotein carbonylation with age may potentially be explained by theobserved age-associated decrease in the ROS scavenging protein cata-lase. Interestingly, exercise training alone and in combination withRSV prevented the age-associated increase in protein carbonylation,which may be due to the observed exercise training and combined ex-ercise training- and RSV-induced increases in anti-oxidant enzymes.These data support that both exercise training and RSV increase theanti-oxidant capacity (Hellsten et al., 1996; Jackson et al., 2011; Leicket al., 2010). However, the observation that RSV supplementationcounteracted the age-associated oxidative stress and at the same timeonly affected GPX1 protein levelsmay suggest that the RSV-mediated de-crease in oxidative stress in part was due to its direct anti-oxidant prop-erties as previously suggested (Howitz et al., 2003; Wood et al., 2004).

Notably, the present finding that young PGC-1α KO mice hadincreased protein carbonylation in conjunction with reduced SOD2levels compared with WT mice indicates that PGC-1α is importantfor the basal ROS handling in young mice (Leick et al., 2010; St-Pierre et al., 2006). Furthermore these findings support that the in-creased oxidative stress in these animals in part is due to reducedROS neutralization capacity as previously reported (St-Pierre et al.,2006). Moreover, while PGC-1α was not required for the observedexercise training and RSV-induced prevention of age-associatedoxidative stress, PGC-1α was required for the exercise training-induced increase in SOD2 protein content and in part also for theexercise training- and combined exercise training and RSV-inducedincrease in catalase and GPX1 protein content. Together these datasupport that changes in the endogenous anti-oxidant system con-tribute to the age-associated increase in oxidative stress as well asthe exercise training-induced prevention of oxidative stress. Howev-er, the precise role of PGC-1α in this is still not clarified.


ED P

RO

OF

A key finding of the present study is that long-term RSV supple-mentation only showed minor effects, which is in contrast to manyprevious studies in mice (Baur et al., 2006; Lagouge et al., 2006;Um et al., 2010). However, a recent study in mice (Menzies et al.,2013) and two independent human studies in obese men and non-obese women (Poulsen et al., 2012; Yoshino et al., 2012) also failedto show any major impact of RSV supplementation. A study in cul-tured human primary muscle cells even showed impaired glucoseuptake after incubation with RSV (Skrobuk et al., 2012). This under-lines the importance of additional studies to fully understand themetabolic effects of RSV. Moreover, the observed additive effect ofexercise training and RSV on V-AT and S-AT mass in the presentstudy is modest compared with previous studies in rodents examin-ing the additive effects of exercise training and RSV (Dolinsky et al.,2012; Menzies et al., 2013). The relatively small effects of RSV inthe present study do not seem to be dose related. The present doseused corresponds to ~0.7 mg RSV per gram mouse per day, whichis similar to previous studies where profound metabolic effectshave been observed (Lagouge et al., 2006; Um et al., 2010). Thelack of RSV-mediated effects may alternatively be related to theduration of the treatment. Hence, while previous studies havefocused on shorter durations (1–4 month), the 12 month treatmentin the present study may have had a desensitizing effect. Anotherlikely explanation may be that the 15 month old “control” animals,despite the observed age-associated deteriorations already discussed,were too metabolically “healthy” to obtain metabolic improvementsthrough the RSV treatment. Hence, in contrast to several previousreports on RSV supplementation (Baur et al., 2006; Lagouge et al.,2006; Um et al., 2010), the old mice from the present study did nothave impaired glucose tolerance (Ringholm et al., in review, Exp.Gerontology) as already mentioned. The novel finding that PGC-1αwas not required for the RSV-induced improvements in systemic IL-6and SkM oxidative stress may indicate that PGC-1α is activated byRSV as suggested (Baur et al., 2006; Lagouge et al., 2006; Timmerset al., 2011), but not mandatory for the metabolic effects of RSV.

In conclusion, the present findings demonstrate that long-termexercise training prevented an age-associated increase in adiposity,systemic low-grade inflammation as well as SkM oxidative stress. Inaddition, these results show that PGC-1α was required for an exercisetraining-induced prevention of an age-associated increase in SkMTNFα protein. Long-term RSV supplementation elicited only few effectson SkM oxidative stress and in part on low-grade inflammation andPGC-1α was not required for these effects. Together, the presentfindings indicate that regular physical activity is a more powerful inter-vention to prevent or postpone age-related inflammation than RSVsupplementation.

5. Funding

This present studywas supported by grants from theDanishMedicalResearch Council (#11-104199) to HP; Novo Nordisk Foundation to HP;Augustinus Foundation to JO; the National Institutes of Health (R01AR42238) to LJG. The Centre of Inflammation and Metabolism (CIM) issupported by a grant from the Danish National Research Foundation(# 02-512-55). CIM is part of the UNIK Project: Food, Fitness & Pharmafor Health and Disease, supported by the Danish Ministry of Science,Technology, and Innovation. CIM is a member of DD2 — the DanishCenter for Strategic Research in Type 2 Diabetes (the Danish Councilfor Strategic Research, grant no. 09-067009 and 09-075724). Thefunders had no role in study design, data collection and analysis,decision to publish, or preparation of the manuscript.

Conflict of interest

The authors have no conflicts of interests.



640

641

642

643

644

645

646647648649650651652653654655656657658Q3659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718

719720721722723724725726


Acknowledgments

We would like to thank Professor Bruce M. Spiegelman (HarvardMedical School, Boston, Massachusetts, USA) for the kind donation ofbreeding pairs to the initial breeding of the whole body PGC-1α knock-out and littermate wild type mice.

T

727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770Q4771772773774775776777778779780781Q5782783784785786787788Q6789790791792793794795796797798799800801802803804

UNCO

RREC

References

Alibegovic, A.C., Sonne, M.P., Hojbjerre, L., Bork-Jensen, J., Jacobsen, S., Nilsson, E., Faerch,K., Hiscock, N., Mortensen, B., Friedrichsen, M., Stallknecht, B., Dela, F., Vaag, A., 2010.Insulin resistance induced by physical inactivity is associatedwithmultiple transcrip-tional changes in skeletal muscle in young men. Am. J. Physiol. Endocrinol. Metab.299, E752–E763.

Alvarez-Guardia, D., Palomer, X., Coll, T., Davidson, M.M., Chan, T.O., Feldman, A.M.,Laguna, J.C., Vazquez-Carrera, M., 2010. The p65 subunit of NF-kappaB binds toPGC-1alpha, linking inflammation and metabolic disturbances in cardiac cells.Cardiovasc. Res. 87, 449–458.

Baar, K., Wende, A.R., Jones, T.E., Marison, M., Nolte, L.A., Chen, M., Kelly, D.P., Holloszy,J.O., 2002. Adaptations of skeletal muscle to exercise: rapid increase in the transcrip-tional coactivator PGC-1. FASEB J. 16, 1879–1886.

Baur, J.A., Pearson, K.J., Price, N.L., Jamieson, H.A., Lerin, C., Kalra, A., Prabhu, V.V.,Allard, J.S., Lopez-Lluch, G., Lewis, K., Pistell, P.J., Poosala, S., Becker, K.G., Boss,O., Gwinn, D., Wang, M., Ramaswamy, S., Fishbein, K.W., Spencer, R.G., Lakatta,E.G., Le, C.D., Shaw, R.J., Navas, P., Puigserver, P., Ingram, D.K., de, C.R., Sinclair,D.A., 2006. Resveratrol improves health and survival of mice on a high-caloriediet. Nature 444, 337–342.

Berg, A.H., Scherer, P.E., 2005. Adipose tissue, inflammation, and cardiovascular disease.Circ. Res. 96, 939–949.

Birk, J.B., Wojtaszewski, J.F., 2006. Predominant alpha2/beta2/gamma3 AMPK activationduring exercise in human skeletal muscle. J. Physiol. 577, 1021–1032.

Borge, B.A., Kalland, K.H., Olsen, S., Bletsa, A., Berggreen, E., Wiig, H., 2009. Cytokines areproduced locally by myocytes in rat skeletal muscle during endotoxemia. Am.J. Physiol. Heart Circ. Physiol. 296, H735–H744.

Brooks, S.V., Faulkner, J.A., 1994. Skeletal muscle weakness in old age: underlying mech-anisms. Med. Sci. Sports Exerc. 26, 432–439.

Bruunsgaard, H., Pedersen, B.K., 2003. Age-related inflammatory cytokines and disease.Immunol. Allergy Clin. North Am. 23, 15–39.

Canto, C., Jiang, L.Q., Deshmukh, A.S., Mataki, C., Coste, A., Lagouge, M., Zierath, J.R.,Auwerx, J., 2010. Interdependence of AMPK and SIRT1 for metabolic adaptation tofasting and exercise in skeletal muscle. Cell Metab. 11, 213–219.

Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidiniumthiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159.

Doherty, T.J., Vandervoort, A.A., Taylor, A.W., Brown, W.F., 1993. Effects of motor unitlosses on strength in older men and women. J. Appl. Physiol. 74, 868–874.

Dolinsky, V.W., Jones, K.E., Sidhu, R.S., Haykowsky, M., Czubryt, M.P., Gordon, T., Dyck, J.R.,2012. Improvements in skeletal muscle strength and cardiac function induced by res-veratrol during exercise training contribute to enhanced exercise performance inrats. J. Physiol. 590, 2783–2799.

Eisele, P.S., Salatino, S., Sobek, J., Hottiger, M.O., Handschin, C., 2013. The peroxisomeproliferator-activated receptor gamma coactivator 1alpha/beta (PGC-1) coactivatorsrepress the transcriptional activity of NF-kappaB in skeletal muscle cells. J. Biol.Chem. 288, 2246–2260.

Frontera, W.R., Meredith, C.N., O'Reilly, K.P., Knuttgen, H.G., Evans, W.J., 1988. Strengthconditioning in older men: skeletal muscle hypertrophy and improved function.J. Appl. Physiol. 64, 1038–1044.

Frost, R.A., Nystrom, G.J., Lang, C.H., 2002. Lipopolysaccharide regulates proinflammatorycytokine expression in mouse myoblasts and skeletal muscle. Am. J. Physiol. Regul.Integr. Comp. Physiol. 283, R698–R709.

Gollisch, K.S., Brandauer, J., Jessen, N., Toyoda, T., Nayer, A., Hirshman, M.F., Goodyear, L.J.,2009. Effects of exercise training on subcutaneous and visceral adipose tissue innormal- and high-fat diet-fed rats. Am. J. Physiol. Endocrinol. Metab. 297, E495–E504.

Guo, S.S., Zeller, C., Chumlea, W.C., Siervogel, R.M., 1999. Aging, body composition, andlifestyle: the Fels Longitudinal Study. Am. J. Clin. Nutr. 70, 405–411.

Handschin, C., Spiegelman, B.M., 2008. The role of exercise and PGC1alpha in inflamma-tion and chronic disease. Nature 454, 463–469.

Handschin, C., Chin, S., Li, P., Liu, F., Maratos-Flier, E., LeBrasseur, N.K., Yan, Z., Spiegelman,B.M., 2007a. Skeletal muscle fiber-type switching, exercise intolerance, and myopathyin PGC-1alpha muscle-specific knock-out animals. J. Biol. Chem. 282, 30014–30021.

Handschin, C., Choi, C.S., Chin, S., Kim, S., Kawamori, D., Kurpad, A.J., Neubauer, N., Hu, J.,Mootha, V.K., Kim, Y.B., Kulkarni, R.N., Shulman, G.I., Spiegelman, B.M., 2007b. Abnor-mal glucose homeostasis in skeletal muscle-specific PGC-1alpha knockout mice re-veals skeletal muscle-pancreatic beta cell crosstalk. J. Clin. Invest. 117, 3463–3474.

Harman, D., 1956. Aging: a theory based on free radical and radiation chemistry.J. Gerontol. 11, 298–300.

Hellsten, Y., Apple, F.S., Sjodin, B., 1996. Effect of sprint cycle training on activities of anti-oxidant enzymes in human skeletal muscle. J. Appl. Physiol. 81, 1484–1487.

Holloszy, J.O., 1967. Biochemical adaptations in muscle. Effects of exercise on mitochon-drial oxygen uptake and respiratory enzyme activity in skeletal muscle. J. Biol.Chem. 242, 2278–2282.

Hotamisligil, G.S., Shargill, N.S., Spiegelman, B.M., 1993. Adipose expression of tumor necro-sis factor-alpha: direct role in obesity-linked insulin resistance. Science 259, 87–91.


ED P

RO

OF

Hotamisligil, G.S., Murray, D.L., Choy, L.N., Spiegelman, B.M., 1994. Tumor necrosis factoralpha inhibits signaling from the insulin receptor. Proc. Natl. Acad. Sci. U. S. A. 91,4854–4858.

Hotamisligil, G.S., Arner, P., Caro, J.F., Atkinson, R.L., Spiegelman, B.M., 1995. Increasedadipose tissue expression of tumor necrosis factor-alpha in human obesity and insu-lin resistance. J. Clin. Invest. 95, 2409–2415.

Howitz, K.T., Bitterman, K.J., Cohen, H.Y., Lamming, D.W., Lavu, S., Wood, J.G., Zipkin,R.E., Chung, P., Kisielewski, A., Zhang, L.L., Scherer, B., Sinclair, D.A., 2003. Smallmolecule activators of sirtuins extend Saccharomyces cerevisiae lifespan. Nature425, 191–196.

Jackson, J.R., Ryan, M.J., Alway, S.E., 2011. Long-term supplementation with resveratrol al-leviates oxidative stress but does not attenuate sarcopenia in agedmice. J. Gerontol. ABiol. Sci. Med. Sci. 66, 751–764.

Jager, S., Handschin, C., St-Pierre, J., Spiegelman, B.M., 2007. AMP-activated protein kinase(AMPK) action in skeletal muscle via direct phosphorylation of PGC-1alpha. Proc.Natl. Acad. Sci. U. S. A. 104, 12017–12022.

Khazen, W., M'bika, J.P., Tomkiewicz, C., Benelli, C., Chany, C., Achour, A., Forest, C., 2005.Expression of macrophage-selective markers in human and rodent adipocytes. FEBSLett. 579, 5631–5634.

Lagouge, M., Argmann, C., Gerhart-Hines, Z., Meziane, H., Lerin, C., Daussin, F., Messadeq,N., Milne, J., Lambert, P., Elliott, P., Geny, B., Laakso, M., Puigserver, P., Auwerx, J., 2006.Resveratrol improves mitochondrial function and protects against metabolic diseaseby activating SIRT1 and PGC-1alpha. Cell 127, 1109–1122.

Leick, L., Wojtaszewski, J.F., Johansen, S.T., Kiilerich, K., Comes, G., Hellsten, Y., Hidalgo, J.,Pilegaard, H., 2008. PGC-1alpha is not mandatory for exercise- and training-inducedadaptive gene responses in mouse skeletal muscle. Am. J. Physiol. Endocrinol. Metab.294, E463–E474.

Leick, L., Lyngby, S.S., Wojtaszewski, J.F., Pilegaard, H., 2010. PGC-1alpha is required fortraining-induced prevention of age-associated decline in mitochondrial enzymes inmouse skeletal muscle. Exp. Gerontol. 45, 336–342.

Lin, J., Wu, H., Tarr, P.T., Zhang, C.Y., Wu, Z., Boss, O., Michael, L.F., Puigserver, P., Isotani, E.,Olson, E.N., Lowell, B.B., Bassel-Duby, R., Spiegelman, B.M., 2002. Transcriptional co-activator PGC-1 alpha drives the formation of slow-twitch muscle fibres. Nature418, 797–801.

Lin, J., Wu, P.H., Tarr, P.T., Lindenberg, K.S., St-Pierre, J., Zhang, C.Y., Mootha, V.K., Jager, S.,Vianna, C.R., Reznick, R.M., Cui, L., Manieri, M., Donovan, M.X., Wu, Z., Cooper, M.P.,Fan, M.C., Rohas, L.M., Zavacki, A.M., Cinti, S., Shulman, G.I., Lowell, B.B., Krainc, D.,Spiegelman, B.M., 2004. Defects in adaptive energy metabolism with CNS-linkedhyperactivity in PGC-1alpha null mice. Cell 119, 121–135.

Ling, C., Poulsen, P., Carlsson, E., Ridderstrale, M., Almgren, P., Wojtaszewski, J., Beck-Nielsen, H., Groop, L., Vaag, A., 2004. Multiple environmental and genetic factorsinfluence skeletal muscle PGC-1alpha and PGC-1beta gene expression in twins.J. Clin. Invest. 114, 1518–1526.

Lumeng, C.N., Liu, J., Geletka, L., Delaney, C., Delproposto, J., Desai, A., Oatmen, K.,Martinez-Santibanez, G., Julius, A., Garg, S., Yung, R.L., 2011. Aging is associatedwith an increase in T cells and inflammatory macrophages in visceral adipose tissue.J. Immunol. 187, 6208–6216.

Lundby, C., Nordsborg, N., Kusuhara, K., Kristensen, K.M., Neufer, P.D., Pilegaard, H., 2005.Gene expression in human skeletal muscle: alternative normalization method andeffect of repeated biopsies. Eur. J. Appl. Physiol. 95, 351–360.

Menzies, K.J., Singh, K., Saleem, A., Hood, D.A., 2013. Sirtuin 1-mediated effects of exerciseand resveratrol on mitochondrial biogenesis. J. Biol. Chem.

Oh-ishi, S., Kizaki, T., Nagasawa, J., Izawa, T., Komabayashi, T., Nagata, N., Suzuki, K.,Taniguchi, N., Ohno, H., 1997. Effects of endurance training on superoxide dismutaseactivity, content and mRNA expression in rat muscle. Clin. Exp. Pharmacol. Physiol.24, 326–332.

Olesen, J., Larsson, S., Iversen, N., Yousafzai, S., Hellsten, Y., Pilegaard, H., 2012. Skeletalmuscle PGC-1alpha is required for maintaining an acute LPS-induced TNFalpharesponse. PLoS One 7, e32222.

Olholm, J., Paulsen, S.K., Cullberg, K.B., Richelsen, B., Pedersen, S.B., 2010. Anti-inflamma-tory effect of resveratrol on adipokine expression and secretion in human adiposetissue explants. Int. J. Obes. (Lond.) 34, 1546–1553.

Pearson, K.J., Baur, J.A., Lewis, K.N., Peshkin, L., Price, N.L., Labinskyy, N., Swindell, W.R.,Kamara, D., Minor, R.K., Perez, E., Jamieson, H.A., Zhang, Y., Dunn, S.R., Sharma, K.,Pleshko, N., Woollett, L.A., Csiszar, A., Ikeno, Y., Le, C.D., Elliott, P.J., Becker, K.G.,Navas, P., Ingram, D.K., Wolf, N.S., Ungvari, Z., Sinclair, D.A., de, C.R., 2008. Resveratroldelays age-related deterioration andmimics transcriptional aspects of dietary restric-tion without extending life span. Cell Metab. 8, 157–168.

Pedersen, B.K., Febbraio, M.A., 2012. Muscles, exercise and obesity: skeletal muscle as asecretory organ. Nat. Rev. Endocrinol.

Pedersen, B.K., Bruunsgaard, H., Ostrowski, K., Krabbe, K., Hansen, H., Krzywkowski, K.,Toft, A., Sondergaard, S.R., Petersen, E.W., Ibfelt, T., Schjerling, P., 2000. Cytokines inaging and exercise. Int. J. Sports Med. 21 (Suppl. 1), S4–S9.

Pilegaard, H., Ordway, G.A., Saltin, B., Neufer, P.D., 2000. Transcriptional regulation of geneexpression in human skeletal muscle during recovery from exercise. Am. J. Physiol.Endocrinol. Metab. 279, E806–E814.

Pilegaard, H., Saltin, B., Neufer, P.D., 2003. Exercise induces transient transcriptionalactivation of the PGC-1alpha gene in human skeletal muscle. J. Physiol. 546,851–858.

Plomgaard, P., Bouzakri, K., Krogh-Madsen, R., Mittendorfer, B., Zierath, J.R., Pedersen, B.K.,2005. Tumor necrosis factor-alpha induces skeletal muscle insulin resistance inhealthy human subjects via inhibition of Akt substrate 160 phosphorylation. Diabetes54, 2939–2945.

Plomgaard, P., Nielsen, A.R., Fischer, C.P., Mortensen, O.H., Broholm, C., Penkowa, M.,Krogh-Madsen, R., Erikstrup, C., Lindegaard, B., Petersen, A.M., Taudorf, S., Pedersen,B.K., 2007. Associations between insulin resistance and TNF-alpha in plasma, skeletal



805806807808809810811Q7812Q8813814815Q9816817Q10818Q11819820821822823824825826827828829830

831832833834835836837838839840841842843844845846847848849850851852853854855856

858


muscle and adipose tissue in humans with and without type 2 diabetes. Diabetologia50, 2562–2571.

Poulsen, M.M., Vestergaard, P.F., Clasen, B.F., Radko, Y., Christensen, L.P., Stodkilde-Jorgensen, H., Moller, N., Jessen, N., Pedersen, S.B., Jorgensen, J.O., 2012. High-doseresveratrol supplementation in obese men: an investigator-initiated, randomized,placebo-controlled clinical trial of substrate metabolism, insulin sensitivity, andbody composition. Diabetes.

Ringholm, S., Olesen, J., Pedersen, J.T., Brandt, C.T., Halling, J.F., Hellsten, Y., Prats, C.,Pilegaard, H., 2013. Effect of lifelong resveratrol supplementation and exercisetraining on skeletal muscle oxidative capacity in aging mice; impact of PGC-1α. Inpending review. Exp. Gerontol.

Schaap, L.A., Koster, A., Visser, M., 2012. Adiposity, muscle mass, and muscle strength inrelation to functional decline in older persons. Epidemiol. Rev.

Skrobuk, P., von, K.S., Semenova, M.M., Zitting, A., Koistinen, H.A., 2012. Acute exposure toresveratrol inhibits AMPK activity in human skeletal muscle cells. Diabetologia 55,3051–3060.

Starkie, R., Ostrowski, S.R., Jauffred, S., Febbraio, M., Pedersen, B.K., 2003. Exercise and IL-6infusion inhibit endotoxin-induced TNF-alpha production in humans. FASEB J. 17,884–886.

St-Pierre, J., Drori, S., Uldry, M., Silvaggi, J.M., Rhee, J., Jager, S., Handschin, C., Zheng, K., Lin,J., Yang, W., Simon, D.K., Bachoo, R., Spiegelman, B.M., 2006. Suppression of reactiveoxygen species and neurodegeneration by the PGC-1 transcriptional coactivators.Cell 127, 397–408.

Timmers, S., Konings, E., Bilet, L., Houtkooper, R.H., van deWeijer, T., Goossens, G.H., Hoeks,J., van der Krieken, S., Ryu, D., Kersten, S., Moonen-Kornips, E., Hesselink, M.K., Kunz, I.,Schrauwen-Hinderling, V.B., Blaak, E.E., Auwerx, J., Schrauwen, P., 2011. Calorie

UNCO

RRECT

857


OF

restriction-like effects of 30 days of resveratrol supplementation on energymetabolismand metabolic profile in obese humans. Cell Metab. 14, 612–622.

Um, J.H., Park, S.J., Kang, H., Yang, S., Foretz, M., McBurney, M.W., Kim, M.K., Viollet, B.,Chung, J.H., 2010. AMP-activated protein kinase-deficient mice are resistant to themetabolic effects of resveratrol. Diabetes 59, 554–563.

Wei, J., Xu, H., Davies, J.L., Hemmings, G.P., 1992. Increase of plasma IL-6 concentrationwith age in healthy subjects. Life Sci. 51, 1953–1956.

Wellen, K.E., Hotamisligil, G.S., 2005. Inflammation, stress, and diabetes. J. Clin. Invest.115, 1111–1119.

Wenz, T., Rossi, S.G., Rotundo, R.L., Spiegelman, B.M., Moraes, C.T., 2009. Increased musclePGC-1alpha expression protects from sarcopenia and metabolic disease during aging.Proc. Natl. Acad. Sci. U. S. A. 106, 20405–20410.

Wood, J.G., Rogina, B., Lavu, S., Howitz, K., Helfand, S.L., Tatar, M., Sinclair, D., 2004. Sirtuin ac-tivators mimic caloric restriction and delay ageing in metazoans. Nature 430, 686–689.

Woods, J.A., Wilund, K.R., Martin, S.A., Kistler, B.M., 2012. Exercise, inflammation andaging. Aging Dis. 3, 130–140.

Wu, Z., Puigserver, P., Andersson, U., Zhang, C., Adelmant, G., Mootha, V., Troy, A., Cinti, S.,Lowell, B., Scarpulla, R.C., Spiegelman, B.M., 1999.Mechanisms controllingmitochondrialbiogenesis and respiration through the thermogenic coactivator PGC-1. Cell 98, 115–124.

Wu, D., Ren, Z., Pae,M., Guo,W., Cui, X., Merrill, A.H., Meydani, S.N., 2007. Aging up-regulatesexpression of inflammatory mediators in mouse adipose tissue. J. Immunol. 179,4829–4839.

Yoshino, J., Conte, C., Fontana, L., Mittendorfer, B., Imai, S., Schechtman, K.B., Gu, C., Kunz,I., Rossi, F.F., Patterson, B.W., Klein, S., 2012. Resveratrol supplementation does notimprove metabolic function in nonobese women with normal glucose tolerance.Cell Metab. 16, 658–664.
O
ED P

R



Study III

Jesper Olesen, Lasse Gliemann, Rasmus S. Biensø, Jakob Schmidt, Ylva Hellsten, Henriette Pilegaard.

Exercise training, but not resveratrol, improves metabolic and inflammatory status in human skeletal muscle

of aged men. Submitted to J. Physiol.

1

Exercise training, but not resveratrol, improves metabolic

and inflammatory status in human skeletal muscle of aged

men.

Jesper Olesen 1, Lasse Gliemann

2, Rasmus Biensø

1, Jakob Schmidt

2, Ylva Hellsten

2, Henriette

Pilegaard 1

1Centre of Inflammation and Metabolism, Department of Biology, University of Copenhagen. 2Integrated Physiology Group, Department of Nutrition,

Exercise and Sports, University of Copenhagen.

Running title: Resveratrol and exercise training in aged human subjects

Key words: Anti-oxidants; inflammation; physical activity, aging

Total word count: 5026 (without references and legends)

Corresponding author:

Henriette Pilegaard

August Krogh Building

Universitetsparken 13, 4th floor

2100 KBH Ø

Denmark

Email: [email protected]

Phone: +4535321687

mailto:[email protected]

2

Key point summary

-Aging is associated with lifestyle-related metabolic diseases and exercise training is suggested to counteract

such metabolic deteriorations.

-The natural anti-oxidant resveratrol has been shown to exert “exercise like” health beneficial metabolic and

anti-inflammatory effects in rodents, but little is known about the metabolic effects of resveratrol

supplementation alone and when combined with exercise training, in aged human subjects.

-The present findings showed that exercise training markedly improved muscle endurance, increased the

content and activity of oxidative proteins in skeletal muscle as well as reduced markers of oxidative stress

and inflammation in skeletal muscle of aged men.

-Resveratrol did not elicit health beneficial metabolic effects in healthy aged subjects and resveratrol even

impaired the exercise training-induced improvements in markers of oxidative stress and inflammation.

Word count key point summary: 123

3

Summary

Aim: The aim was to investigate the potential metabolic and anti-inflammatory effects of resveratrol alone

and combined with exercise training in human skeletal muscle of aged subjects. Material and Methods:

Healthy physically inactive aged men were randomized into either 8 weeks of daily intake of 250 mg

resveratrol or placebo or into 8 weeks of high intensity exercise training with 250 mg resveratrol or placebo.

Before and after the intervention, a one-leg knee-extensor endurance exercise test (KEE) was performed and

resting blood samples and muscle biopsies were obtained. Results: Exercise training increased time to

exhaustion in KEE by ~1.2-fold, PGC-1α mRNA ~1.5-fold, cytochrome c protein ~1.3 fold, cytochrome c

oxidase I protein ~1.5-fold, citrate synthase activity ~1.3-fold, 3-hydroxyacyl-CoA dehydrogenase activity

~1.3-fold, and IκB-α and IκB-β protein content ~1.3-fold in skeletal muscle with no significant additive or

adverse effects of resveratrol on these parameters. Despite an overall ~25% reduction in total acetylation

level in skeletal muscle with resveratrol, no exclusive resveratrol-mediated beneficial effect was observed on

the investigated parameters. Notably, resveratrol blunted the exercise training-induced ~20% decrease in

protein carbonylation and ~40% decrease in TNFα mRNA content in skeletal muscle. Conclusion:

Resveratrol did not elicit health beneficial metabolic effects in aged subjects and resveratrol even impaired

the observed exercise training-induced improvements in markers of oxidative stress and inflammation.

Collectively this highlight the effectiveness of exercise training in preventing age-related metabolic diseases

in healthy subjects and does not support that resveratrol is a potential exercise mimetic in healthy aged

subjects. (Summary word count: 248)

Abbreviations

AMP-activated protein kinase (AMPK); sirtuin (SIRT); peroxisome proliferator-activated receptor-γ co-

activator-1α (PGC-1α); citrate synthase (CS); 3-hydroxyacyl-CoA dehydrogenase (HAD); cytochrome c (cyt

c); cytochrome c oxidase I (COXI); tumor necrosis factor α (TNFα); inducible nitric oxide synthase (iNOS);

interleukin-6 (IL-6); c-reactive protein (CRP); c-Jun N-terminal kinases (JNK); inhibitor of κB-α (IκB-α);

inhibitor of κB-β (IκB-β); inhibitor of κB kinase (IKK); p38 mitogen-activated protein kinases (p38).

4

Introduction

Aging is directly linked to lifestyle-related metabolic diseases (Masoro, 2001;Woods et al., 2012). Although

many tissues and organs are affected by aging, loss of skeletal muscle mass in concert with decreased

oxidative and anti-oxidant capacity of skeletal muscle (Chabi et al., 2008;Conley et al., 2000;Hollmann et

al., 2007) is a hallmark of aging and may have a negative impact on whole body metabolism. Multiple and

divergent mediators may be involved in the initiation and/or progression of metabolic diseases including

chronic low-grade inflammation, which has been associated lifestyle- and age-related metabolic pathologies

(Astrom et al., 2010;Handschin & Spiegelman, 2008). However, other studies have failed to show

associations between low-grade inflammation and aging (Ahluwalia et al., 2001;Beharka et al., 2001), which

may suggest that low-grade inflammation is a secondary event in the pathogenesis of age- and lifestyle-

related metabolic diseases. Nevertheless, chronic low-grade inflammation seems to be a contributing factor

in the pathogenesis of age- and lifestyle-related metabolic diseases, which may ultimately influence quality

of life.

Exercise training elicits numerous health beneficial effects (Pedersen & Saltin, 2006). In aged subjects,

exercise training has been shown to counteract loss of muscle mass and strength (Frontera et al.,

1988;Hollmann et al., 2007), and thus postponing age-related muscle deteriorations. In addition, exercise

training increases the oxidative capacity of skeletal muscle (Gollnick et al., 1973;Henriksson & Reitman,

1977;Holloszy, 1967) and is believed to exert anti-inflammatory effects (Gleeson et al., 2011;Handschin &

Spiegelman, 2008;Woods et al., 2012). This may contribute to the beneficial effects of exercise training

during aging. However, not all people have the ability or desire to perform regular physical activity.

The natural polyphenol resveratrol, present in dark grapes and nuts, has been shown to induce positive

metabolic effects almost similar to those of exercise training (Baur et al., 2006;Lagouge et al., 2006).

Specifically, resveratrol has been shown to extent lifespan of lower species (Howitz et al., 2003;Wood et al.,

2004) and increase exercise endurance and skeletal muscle oxidative capacity in mice (Lagouge et al., 2006).

Additionally, resveratrol protects rodents from diet-induced obesity, insulin resistance and inflammation

(Baur et al., 2006;Dolinsky et al., 2012;Lagouge et al., 2006;Park et al., 2012;Um et al., 2010;Pearson et al.,

2008). Resveratrol is believed to exert its effects through a sirtuin (SIRT)1-mediated deacetylation of

peroxisome proliferator activated receptor-γ co-activator (PGC)-1α (Baur et al., 2006;Lagouge et al., 2006).

However, the exact mechanism is still debated and studies provide evidence suggesting that effects of

resveratrol may also occur via an AMP-activated protein kinase (AMPK)-mediated mechanism (Park et al.,

2012;Um et al., 2010). Regardless, SIRT1 and AMPK both seem pivotal and may in concert mediate the

metabolic adaptations of resveratrol in skeletal muscle through PGC-1α-mediated transcriptional regulation

(Canto et al., 2010). Moreover, resveratrol may not only mediate adaptations in various organs through an

AMPK/SIRT1/PGC-1α axis, but may additionally scavenge excessive ROS production through its alleged

5

anti-oxidant properties (Olas & Wachowicz, 2005;Stojanovic et al., 2001), which may be especially relevant

during exercise training in aged individuals.

Only few resveratrol studies have been conducted in humans and the reports on resveratrol-mediated

metabolic effects in humans have been inconsistent (Brasnyo et al., 2011;Crandall et al., 2012;Poulsen et al.,

2012;Timmers et al., 2011;Yoshino et al., 2012) and modest compared with the marked effects in rodent

models (Baur et al., 2006;Dolinsky et al., 2012;Lagouge et al., 2006;Park et al., 2012;Um et al., 2010).

Previous human studies have primarily focused on individuals with pre-existing metabolic disorders like type

2 diabetes, insulin resistance and obesity (Brasnyo et al., 2011;Crandall et al., 2012;Poulsen et al.,

2012;Timmers et al., 2011). To our knowledge, no previous human studies have examined the metabolic and

anti-inflammatory effects of resveratrol or the combined effects of resveratrol and exercise training on

metabolism and inflammation in healthy aged individuals.

The aim of the present study was to investigate the effects of 8 weeks of daily resveratrol intake either alone

or in combination with high intensity exercise training in healthy aged men. The study tested the hypotheses

that daily intake of resveratrol elicits metabolic adaptations and anti-inflammatory effects in skeletal muscle

of healthy aged subjects similar to exercise training, and that combined resveratrol intake and exercise

training potentiates such metabolic and anti-inflammatory effects.

Methods

Ethical statement and subjects

The study was approved by the ethics Committee of Copenhagen and Frederiksberg communities (H-2-2011-

079) and was conducted in accordance with the guidelines of The Declaration of Helsinki. All subjects

provided written informed consent before the initiation of the study.

The present study was part of a larger study and data covering cardiovascular adaptations to exercise training

with or without resveratrol supplementation have recently been published (Gliemann et al., 2013).

Forty-three aged physically inactive, but otherwise healthy male subjects participated (table S1, suppl.

information and (Gliemann et al., 2013)). All subjects were non-smokers and underwent a medical

examination. None had been diagnosed with cardiovascular disease, hypertension, renal dysfunction, insulin

resistance or type 2 diabetes and all subjects had normal ECG. Two subjects were diagnosed with

hypercholesterolemia regulated by their own physician (medication was maintained during the experimental

period), whereas the other participants had normal cholesterol levels.

6

Experimental setup

Study design

The study was divided into two parts, which both were 8-week randomized, double-blinded placebo

controlled trials. The subjects in the first part were assigned to either a combination of exercise training and

placebo (n = 13) or exercise training and 250 mg resveratrol·day-1

(Fluxome Inc., Stenlose, Denmark, n =

14). The subjects in the second part were assigned to either placebo (n = 7) or 250 mg resveratrol·day-1

(n =

9). The allocation was based on body mass index (BMI), blood glucose, cholesterol and maximal oxygen

consumption (table S1, suppl. information and (Gliemann et al., 2013)). All participants were instructed to

take one tablet each morning. Subjects noted time of consumption for each tablet and any discomfort that

might appear throughout the intervention period. Furthermore, the subjects were instructed not to change

their normal eating or drinking habits as well as their general activity level or way of living throughout the

experimental period.

Exercise training protocol

The exercise training intervention consisted of supervised high intensity interval spinning training (cycle

ergometer) 2 times/week and full body circuit training (crossfit) 1 time/week. In addition, the subjects were

instructed to walk 5 km ones per week. The intensity of the exercise was controlled by TEAM2 WearLink+

heart rate monitors (Polar, Kempele, Finland).

Endurance test

On the first experimental day, a dual-energy X-ray absorptiometry (DXA) scanning was performed in

addition to an incremental time to exhaustion dynamic one-leg knee-extensor exercise test. After

acclimatizing and a short warm-up, the test started at 6 W and gradually increased with 6 W every 5 minute

until exhaustion. The total energy output (KJ) was calculated based on duration (seconds) and workload

(Watt).

Muscle biopsies and blood samples

On the second experimental day (minimum 48 hours after the first experimental day), the subjects arrived

after an overnight fast and resting blood samples were taken from an arm vein and a vastus lateralis muscle

biopsy was obtained under local anesthesia (lidocaine; AstraZeneca, Södertälje, Sweden) using the

percutaneous needle biopsy technique (Bergstrom, 1975) with suction. Muscle biopsies were quick-frozen in

liquid nitrogen and stored (-80° C) until analysis. The two experimental days were repeated in the same order

after the 8-week intervention.

7

Analyses

Plasma cytokines

Plasma cytokines were analyzed using an ultra-sensitive MSD multi-spot 96 well assay system pre-coated

with antibodies (MesoScaleDiscovery, Gaithersburg, Maryland, USA) according to manufacturer’s protocol.

The MSD plates were measured on a MSD Sector Imager 2400 plate reader. Raw data were measured as

electrochemiluminescence signal (light) detected by photodetectors and analyzed using the Discovery

Workbench 3.0 software (MSD). A standard curve was generated for each analyte and used to determine the

concentration of analytes in each sample.

RNA isolation, RT and real time PCR

Total RNA was isolated from 15-20 mg muscle tissue by a modified guanidinium thiocyanate-phenol-

chloroform extraction method (Chomczynski & Sacchi, 1987) as described previously (Pilegaard et al.,

2000), except that the tissue was homogenized for 2 min at 30 s−1

in a TissueLyserII (Qiagen, Germany). The

final pellets were re-suspended in DEPC treated H2O containing 0.1mM EDTA. RNA was quantified based

on the absorbance at 260 nm (Nanodrop 1000, Thermo Scientific, Rockford, IL, USA). Purity of the RNA

samples was evaluated from 260nm/280nm ratio and all samples were above 1.8.

Superscript II RNase H− system and Oligo dT (Invitrogen, Carlsbad, CA, USA) were used to reverse

transcribe the mRNA to cDNA as described previously (Pilegaard et al., 2000). The amount of single-

stranded DNA (ssDNA) was determined in each cDNA sample by use of OliGreen reagent (Molecular

Probes, Leiden, The Netherlands) as described previously (Lundby et al., 2005).

Real-time PCR was performed using an ABI 7900 sequence-detection system (Applied Biosystems, Foster

City, CA, USA). Primers and TaqMan probes for amplifying gene-specific mRNA fragments were designed

using the human-specific database from ensemble.org and Primer Express (Applied Biosystems). All

TaqMan probes were 5´-FAM and 3´-TAMRA labeled, and primers and Taqman probes were obtained from

TAG Copenhagen (Copenhagen, Denmark) (table 1). Real-time PCR was performed in triplicates in a total

reaction volume of 10 µl using Universal Mastermix (Applied Biosystems). Cycle threshold (Ct) was

converted to a relative amount by use of a standard curve constructed from a serial dilution of a pooled RT

sample analyzed together with the samples. For each sample, target gene mRNA content was normalized to

ssDNA content.

Lysate generation and protein determination

Freeze-dried muscle biopsies were dissected free of visible fat, blood and connective tissue under a stereo

microscope in a temperature (~18° C ) and humidity (<30 %) controlled room. Muscle lysate was produced

from ~5-10 mg dry weight as previously described (Birk & Wojtaszewski, 2006), except that the tissue was

8

homogenized for 3 min at 30 s−1

in a TissueLyser (TissueLyser II; QIAGEN, Germany). Homogenates were

centrifuged for 20 min, at 16000 g, 4° and lysates (supernatant) were collected. Protein content in lysates was

measured by the bicinchoninic acid (BCA) method (Thermo Scientific, Rockford, IL, USA).

SDS-PAGE and western blotting

Protein content and phosphorylation levels were measured in muscle lysates by SDS-PAGE and western

blotting. Equal amounts of total protein were loaded of each sample. Band intensity was quantified using

Carestream IS 4000 MM (Fisher Scientific, ThermoFisher Scientific, Waltman, MA, USA) and Carestream

Health Molecular Imaging software. Commercially available antibodies were used to detect, AMPKThr172

(#2535), SIRT1 (#2493), acetylated lysine residues (#9441), TNFα (#3707), iNOS (#2977), IκB-α (#9242),

IκB-β (#9248), p65 (#4764), p65ser536

(#3033), IKK (#2678), IKKser176, 180

(#2697), JNK (#9252), JNKThr183,

Tyr185 (#9251), p38 (#9212), p38

Thr180, Tyr182 (# 4511) and GAPDH (#2118) all from Cell Signalling. Cyt c

(#556433, BD Pharmigen), COXI (#459600, Invitrogen) and AMPKα2 (a kind gift from Grahame Hardie,

Dundee). Protein content as well as phosphorylation levels were expressed as arbitrary units relative to

control samples loaded on each site of each gel and normalized to GAPDH protein.

Enzyme activities

A portion of the freeze-dried muscle biopsies, free of connective tissue, blood and visible fat was

homogenized (1:400) in 0.3 M phosphate-buffer (pH 7.7) containing 0.05% bovine serum albumin by use of

a TissueLyser (3 min, 30 s−1

). Maximal citrate synthase (CS) activity was determined according to

manufacturers protocol (Sigma-Aldrich, MO, USA) with absorbance kinetically measured at 405 nm

(Multiscan, Thermo Scientific) at baseline and after addition of oxaloacetate (Sigma-Aldrich). 3-

hydroxyacyl-CoA dehydrogenase (HAD) activity was kinetically determined at 355nm/460nm

(excitation/emission, Fluoroscan, Thermo Scientific) as previously described (Lowry et al., 1978). After

addition of acetoacetyl-CoA (Sigma-Aldrich), the delta emission was converted to activity. CS and HAD

activities were normalized to protein content measured in the respective samplesby the BCA method.

Protein carbonylation

Protein carbonylation was determined in homogenates made in a 0.3 M phosphate-buffer (described above)

using an OxiSelectTM

ELISA-kit (Cell Biolabs, SD, USA) according to the protocol of the manufacturer with

absorbance measured at 450 nm (Multiscan, Thermo Scientific). Based on a serial diluted oxidized/reduced

BSA standard curve, the absorbance was converted to protein carbonyl concentration and normalized to

protein content in the samples as determined by the BCA method.

9

Statistical analysis

Two-way repeated measures ANOVA was applied to test the effect of resveratrol vs. placebo and the

combined resveratrol/exercise training vs. combined placebo/exercise training. If normality or variance of

the data-set were skewed, the data were logarithmically transformed before applying the ANOVA. If a main

effect was observed, pairwise differences were located by Student Newman Keul’s multiple comparison post

hoc test. In addition, within-group comparisons were analyzed by paired T-test. P < 0.05 was considered

significant and a tendency is reported for 0.05 ≤ P < 0.1. All values are presented as means ± SE.

Results

Compliance

Based on self-reports, all subjects took their provided daily tablet and none of them reported any significant

side-effects throughout the intervention.

The subjects enrolled in the exercise protocol had a high compliance and completed 1.9 ± 0.1 (spinning) and

1.0 ± 0.09 (Crossfit) sessions per week. The intensity during the supervised training was equal between the

two groups and was on average above 70% HRmax during 67% of each training session and above 90% HRmax

during 14% of each training session (Gliemann et al. 2013).

Subject characteristics

Baseline characteristics as well as results after the exercise training intervention with or without resveratrol

supplementation for the subjects enrolled in the exercise training protocol have been characterized in detail

(Gliemann et al., 2013). Detailed baseline characteristics as well as the results after the intervention of the

subjects enrolled in the placebo and resveratrol intervention without exercise training are given in the

supplementary information (table S1, suppl. information). Briefly, general blood profile (fasting glucose,

total cholesterol, HDL, LDL, HDL/LDL, triglycerides) were similar in all groups before the intervention

(table S1, suppl. information and Gliemann et al. 2013) and did not change with either placebo or resveratrol

supplementation (table S1). Age, BMI, body fat percentage and mean arterial pressure (MAP) were similar

between all groups before the intervention (table S1, suppl. information and Gliemann et al. 2013) and did

not change with either placebo or resveratrol supplementation (table S1).

Endurance

Eight weeks of resveratrol or placebo intake did not affect endurance in the a one-leg knee extensor exercise

test (presented as % change in energy output in kilojoule). However, endurance was increased (P < 0.05)

~1.2 fold (from 25.8 ± 1.7 KJ to 31.6 ± 2.1 KJ) with exercise training and ~1.4 fold (from 22.9 ± 2.4 KJ to

10

32.2 ± 2.6 KJ) with exercise training combined with resveratrol supplementation, with no significant additive

effects of resveratrol supplementation (figure 1).

AMPK phosphorylation, SIRT1 protein and total acetylation in skeletal muscle

The two suggested mediators of both resveratrol and exercise training-induced metabolic adaptations in

skeletal muscle, AMPK and SIRT1, were determined. While resveratrol alone did not affect AMPK

phosphorylation (figure 2a) or SIRT1 protein content in skeletal muscle (figure S1, suppl. information), the

overall acetylation level decreased (P < 0.05) on average 27% with resveratrol (figure 2b). Moreover,

exercise training tended to increase (P = 0.06) AMPK phosphorylation 1.1 fold only in the exercise

training/placebo group (figure 2a). SIRT1 protein data from the exercise training part of the study has

recently been published (Gliemann et al., 2013).

PGC-1α mRNA content in skeletal muscle

The mRNA level of the downstream target of AMPK and SIRT1, PGC-1α, was determined in skeletal

muscle. While resveratrol alone did not affect PGC-1α mRNA content in skeletal muscle, exercise training

increased (p<0.05) PGC-1α mRNA content ~1.5 fold with no significant additional effect of resveratrol

(figure 3).

Enzyme activities and oxidative protein content in skeletal muscle

Resveratrol supplementation alone had no effect on the CS and HAD activity or the cyt c and COXI protein

content in skeletal muscle. In contrast, exercise training increased (P < 0.05) CS and HAD activity ~1.3 fold,

cyt c protein ~1.2 fold and COXI protein content ~1.5 fold, with no significant additional effects of

resveratrol supplementation when combined with exercise training (figure 4).

Protein carbonylation in skeletal muscle

The protein carbonylation level in skeletal muscle was evaluated as a marker of oxidative stress. While

resveratrol intake alone did not affect protein carbonylation in skeletal muscle, exercise training decreased (P

< 0.05) protein carbonyl levels ~20%. Resveratrol combined with exercise training blunted the exercise

training-induced reduction in protein carbonylation in skeletal muscle (figure 5).

Intracellular signalling

The three major pro-inflammatory signalling pathways inhibitor of κB kinase (IKK)/nuclear factor kappa-

light-chain-enhancer of activated B cells (NF-κB), c-Jun N-terminal kinase (JNK) and p38 mitogen-

activated protein kinases (p38) were analysed in skeletal muscle in order to examine whether the

interventions affected these signalling pathways.

Only a small tendency (P = 0.079) for a significant decrease was observed in p65 phosphorylation with

resveratrol intake, and the remaining inflammatory signalling pathways were unaffected by resveratrol

11

supplementation (figure 6 and suppl. figure S2). Exercise training increased (P < 0.05) the protein abundance

of both isoforms of the Inhibitor of κB (IκB)-α and IκB-β 1.2-1.3 fold and tended to decrease (P = 0.07) the

JNK phosphorylation level 16% in skeletal muscle. In addition, exercise training combined with resveratrol

intake decreased the p65 phosphorylation level by 25% (P < 0.05) (figure 6). The upstream kinase of IκB

(IKK) and p38 signaling were not affected by any of the interventions (figure S2, suppl. information).

Inflammatory markers

Resveratrol did not affect TNFα mRNA (figure 7) and protein or iNOS mRNA and iNOS protein content in

skeletal muscle (suppl. figure S3). However, exercise training decreased (P < 0.05) TNFα mRNA content

~40%, whereas resveratrol combined with exercise training blunted the exercise training-induced reduction

in TNFα mRNA content (figure 7). Muscle TNFα and iNOS protein was not affected by training with or

without resveratrol (figure S3, suppl. information).

Plasma levels of inflammatory markers

No differences were observed in systemic levels of TNFα, IL-6 or C-reactive protein in any of the

interventions (table 2).

Discussion

The main findings of the present study were that while exercise training markedly increased muscle

endurance, the content and/or activity of oxidative proteins and reduced markers of oxidative stress and

inflammation, resveratrol on the other hand did not affect these parameters or the blood lipid profile, body fat

percentage and MAP. In fact, resveratrol even impaired the exercise training-induced improvements in

markers of oxidative stress and inflammation. This finding contradicts our hypotheses and earlier studies in

animals and contrasts the observed profound beneficial metabolic effects observed with exercise training.

The observation that resveratrol did not exert any effects on oxidative proteins, inflammatory markers and

protein carbonylation in skeletal muscle or in metabolic parameters like blood lipid profile, body fat

percentage and MAP (suppl. table S1.) is in contrast to previous reports in rodents (Baur et al.,

2006;Dolinsky et al., 2012;Lagouge et al., 2006;Park et al., 2012;Um et al., 2010) and a recent study in

obese human subjects (Timmers et al., 2011). However, two recent studies in humans failed to observe any

metabolic effects of resveratrol supplementation (Poulsen et al., 2012;Yoshino et al., 2012). In accordance

with these studies, the present study did not show any resveratrol-mediated phosphorylation of AMPK or

increase in SIRT1 protein content in skeletal muscle. Although indicative, resveratrol did seem to affect

overall acetylation level in skeletal muscle, which may reflect increased SIRT1 activity as previously

suggested (Baur et al., 2006;Lagouge et al., 2006). However, the present data cannot exclude the possibility

that other members of the sirtuin family (SIRT2-7) may have contributed to the observed effect on

12

acetylation. The lack of effects of resveratrol on PGC-1α mRNA and its downstream targets cyt c and COXI

collectively indicates that the apparent resveratrol-mediated SIRT1 activation was inadequate to induce any

metabolic effects on oxidative proteins in the present settings. The discrepancies between the various human

studies do not seem to be dose-related, as the study by Timmers et al. (Timmers et al., 2011) with clear

metabolic improvements used an intermediate dose of resveratrol (150 mg/day for 30 days), whereas the

study by Youshino et al. (Yoshino et al., 2012) (75 mg/day for 12 weeks) and the study by Poulsen et al.

(Poulsen et al., 2012) (500 mg/day for 4 weeks) and the present study (250 mg/day for 8 weeks) did not

observe similar metabolic effects. It may be argued that the pharmacological window of resveratrol-mediated

metabolic effects is narrow, but growing evidence in rodent models does not support this. Specifically, doses

ranging from ~ 20 mg·kg-1

animal·day-1

to ~ 1 g·kg-1

animal·day-1

resveratrol have all shown profound

metabolic effects in rodents (Baur et al., 2006;Dolinsky et al., 2012;Lagouge et al., 2006;Price et al., 2012).

Alternatively, it may be speculated that the subjects in the present study as well as in the study by Youshino

et al. with non-obese women (Yoshino et al., 2012), were too metabolically healthy to observe any

improvements with resveratrol. Accordingly, resveratrol does not appear to improve plasma lipid profile,

glucose tolerance, insulin sensitivity or lifespan in metabolically healthy rodents (Jeon et al., 2012;Juan et

al., 2002;Miller et al., 2011;Strong et al., 2013;Turrens et al., 1997). Collectively, these observations

indicate that a certain level of metabolic dysfunction is a prerequisite to obtain favourable effects of

resveratrol. Further studies in humans are however required to verify this suggestion.

The present observations that exercise training led to a coordinated increase in muscle endurance,

mitochondrial oxidative proteins and CS and HAD enzyme activity are in accordance with previous studies

in young (Gollnick et al., 1973;Henriksson & Reitman, 1977;Holloszy, 1967) and aged (Iversen et al.,

2011;Suominen et al., 1977) subjects and emphasize the high plasticity of skeletal muscle even with

increasing age (Ghosh et al., 2011;Short et al., 2003). These data collectively support that exercise training

improves muscle function in aged individuals via several intracellular metabolic adaptations. The observed

concomitant increase in PGC-1α mRNA in skeletal muscle with exercise training may suggest that elevated

PGC-1α expression underlies the coordinated exercise training-induced adaptations in skeletal muscle. PGC-

1α is regarded as a “master” regulator of mitochondrial biogenesis and oxidative metabolism (Lin et al.,

2002;Puigserver et al., 1998;Wu et al., 1999). Previous reports have shown that PGC-1α mRNA content

increases in recovery from a single exercise bout (Baar et al., 2002;Pilegaard et al., 2003) and with exercise

training (Russell et al., 2003;Short et al., 2003). Moreover, several studies in genetic PGC-1α mice models

have emphasized the role of PGC-1α in basal mitochondrial function (Handschin et al., 2007;Lin et al.,

2002;Lin et al., 2004;Wende et al., 2007) and training-induced mitochondrial adaptations (Geng et al.,

2010;Leick et al., 2010). Taken together these data suggest that PGC-1α may have mediated the present

exercise training-induced metabolic adaptations in skeletal muscle of aged individuals.

13

The present findings that combined exercise training and resveratrol supplementation did not elicit additive

effects on oxidative protein content in skeletal muscle are in contrast to a previous study in mice showing

additive effects of exercise training and resveratrol supplementation on cyt c protein content and COX

activity in skeletal muscle (Menzies et al., 2013). However the present data may be explained by the

observed similar increase in PGC-1α mRNA in the two exercise training groups.

The present exercise training-induced reduction in protein carbonylation indicates that the subjects had a

lower level of oxidative stress after the training period. This may be due to several processes including

increased anti-oxidant capacity, decreased ROS production and increased capacity for removal of oxidized

proteins and these possibilities are not mutually exclusive. Although only SOD2 protein was found to

increase in the current study (Gliemann et al. 2013), other anti-oxidant systems not measured may also have

been increased and possibly explain the reduced protein carbonyl level. The novel observation that

resveratrol impaired the exercise training-induced reduction in protein carbonylation and thereby likely also

the reduced oxidative stress is surprising. However, previous studies have similarly reported that

supplementation with various anti-oxidants in combination with exercise training blunts exercise training-

induced adaptations (Gomez-Cabrera et al., 2008;Ristow et al., 2009). Reports have shown that ROS are

important inducers of anti-oxidant enzymes (Jackson et al., 2002) and although speculative, the previously

reported direct anti-oxidant properties of resveratrol (Olas & Wachowicz, 2005;Stojanovic et al., 2001) may

therefore suggest that resveratrol blunted a ROS-induced production of anti-oxidant enzymes and hence the

exercise training-induced reduction in oxidative stress. However, the observed similar protein levels of

catalase, GPX1 and SOD2 in the two exercise training groups (Gliemann et al. 2013) does not support this,

underlining that additional studies are needed to elucidate the effects of resveratrol on ROS levels during

exercise.

The increased abundance of IκB-α and IκB-β protein content and decreased p65 and JNK signalling

observed with exercise training indicate that exercise training led to reduced inflammatory signalling. These

findings are intriguing and may explain the observed reduced TNFα mRNA level and support that exercise

training has anti-inflammatory properties (Gleeson et al., 2011;Pedersen & Saltin, 2006;Woods et al., 2012).

A previous study has shown that adenoviral overexpression of PGC-1α in primary human muscle cells

increases the expression of IκB-α (Mormeneo et al., 2012), which binds NF-κB and keeps it sequestered in

an inactive state in the cytosol. Thus, it may be speculated that an exercise training-induced increase in PGC-

1α protein inhibited NF-κB activity by inducing the expression of IκB-α and IκB-β, and that this can explain

the exercise training-induced decrease in TNFα mRNA in the elderly subjects.

The novel finding that resveratrol blunted the exercise training-induced reduction in TNFα mRNA content is

interesting, but difficult to explain based on the somewhat inconclusive findings on inflammatory signalling

in the subjects. However it may indirectly be explained by the impaired exercise training-induced reduction

14

in oxidative stress in these subjects, although the apparent similar exercise training-induced increase in IκB

protein in skeletal muscle of both exercise trained groups does not fully support this. It should also be noted

that the decreased TNFα mRNA level with exercise training was not reflected at the protein level in skeletal

muscle or in plasma, which again suggest that the subjects have been too metabolically healthy for systemic

anti-inflammatory effects to be measurable?

In conclusion, the present findings indicate that resveratrol supplementation did not elicit health beneficial

metabolically related effects in healthy aged subjects. In contrast, resveratrol even impaired the observed

exercise training-induced improvements in markers of oxidative stress and inflammation in skeletal muscle.

This finding contradicts our hypotheses and earlier studies in animals (Baur et al., 2006;Jackson et al.,

2011;Kim et al., 2007) and contrasts the observed profound improvements in muscle endurance, oxidative

proteins and markers of oxidative stress and inflammation in skeletal muscle with exercise training in the

current study. The latter findings highlight the remarkable plasticity of skeletal muscle and efficiency of

exercise training even with increasing age. The present data support the notion that exercise training may

have numerous health beneficial effects in aged individuals potentially postponing age-related metabolic

diseases, while use of resveratrol as a daily supplement in conjunction with exercise training may be

questioned in healthy aged people.

15

Reference List

Ahluwalia N, Mastro AM, Ball R, Miles MP, Rajendra R, & Handte G (2001). Cytokine production by

stimulated mononuclear cells did not change with aging in apparently healthy, well-nourished women. Mech

Ageing Dev 122, 1269-1279.

Astrom MB, Feigh M, & Pedersen BK (2010). Persistent low-grade inflammation and regular exercise.

Front Biosci (Schol Ed) 2, 96-105.

Baar K, Wende AR, Jones TE, Marison M, Nolte LA, Chen M, Kelly DP, & Holloszy JO (2002).

Adaptations of skeletal muscle to exercise: rapid increase in the transcriptional coactivator PGC-1. FASEB J

16, 1879-1886.

Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C, Kalra A, Prabhu VV, Allard JS, Lopez-Lluch G,

Lewis K, Pistell PJ, Poosala S, Becker KG, Boss O, Gwinn D, Wang M, Ramaswamy S, Fishbein KW,

Spencer RG, Lakatta EG, Le CD, Shaw RJ, Navas P, Puigserver P, Ingram DK, de CR, & Sinclair DA

(2006). Resveratrol improves health and survival of mice on a high-calorie diet. Nature 444, 337-342.

Beharka AA, Meydani M, Wu D, Leka LS, Meydani A, & Meydani SN (2001). Interleukin-6 production

does not increase with age. J Gerontol A Biol Sci Med Sci 56, B81-B88.

Bergstrom J (1975). Percutaneous needle biopsy of skeletal muscle in physiological and clinical research.

Scand J Clin Lab Invest 35, 609-616.

Birk JB & Wojtaszewski JF (2006). Predominant alpha2/beta2/gamma3 AMPK activation during exercise in

human skeletal muscle. J Physiol 577, 1021-1032.

Brasnyo P, Molnar GA, Mohas M, Marko L, Laczy B, Cseh J, Mikolas E, Szijarto IA, Merei A, Halmai R,

Meszaros LG, Sumegi B, & Wittmann I (2011). Resveratrol improves insulin sensitivity, reduces oxidative

stress and activates the Akt pathway in type 2 diabetic patients. Br J Nutr 106, 383-389.

Canto C, Jiang LQ, Deshmukh AS, Mataki C, Coste A, Lagouge M, Zierath JR, & Auwerx J (2010).

Interdependence of AMPK and SIRT1 for metabolic adaptation to fasting and exercise in skeletal muscle.

Cell Metab 11, 213-219.

Chabi B, Ljubicic V, Menzies KJ, Huang JH, Saleem A, & Hood DA (2008). Mitochondrial function and

apoptotic susceptibility in aging skeletal muscle. Aging Cell 7, 2-12.

16

Chomczynski P & Sacchi N (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-

phenol-chloroform extraction. Anal Biochem 162, 156-159.

Conley KE, Jubrias SA, & Esselman PC (2000). Oxidative capacity and ageing in human muscle. J Physiol

526 Pt 1, 203-210.

Crandall JP, Oram V, Trandafirescu G, Reid M, Kishore P, Hawkins M, Cohen HW, & Barzilai N (2012).

Pilot study of resveratrol in older adults with impaired glucose tolerance. J Gerontol A Biol Sci Med Sci 67,

1307-1312.

Dolinsky VW, Jones KE, Sidhu RS, Haykowsky M, Czubryt MP, Gordon T, & Dyck JR (2012).

Improvements in skeletal muscle strength and cardiac function induced by resveratrol during exercise

training contribute to enhanced exercise performance in rats. J Physiol 590, 2783-2799.

Frontera WR, Meredith CN, O'Reilly KP, Knuttgen HG, & Evans WJ (1988). Strength conditioning in older

men: skeletal muscle hypertrophy and improved function. J Appl Physiol 64, 1038-1044.

Geng T, Li P, Okutsu M, Yin X, Kwek J, Zhang M, & Yan Z (2010). PGC-1alpha plays a functional role in

exercise-induced mitochondrial biogenesis and angiogenesis but not fiber-type transformation in mouse

skeletal muscle. Am J Physiol Cell Physiol 298, C572-C579.

Ghosh S, Lertwattanarak R, Lefort N, Molina-Carrion M, Joya-Galeana J, Bowen BP, Garduno-Garcia JJ,

Abdul-Ghani M, Richardson A, DeFronzo RA, Mandarino L, Van RH, & Musi N (2011). Reduction in

reactive oxygen species production by mitochondria from elderly subjects with normal and impaired glucose

tolerance. Diabetes 60, 2051-2060.

Gleeson M, Bishop NC, Stensel DJ, Lindley MR, Mastana SS, & Nimmo MA (2011). The anti-inflammatory

effects of exercise: mechanisms and implications for the prevention and treatment of disease. Nat Rev

Immunol 11, 607-615.

Gliemann L, Schmidt JF, Olesen J, Bienso RS, Peronard SL, Grandjean SU, Mortensen SP, Nyberg M,

Bangsbo J, Pilegaard H, & Hellsten Y (2013). Resveratrol Blunts the Positive Effects of Exercise Training

on Cardiovascular Health in Aged Men. J Physiol.

Gollnick PD, Armstrong RB, Saltin B, Saubert CW, Sembrowich WL, & Shepherd RE (1973). Effect of

training on enzyme activity and fiber composition of human skeletal muscle. J Appl Physiol 34, 107-111.

Gomez-Cabrera MC, Domenech E, Romagnoli M, Arduini A, Borras C, Pallardo FV, Sastre J, & Vina J

(2008). Oral administration of vitamin C decreases muscle mitochondrial biogenesis and hampers training-

induced adaptations in endurance performance. Am J Clin Nutr 87, 142-149.

Handschin C, Choi CS, Chin S, Kim S, Kawamori D, Kurpad AJ, Neubauer N, Hu J, Mootha VK, Kim YB,

Kulkarni RN, Shulman GI, & Spiegelman BM (2007). Abnormal glucose homeostasis in skeletal muscle-

17

specific PGC-1alpha knockout mice reveals skeletal muscle-pancreatic beta cell crosstalk. J Clin Invest 117,

3463-3474.

Handschin C & Spiegelman BM (2008). The role of exercise and PGC1alpha in inflammation and chronic

disease. Nature 454, 463-469.

Henriksson J & Reitman JS (1977). Time course of changes in human skeletal muscle succinate

dehydrogenase and cytochrome oxidase activities and maximal oxygen uptake with physical activity and

inactivity. Acta Physiol Scand 99, 91-97.

Hollmann W, Struder HK, Tagarakis CV, & King G (2007). Physical activity and the elderly. Eur J

Cardiovasc Prev Rehabil 14, 730-739.

Holloszy JO (1967). Biochemical adaptations in muscle. Effects of exercise on mitochondrial oxygen uptake

and respiratory enzyme activity in skeletal muscle. J Biol Chem 242, 2278-2282.

Howitz KT, Bitterman KJ, Cohen HY, Lamming DW, Lavu S, Wood JG, Zipkin RE, Chung P, Kisielewski

A, Zhang LL, Scherer B, & Sinclair DA (2003). Small molecule activators of sirtuins extend Saccharomyces

cerevisiae lifespan. Nature 425, 191-196.

Iversen N, Krustrup P, Rasmussen HN, Rasmussen UF, Saltin B, & Pilegaard H (2011). Mitochondrial

biogenesis and angiogenesis in skeletal muscle of the elderly. Exp Gerontol 46, 670-678.

Jackson JR, Ryan MJ, & Alway SE (2011). Long-term supplementation with resveratrol alleviates oxidative

stress but does not attenuate sarcopenia in aged mice. J Gerontol A Biol Sci Med Sci 66, 751-764.

Jackson MJ, Papa S, Bolanos J, Bruckdorfer R, Carlsen H, Elliott RM, Flier J, Griffiths HR, Heales S, Holst

B, Lorusso M, Lund E, Oivind MJ, Moser U, Di PM, Polidori MC, Signorile A, Stahl W, Vina-Ribes J, &

Astley SB (2002). Antioxidants, reactive oxygen and nitrogen species, gene induction and mitochondrial

function. Mol Aspects Med 23, 209-285.

Jeon BT, Jeong EA, Shin HJ, Lee Y, Lee DH, Kim HJ, Kang SS, Cho GJ, Choi WS, & Roh GS (2012).

Resveratrol attenuates obesity-associated peripheral and central inflammation and improves memory deficit

in mice fed a high-fat diet. Diabetes 61, 1444-1454.

Juan ME, Vinardell MP, & Planas JM (2002). The daily oral administration of high doses of trans-resveratrol

to rats for 28 days is not harmful. J Nutr 132, 257-260.

Kim HJ, Park KG, Yoo EK, Kim YH, Kim YN, Kim HS, Kim HT, Park JY, Lee KU, Jang WG, Kim JG,

Kim BW, & Lee IK (2007). Effects of PGC-1alpha on TNF-alpha-induced MCP-1 and VCAM-1 expression

and NF-kappaB activation in human aortic smooth muscle and endothelial cells. Antioxid Redox Signal 9,

301-307.

18

Lagouge M, Argmann C, Gerhart-Hines Z, Meziane H, Lerin C, Daussin F, Messadeq N, Milne J, Lambert

P, Elliott P, Geny B, Laakso M, Puigserver P, & Auwerx J (2006). Resveratrol improves mitochondrial

function and protects against metabolic disease by activating SIRT1 and PGC-1alpha. Cell 127, 1109-1122.

Leick L, Lyngby SS, Wojtaszewski JF, & Pilegaard H (2010). PGC-1alpha is required for training-induced

prevention of age-associated decline in mitochondrial enzymes in mouse skeletal muscle. Exp Gerontol 45,

336-342.

Lin J, Wu H, Tarr PT, Zhang CY, Wu Z, Boss O, Michael LF, Puigserver P, Isotani E, Olson EN, Lowell

BB, Bassel-Duby R, & Spiegelman BM (2002). Transcriptional co-activator PGC-1 alpha drives the

formation of slow-twitch muscle fibres. Nature 418, 797-801.

Lin J, Wu PH, Tarr PT, Lindenberg KS, St-Pierre J, Zhang CY, Mootha VK, Jager S, Vianna CR, Reznick

RM, Cui L, Manieri M, Donovan MX, Wu Z, Cooper MP, Fan MC, Rohas LM, Zavacki AM, Cinti S,

Shulman GI, Lowell BB, Krainc D, & Spiegelman BM (2004). Defects in adaptive energy metabolism with

CNS-linked hyperactivity in PGC-1alpha null mice. Cell 119, 121-135.

Lowry CV, Kimmey JS, Felder S, Chi MM, Kaiser KK, Passonneau PN, Kirk KA, & Lowry OH (1978).

Enzyme patterns in single human muscle fibers. J Biol Chem 253, 8269-8277.

Lundby C, Nordsborg N, Kusuhara K, Kristensen KM, Neufer PD, & Pilegaard H (2005). Gene expression

in human skeletal muscle: alternative normalization method and effect of repeated biopsies. Eur J Appl

Physiol 95, 351-360.

Masoro EJ (2001). Physiology of aging. Int J Sport Nutr Exerc Metab 11 Suppl, S218-S222.

Menzies KJ, Singh K, Saleem A, & Hood DA (2013). Sirtuin 1-mediated effects of exercise and resveratrol

on mitochondrial biogenesis. J Biol Chem.

Miller RA, Harrison DE, Astle CM, Baur JA, Boyd AR, de CR, Fernandez E, Flurkey K, Javors MA, Nelson

JF, Orihuela CJ, Pletcher S, Sharp ZD, Sinclair D, Starnes JW, Wilkinson JE, Nadon NL, & Strong R (2011).

Rapamycin, but not resveratrol or simvastatin, extends life span of genetically heterogeneous mice. J

Gerontol A Biol Sci Med Sci 66, 191-201.

Mormeneo E, Jimenez-Mallebrera C, Palomer X, De N, V, Vazquez-Carrera M, Orozco A, Nascimento A,

Colomer J, Lerin C, & Gomez-Foix AM (2012). PGC-1alpha induces mitochondrial and myokine

transcriptional programs and lipid droplet and glycogen accumulation in cultured human skeletal muscle

cells. PLoS One 7, e29985.

Olas B & Wachowicz B (2005). Resveratrol, a phenolic antioxidant with effects on blood platelet functions.

Platelets 16, 251-260.

19

Park SJ, Ahmad F, Philp A, Baar K, Williams T, Luo H, Ke H, Rehmann H, Taussig R, Brown AL, Kim

MK, Beaven MA, Burgin AB, Manganiello V, & Chung JH (2012). Resveratrol ameliorates aging-related

metabolic phenotypes by inhibiting cAMP phosphodiesterases. Cell 148, 421-433.

Pearson KJ, Baur JA, Lewis KN, Peshkin L, Price NL, Labinskyy N, Swindell WR, Kamara D, Minor RK,

Perez E, Jamieson HA, Zhang Y, Dunn SR, Sharma K, Pleshko N, Woollett LA, Csiszar A, Ikeno Y, Le CD,

Elliott PJ, Becker KG, Navas P, Ingram DK, Wolf NS, Ungvari Z, Sinclair DA, & de CR (2008). Resveratrol

delays age-related deterioration and mimics transcriptional aspects of dietary restriction without extending

life span. Cell Metab 8, 157-168.

Pedersen BK & Saltin B (2006). Evidence for prescribing exercise as therapy in chronic disease. Scand J

Med Sci Sports 16 Suppl 1, 3-63.

Pilegaard H, Ordway GA, Saltin B, & Neufer PD (2000). Transcriptional regulation of gene expression in

human skeletal muscle during recovery from exercise. Am J Physiol Endocrinol Metab 279, E806-E814.

Pilegaard H, Saltin B, & Neufer PD (2003). Exercise induces transient transcriptional activation of the PGC-

1alpha gene in human skeletal muscle. J Physiol 546, 851-858.

Poulsen MM, Vestergaard PF, Clasen BF, Radko Y, Christensen LP, Stodkilde-Jorgensen H, Moller N,

Jessen N, Pedersen SB, & Jorgensen JO (2012). High-Dose Resveratrol Supplementation in Obese Men: An

Investigator-Initiated, Randomized, Placebo-Controlled Clinical Trial of Substrate Metabolism, Insulin

Sensitivity, and Body Composition. Diabetes.

Price NL, Gomes AP, Ling AJ, Duarte FV, Martin-Montalvo A, North BJ, Agarwal B, Ye L, Ramadori G,

Teodoro JS, Hubbard BP, Varela AT, Davis JG, Varamini B, Hafner A, Moaddel R, Rolo AP, Coppari R,

Palmeira CM, de CR, Baur JA, & Sinclair DA (2012). SIRT1 is required for AMPK activation and the

beneficial effects of resveratrol on mitochondrial function. Cell Metab 15, 675-690.

Puigserver P, Wu Z, Park CW, Graves R, Wright M, & Spiegelman BM (1998). A cold-inducible coactivator

of nuclear receptors linked to adaptive thermogenesis. Cell 92, 829-839.

Ristow M, Zarse K, Oberbach A, Kloting N, Birringer M, Kiehntopf M, Stumvoll M, Kahn CR, & Bluher M

(2009). Antioxidants prevent health-promoting effects of physical exercise in humans. Proc Natl Acad Sci U

S A 106, 8665-8670.

Russell AP, Feilchenfeldt J, Schreiber S, Praz M, Crettenand A, Gobelet C, Meier CA, Bell DR, Kralli A,

Giacobino JP, & Deriaz O (2003). Endurance training in humans leads to fiber type-specific increases in

levels of peroxisome proliferator-activated receptor-gamma coactivator-1 and peroxisome proliferator-

activated receptor-alpha in skeletal muscle. Diabetes 52, 2874-2881.

Short KR, Vittone JL, Bigelow ML, Proctor DN, Rizza RA, Coenen-Schimke JM, & Nair KS (2003). Impact

of aerobic exercise training on age-related changes in insulin sensitivity and muscle oxidative capacity.

Diabetes 52, 1888-1896.

20

Stojanovic S, Sprinz H, & Brede O (2001). Efficiency and mechanism of the antioxidant action of trans-

resveratrol and its analogues in the radical liposome oxidation. Arch Biochem Biophys 391, 79-89.

Strong R, Miller RA, Astle CM, Baur JA, de CR, Fernandez E, Guo W, Javors M, Kirkland JL, Nelson JF,

Sinclair DA, Teter B, Williams D, Zaveri N, Nadon NL, & Harrison DE (2013). Evaluation of resveratrol,

green tea extract, curcumin, oxaloacetic acid, and medium-chain triglyceride oil on life span of genetically

heterogeneous mice. J Gerontol A Biol Sci Med Sci 68, 6-16.

Suominen H, Heikkinen E, Liesen H, Michel D, & Hollmann W (1977). Effects of 8 weeks' endurance

training on skeletal muscle metabolism in 56-70-year-old sedentary men. Eur J Appl Physiol Occup Physiol

37, 173-180.

Timmers S, Konings E, Bilet L, Houtkooper RH, van de Weijer T, Goossens GH, Hoeks J, van der Krieken

S, Ryu D, Kersten S, Moonen-Kornips E, Hesselink MK, Kunz I, Schrauwen-Hinderling VB, Blaak EE,

Auwerx J, & Schrauwen P (2011). Calorie restriction-like effects of 30 days of resveratrol supplementation

on energy metabolism and metabolic profile in obese humans. Cell Metab 14, 612-622.

Turrens JF, Lariccia J, & Nair MG (1997). Resveratrol has no effect on lipoprotein profile and does not

prevent peroxidation of serum lipids in normal rats. Free Radic Res 27, 557-562.

Um JH, Park SJ, Kang H, Yang S, Foretz M, McBurney MW, Kim MK, Viollet B, & Chung JH (2010).

AMP-activated protein kinase-deficient mice are resistant to the metabolic effects of resveratrol. Diabetes

59, 554-563.

Wende AR, Schaeffer PJ, Parker GJ, Zechner C, Han DH, Chen MM, Hancock CR, Lehman JJ, Huss JM,

McClain DA, Holloszy JO, & Kelly DP (2007). A role for the transcriptional coactivator PGC-1alpha in

muscle refueling. J Biol Chem 282, 36642-36651.

Wood JG, Rogina B, Lavu S, Howitz K, Helfand SL, Tatar M, & Sinclair D (2004). Sirtuin activators mimic

caloric restriction and delay ageing in metazoans. Nature 430, 686-689.

Woods JA, Wilund KR, Martin SA, & Kistler BM (2012). Exercise, inflammation and aging. Aging Dis 3,

130-140.

Wu Z, Puigserver P, Andersson U, Zhang C, Adelmant G, Mootha V, Troy A, Cinti S, Lowell B, Scarpulla

RC, & Spiegelman BM (1999). Mechanisms controlling mitochondrial biogenesis and respiration through

the thermogenic coactivator PGC-1. Cell 98, 115-124.

Yoshino J, Conte C, Fontana L, Mittendorfer B, Imai S, Schechtman KB, Gu C, Kunz I, Rossi FF, Patterson

BW, & Klein S (2012). Resveratrol supplementation does not improve metabolic function in nonobese

women with normal glucose tolerance. Cell Metab 16, 658-664.

21

Competing interests The authors declare no conflict of interests

Author contributions

JO, LG, RB, YH and HP designed and conceived the study. YH and HP acquired funding for the study and

the analyses. JO, LG, RB, JS and HP performed the analyses. JO and HP wrote the manuscript. LG, RB, JS

and YH reviewed the manuscript.

Funding

This study was supported by The Danish Ministry of Culture for Sports Research, The Danish Council for

Independent Research – Medical Sciences. The Centre of Inflammation and Metabolism (CIM) is supported

by a grant from the Danish National Research Foundation (# 02-512-55). CIM is part of the UNIK Project:

Food, Fitness & Pharma for Health and Disease, supported by the Danish Ministry of Science, Technology,

and Innovation. CIM is a member of DD2 - the Danish Center for Strategic Research in Type 2 Diabetes (the

Danish Council for Strategic Research, grant no. 09-067009 and 09-075724). The funders had no role in

study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Acknowledgements

The authors would like thank Fluxome Inc. (Stenlose, Denmark) for providing us with Trans-resveratrol and

placebo tablets. A special thanks to Iben Plate for establishing the contact with Fluxome, Ninna Iversen and

Anne Finne Pihl for technical assistance and Line Nielsen, Marie Linander Henriksen, Sebastian Peronard

and Simon Grandjean for their skillful training of the subjects.

22

Tables

Table 1. Primer and TaqMan probe sequences

Forward primer Reverse primer Probe

PGC-1α 5' CAAGCCAAACCAACAACTTTATCTCT 3' 5' CACACTTAAGGTGCGTTCAATAGTC 3' 5' AGTCACCAAATGACCCCAAGGGTTCC 3'

TNFα 5' TCTGGCCCAGGCAGTCAGAT 3' 5' AGCTGCCCCTCAGCTTGA 3' 5' CAAGCCTGTAGCCCATGTTGTAGCAAACC 3'

iNOS 5' AGCGGGATGACTTTCCAAGA 3' 5' TAATGGACCCCAGGCAAGATT 3' 5' CCTGCAAGTTAAAATCCCTTTGGCCTTATG 3'

Table 1. Peroxisome proliferator-activated receptor-γ co-activator (PGC)-1α, tumor necrosis factor (TNF)α and inducible nitric oxide

synthase (iNOS) primer and TaqMan probe sequences used for real time.

Table 2. Plasma cytokines

Placebo RSV T-Placebo T-RSV

Pre post Pre post Pre post Pre post

TNFα (pg/ml) 3.5±0.5 3.2±0.5 3.2±0.4 3.7±0.5 3.2±0.3 3.2±0.1 3.4±0.3 3.7±0.3

IL-6 (pg/ml) 2.0±0.4 2.5±0.6 2.3±0.3 2.5±0.3 2.4±0.5 2.0±0.3 2.8±0.4 3.0±0.4

CRP (mg/l) 1.6±0.4 1.7±0.2 2.6±0.8 2.6±0.7 1.1±0.1 1.1±0.1 3.0±0.7 2.4±0.7

Table 2. Plasma tumor necrosis factor (TNF)α, interleukin (IL)-6 and c-reactive protein (CRP) obtained from placebo (n = 7),

resveratrol (RSV) (n = 9; 250 mg per day), exercise training and placebo (n = 13) and exercise training and RSV (n = 14; 250 mg per

day) supplemented subjects pre and post 8 weeks of intervention. Values are presented as means ± SE.

23

Figures

Figure 1.

Muscle endurance

Placebo RSV T-Placebo T-RSV

Change in e

ndura

nce (

%)

-20

0

20

40

60

80

100

*

*

24

Figure 2.

25

Figure 3.

PGC-1mRNA expression


PG

C-1m

RN

A/s

sD

NA

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0Pre

Post **

26

Figure 4.

27

Figure 5.

Protein carbonyls


Pro

tein

carb

onyls

(%

of P

re level)

0

20

40

60

80

100

120Pre

Post

*

28

Figure 6.

29

Figure 7.

TNF mRNA


TN

F

mR

NA

/ssD

NA

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4 Pre

Post

*

30

Legends

Figure 1. Percentage change in endurance from before (Pre) to after (Post) 8 weeks of intervention in placebo

(n = 7), resveratrol (RSV) (n = 9; 250 mg per day), exercise trained and placebo (n = 13) and exercise trained

and RSV (n = 14; 250 mg per day) supplemented subjects. Values are presented as means ± SE. *:

Significantly different from Pre within treatment, P < 0.05.

Figure 2. AMP-activated protein kinase (AMPK)Thr182

phosphorylation normalized to AMPK-α2 protein

content (a) and total acetylation of lysine residues normalized to glyceraldehyde 3-phosphate dehydrogenase

(GAPDH) protein (b) in muscle lysates from placebo (n = 7), resveratrol (RSV) (n = 9; 250 mg per day),

exercise trained and placebo (n = 13) and exercise trained and RSV (n = 14; 250 mg per day) supplemented

subjects before (Pre) and after (Post) 8 weeks of intervention. Protein content and acetylation are given in

arbitrary units (AU). Representative blots are shown on each figure with samples loaded in the same order as

depicted on the graph. Values are presented as means ± SE.*: Significantly different from Pre within

treatment, P <0.05. (*): Tends to be significantly different from Pre within treatment, 0.05 < P ≤ 01.

Figure 3. Peroxisome proliferator-activated receptor-γ co-activator (PGC)-1α mRNA normalized to single

stranded DNA content (ssDNA) in skeletal muscle from placebo (n = 7), resveratrol (RSV) (n = 9; 250 mg

per day), exercise trained and placebo (n =13) and exercise trained and RSV (n = 14; 250 mg per day)

supplemented subjects before (Pre) and after (Post) 8 weeks of intervention. Values are presented as means ±

SE. *: Significantly different from Pre within treatment, P <0.05.

Figure 4. Citrate synthase (CS) activity (a), 3-hydroxyacyl-CoA dehydrogenase (HAD) activity (b),

cytochrome c (Cyt c) protein (c) and cytochrome c oxidase (COX)1 protein (d) in skeletal muscle from

placebo (n = 7), resveratrol (RSV) (n = 9; 250 mg per day), exercise trained and placebo (n = 13) and

exercise trained and RSV (n =14; 250 mg per day) supplemented subjects before (Pre) and after (Post) 8

weeks of intervention. Protein content is given in arbitrary units (AU). Representative blots are shown for cyt

c and COX1 with samples loaded in the same order as depicted on the graph. Values are presented as means

± SE. *: Significantly different from Pre within treatment, P < 0.05.

Figure 5. Protein carbonylation levels in skeletal muscle from placebo (n = 7), resveratrol (RSV) (n = 9; 250

mg per day), exercise trained and placebo (n = 13) and exercise trained and RSV (n = 14; 250 mg per day)

31

supplemented subjects before (Pre) and after (Post) 8 weeks of intervention. Values are presented as means ±

SE. *: Significantly different from Pre within treatment, P < 0.05.

Figure 6. C-Jun N-terminal kinase (JNK)Thr183, Tyr185

phosphorylation normalized to JNK protein (a), p65ser536

phosphorylation normalized to p65 protein (b), inhibitor of κB (IκB)-α (c) and IκB-β protein in muscle

lysates from placebo (n = 7), resveratrol (RSV) (n = 9; 250 mg per day), exercise trained and placebo (n =

13) and exercise trained and RSV (n = 14; 250 mg per day) supplemented subjects before (Pre) and after

(Post) 8 weeks of intervention. Phosphorylation and protein content are given in arbitrary units (AU).

Representative blots are shown on each figure with samples loaded in the same order as depicted on the

graph. Values are presented as means ± SE. *: Significantly different from Pre within treatment, P < 0.05.

(*): Tends to be significantly different from Pre within treatment, 0.05 < P ≤ 01.

Figure 7. Tumor necrosis factor (TNF)α mRNA content in skeletal muscle from placebo (n = 7), resveratrol

(RSV) (n = 9; 250 mg per day), exercise trained and placebo (n = 13) and exercise trained and RSV (n = 14;

250 mg per day) supplemented subjects before (Pre) and after (Post) 8 weeks of intervention. TNFα mRNA

is normalized to single stranded DNA (ssDNA). Values are presented as means ± SE. *: Significantly

different from pre within treatment, P < 0.05.

32

Supplementary data

Table S1. Basic characteristics.

Table S1. Basic characteristics before (Pre) and after (Post) 8 weeks of placebo (n = 7) or resveratrol (RSV)

(n = 9; 250 mg per day) treatment. Values are presented as means ± SE. Mean arterial pressure (MAP), high

density lipoproteins (HDL), low density lipoproteins (LDL).

Placebo (n=7) Resveratrol (n=9)

Pre Post Pre Post

Age (years) 65.1±1.5 65.2±1.0

Weight (kg) 78.3±3.0 78.7±3.2 83.9±3.7 84.1±3.1

Body mass index (kg·m-2) 25.2±0.9 25.5±1.0 26.1±0.6 26.2±0.9

Body fat (%) 26.2±0.5 28.7±0.5 26.7±1.5 27.4±1.4

Lean body mass (kg) 55.3±1.8 54.0±2.0 59.9±1.0 59.2±1.5

MAP (mmHg) 92.0±2.8 90.2±3.3 93.1±3.5 91.4±3.3

Fasting glucose (mM) 5.1±0.1 5.2±0.1 5.3±0.2 5.2±0.2

Total cholesterol (mM) 6.0±0.4 6.1±0.4 5.7±0.3 5.7±0.3

HDL (mM) 1.7±0.2 1.7±0.1 1.6±0.1 1.6±0.1

LDL (mM) 3.8±0.3 4.3±0.6 3.5±0.3 3.6±0.3

HDL/LDL ratio 0.5±0.04 0.4±0.05 0.5±0.05 0.4±0.05

Triglycerides (mM) 1.1±0.1 1.3±0.1 1.2±0.2 1.3±0.2

33

Figure S1.

Figure S1. SIRT1 protein content in skeletal muscle before (pre) and after (post) 8 weeks of placebo (n = 7)

or resveratrol (RSV) (n = 9; 250 mg per day) supplementation. A representative blot of SIRT1 is shown

above the bar graph with samples loaded in the same order as depicted on the graph. Values are presented as

means ± SE.

34

Figure S2.

Figure S2. IKK phosphorylation normalized to IKK protein (a) and p38 phosphorylation normalized to p38

protein in muscle lysates from placebo (n = 7), resveratrol (RSV) (n = 9; 250 mg per day), exercise trained

and placebo (n = 13) and exercise trained and RSV (n = 14; 250 mg per day) supplemented subjects before

(Pre) and after (Post) 8 weeks of intervention. Phosphorylation is given in arbitrary units (AU). Values are

presented as means ± SE.

35

Figure S3.

Figure S3. iNOS mRNA (a), iNOS protein (b) and TNFα protein (c) in skeletal muscle from placebo (n = 7),

resveratrol (RSV) (n = 9; 250 mg per day), exercise trained and placebo (n = 13) and exercise trained and

RSV (n = 14; 250 mg per day) supplemented subjects before (Pre) and after (Post) 8 weeks of intervention.

iNOS mRNA is normalized to single stranded DNA (ssDNA), and protein content is given in arbitrary units

(AU). Values are presented as means ± SE.

Role of PGC-1α in acute and low-grade inflammation Olesen.pdf · Summary The aim of the present...

Documents

Transcript of Role of PGC-1α in acute and low-grade inflammation Olesen.pdf · Summary The aim of the present...