You are on page 1of 9

REVIEWS

Muscles, exercise and obesity: skeletal muscle as a secretory organ


Bente K. Pedersen and Mark A. Febbraio
Abstract | During the past decade, skeletal muscle has been identified as a secretory organ. Accordingly, we have suggested that cytokines and other peptides that are produced, expressed and released by muscle fibres and exert either autocrine, paracrine or endocrine effects should be classified as myokines. The finding that the muscle secretome consists of several hundred secreted peptides provides a conceptual basis and a whole new paradigm for understanding how muscles communicate with other organs, such as adipose tissue, liver, pancreas, bones and brain. However, some myokines exert their effects within the muscle itself. Thus, myostatin, LIF, IL6 and IL7 are involved in muscle hypertrophy and myogenesis, whereas BDNF and IL6 are involved in AMPKmediated fat oxidation. IL6 also appears to have systemic effects on the liver, adipose tissue and the immune system, and mediates crosstalk between intestinal L cells and pancreatic islets. Other myokines include the osteogenic factors IGF1 and FGF2; FSTL1, which improves the endothelial function of the vascular system; and the PGC1dependent myokine irisin, which drives brownfatlike development. Studies in the past few years suggest the existence of yet unidentified factors, secreted from muscle cells, which may influence cancer cell growth and pancreas function. Many proteins produced by skeletal muscle are dependent upon contraction; therefore, physical inactivity probably leads to an altered myokine response, which could provide a potential mechanism for the association between sedentary behaviour and many chronic diseases.
Pedersen, B. K. & Febbraio, M. A. Nat. Rev. Endocrinol. advance online publication 3 April 2012; doi:10.1038/nrendo.2012.49

Introduction
The views on hormonal regulation of metabolism in health and disease have markedly changed over the past 20years, principally owing to extensive research into adipose tissue. Initially considered an inert storage compartment for triglycerides, pioneering work in the mid 1980s demonstrated that adipocytes are an abundant source of a specific secretory protein called complement factor D or adipsin.1 A little over a decade ago, in a landmark finding, Zhang etal.2 identified leptin as a fat-cell-specific secretory factorlacking in the obese ob/ob mousethat mediates a canonical hormonal signal between adipose tissue and the brain. Since then, adiponectin, resistin, nicotinamide phosphoribosyltransferase (also known as visfatin) and retinol-binding protein 4 have joined the growing list of adipokines.3 Although adipokines have been the focus of much research in terms of their role as circulatory factors with effects on metabolically active tissue, it should be noted that during contractions, muscle cells undergo one of the most marked alterations to cellular quiescence in physiology and indeed pathophysiology. In addition, exercise influences the metabolism and function of several organs. Muscles, therefore, could represent an important source of secretory molecules with either local or endocrine effects. Apart from adiponectin, 4 most of the factors that are produced by adipocytes are proinflammatoryfor
Competing interests The authors declare no competing interests.

example, TNF, CCL2 and PAI-1and are potentially harmful with regard to the development of obesityinduced metabolic and cardiovascular diseases. This understanding raises the important question of which tissue or tissues could be protective and provide a counterbalance to the proinflammatory factors that are produced by adipocytes. Even short periods of physical inactivity are associated with metabolic changes, including decreased insulin sensitivity, attenuation of postprandial lipid metabolism, loss of muscle mass and accumulation of visceral adipose tissue.5,6 Such abnormalities probably represent a link between reduced exercise and increased risks of the progression of chronic disorders and premature mortality.7 Physical inactivity increases the risk of type2 diabetes mellitus (T2DM), 8 cardiovascular diseases, 9 colon cancer, 10 postmenopausal breast cancer 11 and osteoporosis.12 A reasonable suggestion is that skeletal muscle might mediate some of the well-established protective effects of exercise via secretion of proteins that could counteract the harmful effects of proinflammatory adipokines (Figure1). The idea that muscle cells might produce and release a humoral factor dates back many years before the identification of adipose tissue as an endocrine organ. For nearly half a century, researchers had hypothesized that skeletal muscle cells possess a humoral factor that is released in response to increased glucose demand during contraction.13 To date, owing to lack of more precise

The Centre of Inflammation and Metabolism, Department of Infectious Diseases and CMRC, Rigshospitalet, Section 7641, Faculty of Health Sciences, University of Copenhagen, Blegdamsvej 9, DK2100, Copenhagen, Denmark (B.K.Pedersen). Cellular and Molecular Metabolism Laboratory, Baker IDI Heart and Diabetes Institute, 75Commercial Road, Melbourne, VIC 3004, Australia (M. A. Febbraio). Correspondence to: B. K. Pedersen bkp@rh.dk

NATURE REVIEWS | ENDOCRINOLOGY


2012 Macmillan Publishers Limited. All rights reserved

ADVANCE ONLINE PUBLICATION | 1

REVIEWS
Key points
Myokines are cytokines or other peptides that are produced, expressed and released by muscle fibres Myokines may exert autocrine, paracrine or endocrine effects Myokines may balance and counteract the effects of adipokines The musclecell secretome consists of several hundred secreted products Identified myokines include myostatin, LIF, IL6, IL7, BDNF, IGF1, FGF2, FSTL1 and irisin Myokines may mediate protective effects of muscular exercise, with regard to diseases associated with a physically inactive lifestyle

Myokines

Identification of skeletal muscle as a secretory organ has created a new paradigm: muscles produce and release myokines, which work in a hormone-like fashion and exert specific endocrine effects on distant organs. Other proteins produced by skeletal muscle that are not released into the circulation, could work via autocrine or paracrine mechanisms, exerting their effects on signalling pathways within the muscle itself.2025 Myokines could, therefore, be involved in mediating the multiple health benefits of exercise. This Review provides an update on some of the muscle-derived cytokines that have been identified to date (Figure2). Furthermore, the identification of skeletal muscle as an endocrine organ has clinical implications, which are highlighted in the Review, such as the central part that skeletal muscle plays in organ crosstalk, including muscleliver and muscleadipose tissue crosstalk.

Myostatin
Myokines Type 2 diabetes mellitus, cardiovascular disease, cancer, osteoporosis Proin ammatory adipokines

Figure 1 | Interplay between adipokines and myokines represent a yinyang balance. Especially under conditions of obesity, adipose tissue secretes adipokines, which contribute to establish a chronic inflammatory environment that promotes pathological processes such as atherosclerosis and insulin resistance. Skeletal muscles are capable of producing myokines that confer some of the health benefits of exercise. Such myokines might counteract the harmful effects of proinflammatory adipokines.

knowledge, the unidentified contraction-induced factor has been named the work stimulus, the work factor or the exercise factor.14 In our view, the plural form exercise factors would be more applicable, given the fact that multiple metabolic and physiologic changes are induced by exercise. The early view on the exercise factor concept was predicated by the fact that contracting skeletal muscle mediates metabolic and physiologic responses in other organs that are not mediated via the nervous system. Namely, electrical stimulation of paralysed muscles in patients with spinal cord injury with no afferent or efferent impulses induces many of the same physiological changes as in uninjured individuals.15,16 Contracting skeletal muscles must, therefore, be able to communicate to other organs via humoral factors, which are released into the circulation during physical activity. Such factors might directly or indirectly influence the function of other organs such as adipose tissue, liver, the cardiovascular system and the brain. Skeletal muscle represents approximately 40% of the body weight in lean men and women and, therefore, constitutes the largest organ in nonobese humans. During the past decade, muscle cells have been identified as cells with a high secretory capacity, in support of the concept of adipocytes being major endocrine cells. Muscle cells are thought to have the capacity to produce several hundred secreted factors.1719 In 2003, Pedersen etal. suggested that cytokines or other peptides that are produced, expressed and released by muscle fibres and exert endocrine effects should be classified as myokines.14
2 | ADVANCE ONLINE PUBLICATION

Myostatin (also known as growth/differentiation factor8), was the first secreted muscle factor to fulfil the criteria of a myokine. This protein is secreted into the circulation. Myostatin is a highly conserved member of the TGF- superfamily, and inactivation of the myostatin gene (knockout) results in extensive skeletal muscle hypertrophy in mice,26 cattle and humans.27 In addition to the regulatory roles of myostatin on skeletal muscle growth, this myokine is also involved in the maintenance of metabolic homeostasis and in modulation of adipose tissue function and mass.2831 Deletion of myostatin in mice produces concomitant skeletal muscle hypertrophy and reduction in total body adipose tissue.32,33 In humans and rodents, aerobic and strength exercise attenuate myostatin expression, whereas myostatin inactivation seems to potentiate the beneficial effects of endurance exercise on metabolism.34 Several lines of evidence demonstrate that obesity is associated with increased myostatin expression. Muscle and circulating myostatin protein levels are increased in individuals with obesity; furthermore, myostatin secretion from myotubes derived from myoblasts isolated from muscle biopsy samples is increased in women with obesity compared with lean women.35 Follistatin, another member of the TGF- superfamily, is a naturally occurring inhibitor of myostatin with regard to its regulatory role in skeletal muscle. Plasma follistatin levels increased in healthy individuals following acute bicycle exercise, with a peak increase of sevenfold. However, there was no net release of follistatin from the exercising limb, which suggests that contracting skeletal muscle was not the source of follistatin. In mice performing a bout of swimming exercise, a marked increase occurred both in plasma follistatin levels as well as follistatin mRNA and protein expression in the liver. The marked increase in systemic levels of follistatin could, in principle, contribute to the regulation of muscular expression of myostatin in relation to exercise. Although follistatin should probably be classified as a hepatokine rather than a myokine, these data suggest the existence of possible muscleliver crosstalk during and following exercise.36
www.nature.com/nrendo

2012 Macmillan Publishers Limited. All rights reserved

REVIEWS
IL6
The cytokine IL-6 was the first myokine found to be secreted into the bloodstream in response to muscle contractions.23 The cytokine was serendipitously discovered as a myokine because of the observation that its levels increased in an exponential fashion proportional to the length of exercise and the amount of muscle mass engaged in the exercise. Thus, plasma IL-6 levels can increase up to 100-fold in response to exercise, although less dramatic increases are more frequent. 37 IL-6 is expressed by human myoblasts,38,39 human cultured myotubes,40 growing murine myofibres and associated muscle stem cells (satellite cells).41 In addition, IL-6 is released from human primary muscle cell cultures from healthy individuals and from patients with T2DM.42,43 Interestingly, the increase in IL-6 levels in the circulation occurs during exercise without any sign of muscle damage.37 Until the beginning of this millennium, it was commonly thought that the increase in IL-6 levels during exercise was a consequence of an immune response owing to local damage in the working muscles,44 and macrophages were hypothesized to be responsible for this increase.45 However, the IL6 mRNA level in monocytes does not increase as a result of exercise.46 This finding was confirmed at the protein level.47,48 Today, muscle cells are known to be the dominant source of IL-6 production during exercise. Furthermore, the hepatosplanchnic viscera remove IL-6 from the circulation in humans during exercise.49 The removal of IL-6 by the liver could constitute a mechanism that limits the negative metabolic effects of chronically elevated levels of circulating IL-6. Several pieces of evidence support the notion that IL-6 is produced by muscle cells during exercise. The nuclear transcription rate for IL-6 and IL6 mRNA levels increase rapidly and markedly within 30 min of the onset of exercise.50,51 which suggests that a factor associated with contraction is involved in the regulation of IL-6 transcription within the nuclei of muscle cells. Further evidence that contracting muscle fibres are themselves a source of IL6 mRNA and protein has been gained from analysis of biopsy samples from the human vastus lateralis muscle using insitu hybridization and immunohistochemistry techniques.52 Microdialysis studies suggest that the concentration of IL-6 within the contracting skeletal muscle could be fivefold to 100-fold higher than the levels found in the circulation and supports the idea that IL-6 accumulates within muscle fibres as well as in the interstitium during and following exercise.53 The simultaneous measurement of arteriovenous IL-6 concentrations and blood flow across an exercising leg has demonstrated that large amounts of IL-6 are also released into the circulation from the exercising muscle.54 Human skeletal muscle is unique in that it can produce IL-6 during contraction in a strictly TNF-independent fashion. 40 This finding suggests that muscular IL-6 has a role in metabolism rather than in inflammation. In support of this hypothesis, both intramuscular IL6 mRNA expression55 and IL-6 protein release56 are markedly enhanced when intramuscular glycogen levels
NATURE REVIEWS | ENDOCRINOLOGY
2012 Macmillan Publishers Limited. All rights reserved
Lipolysis

UCP-1

IL-6 Irisin IL-6 BDNF IL-6

Follistatin LIF IL-4 IL-6 IL-7 IL-15 Myostatin Unknown exercise Hypertrophy stimulus IL-6

Hepatic glucose production during exercise Hepatic CXCL-1 production

IL-6

Lipolysis

AMPK Glucose uptake Fat oxidation Angiogenesis IGF-1 FGF-2 FGF-21 IL-8? IL-6 Insulin secretion CXCL-1? via GLP-1 Follistatin-related protein 1 Promotes endothelial function and revascularization

Figure 2 | Skeletal muscle is a secretory organ. LIF, IL4, IL6, IL7 and IL15 promote muscle hypertrophy. Myostatin inhibits muscle hypertrophy and exercise provokes the release of a myostatin inhibitor, follistatin, from the liver. BDNF and IL6 are involved in AMPKmediated fat oxidation and IL6 enhances insulin stimulated glucose uptake. IL6 appears to have systemic effects on the liver and adipose tissue and increases insulin secretion via upregulation of GLP1. IGF1 and FGF2 are involved in bone formation, and follistatinrelated protein 1 improves endothelial function and revascularization of ischaemic vessels. Irisin has a role in browning of white adipose tissue.

are low, which suggests that IL-6 works as an energy sensor.5760 This idea is supported by numerous studies showing that glucose ingestion during exercise attenuates the exercise-induced increase in plasma IL-623 and inhibits the IL-6 release from contracting skeletal muscle in humans.23,61 Skeletal muscle is a tissue that is capable of altering the type and amount of protein in response to regular exercise. Exercise-induced adaptation in skeletal muscle increases pre-exercise skeletal muscle glycogen content, enhances activity of key enzymes involved in -oxidation, increases sensitivity of adipose tissue to epinephrinestimulated lipolysis, and increases oxidation of intramuscular triglycerides. As a consequence, the trained skeletal muscle can utilize fat as a substrate and is less dependent on plasma glucose and muscle glycogen as substrates during exercise.23,62 Several epidemiological studies have reported a negative association between the amount of regular physical activity and resting plasma IL-6 levels: the more physical activity, the lower the basal plasma IL-6 level.37 By contrast, high basal plasma levels of IL-6 are closely associated with physical inactivity and the metabolic syndrome. Moreover, basal levels of IL-6 are reduced after endurance training.37 In addition, the exercise-induced increase in plasma IL-6 and muscular IL6 mRNA levels is diminished by endurance training.63 Interestingly, although resting plasma IL-6 levels are downregulated by endurance training, the resting muscular expression of IL-6R is upregulated. In response to endurance training, the basal IL6R mRNA content in muscle is increased by ~100%.55 Thus, with exercise
ADVANCE ONLINE PUBLICATION | 3

REVIEWS
training, the downregulation of IL-6 is partially counteracted by an enhanced expression of IL-6R, such that sensitivity to IL-6 is increased. Our hypothesis is that muscle disuse leads to IL-6 resistance and elevated circulating levels of IL-6, which would parallel the insulin resistance that is accompanied by hyperinsulinaemia and the leptin resistance that reflects chronic, high circulating levels ofleptin. Acute treatment of rat L6 muscle cells invitro with IL-6 increased basal glucose uptake and the translocation of the glucose transporter GLUT4.64 Moreover, IL-6 increased insulin-stimulated glucose uptake in muscle cells invitro. The findings appear to be of clinical relevance, as infusion of recombinant human IL-6 into healthy individuals during a hyperinsulinaemic, euglycaemic clamp procedure enhanced whole-body insulin sensitivity. Treatment with IL-6 increased the glucose infusion rate without having any influence on the total suppression of endogenous glucose production.64 Invitro, the effects of IL-6 on glucose uptake appeared to be mediated by activation of AMPK, as the results were abolished in cells infected with a recombinant adenovirus ex pressing dominant-negative AMPK.64 Several studies have reported that IL-6 might also increase intramyocellular 6466 or whole-body 67 fatty acid oxidation via AMPK.64,68 IL-6 acutely mediates signalling through the receptor IL-6R (also known as glycoprotein 130 [gp130]) and exhibits many leptin-like actions, such as activation of AMPK6971 and insulin signalling.72 Interestingly, IL-6 knockout mice develop late-onset obesity and glucose intolerance,73 which supports the notion that IL-6 exerts beneficial effects on metabolism. Importantly, IL-6 is a myokine with cardinal biological activity, as it contributes to hepatic glucose production during exercise.74 The mechanisms that mediate the tightly controlled production and clearance of glucose during muscular work are unclear. An unidentified work factor has been suggested to exist that influences the contraction-induced increase in endogenous glucose production. Acute administration of recombinant human IL-6 infused at physiological concentrations into resting human individuals has no effect on whole-body glucose disposal, glucose uptake or endogenous glucose production.66,75,76 By contrast, IL-6 contributes to the contractioninduced increase in endogenous glucose production. When recombinant human IL-6 was infused into healthy volunteers during low-intensity exercise, to mimic the circulating concentration of IL-6 observed during highintensity exercise, the glucose output was as high as during high-intensity exercise. The study demonstrated a direct muscleliver crosstalk. IL-6 appeared to have a role in endogenous glucose production during exercise in humans; however, its action on the liver was dependent on a yet unidentified muscle contraction-induced factor.74 Infusion of recombinant human IL-6 into healthy individuals also caused an increase in lipolysis in the absence of hypertriglyceridaemia or changes in catecholamines, glucagon, insulin or any adverse effects.66,67,76 These findings combined with cell culture experiments show that IL-6 has direct effects on both lipolysis and
4 | ADVANCE ONLINE PUBLICATION
2012 Macmillan Publishers Limited. All rights reserved

fatty acid oxidation and identify IL-6 as a lipolytic factor.66 Infusion of IL-6 into healthy humans at a physiological level primarily stimulates lipolysis in skeletal muscle, whereas adipose tissue is unaffected.77 IL-6 probably also mediates some of the antiinflammatory and immunoregulatory effects of exercise.78,79 IL-6 inhibits lipopolysaccharide-induced TNF production in cultured human monocytes,80 and levels of TNF are markedly elevated in mice treated with an anti-IL-6 antibody and in IL6 knockout mice,81 which suggests that circulating IL-6 is involved in the regulation of TNF levels. In addition, both recombinant human IL-6 infusion and exercise inhibit the endotoxin-induced increase in circulating levels of TNF in healthy individuals.82 The anti-inflammatory effects of IL-6 are also demonstrated by IL-6 stimulating the production of the classic anti-inflammatory cytokines IL-1ra and IL-10.83

IL7
Haugen etal. identified IL-7 as a myokine.42 IL-7 is a cytokine that is required for T-cell and B-cell development, whereas possible biological functions of IL-7 in nonimmune cells have not been explored. IL7 mRNA and protein were detected in conditioned media from primary cultures of human myotubes as well as inside the myotube. The amount of IL-7 in the medium increased with incubation time.42 Incubations with recombinant IL-7 during differentiation of human myoblasts induced a reduction in mRNA levels of the terminal myogenic markers myosin heavy chain 2 and myogenin. This finding suggests that IL-7 might act on satellite cells, which are small mononuclear progenitor cells found in mature muscle. Haugen and co-workers also demonstrated that the muscular expression of IL7 mRNA was increased several fold in biopsy samples from resting vastus lateralis and trapezius muscles taken from male individuals undergoing a strength training program.42

IL8 and CXCL1


IL-8 belongs to a large family of chemokines. This myokine is mainly produced by macrophages and endothelial cells and exerts marked chemotactic activity towards leukocytes, in addition to being an angiogenic factor. The murine chemokine CXC ligand 1 (CXCL-1) shares the highest sequence homology with human CXCL-1, but it is often mentioned as the functional homologue to human IL-8.84 CXCL-1 and IL-8 possess neutrophil chemoattractant activity. In addition, they are involved in the processes of angiogenesis.85 The capacity of IL-8 to induce angiogenesis is distinct from its capacity to induce inflammation.86 IL-8 binds to the CXC chemokine receptors CXCR-1 and CXCR-2.87 CXCR-2 is expressed by human microvascular endothelial cells and is the receptor responsible for IL-8-induced angiogenesis.88 The production of different CXC chemokines is induced by IL-6.89 The role of exercise and IL-6 in the regulation of murine CXCL-1 has, therefore, been studied.90 Following a single bout of exercise, CXCL1 mRNA levels increased in serum, muscle and liver. The exercise-induced regulation of liver CXCL1 mRNA expression was completely
www.nature.com/nrendo

REVIEWS
blunted in IL6 knockout mice. When IL-6 was overexpressed in murine muscles, a marked increase in serum CXCL-1 and liver CXCL1 mRNA expression occurred. These data demonstrate a robust muscleliver crosstalk during exercise, in which exercise-induced IL-6 production stimulates the liver to produce CXCL-1. The study found a higher expression of CXCL-1 in liver compared with muscle. However, muscular IL8 mRNA levels are enhanced by exercise,91 and IL-8 is released by human primary cultured myotubes.42 The biological role of both liver-derived and muscle-cell-derived IL-8 remains to bedefined.

Other myokines
Generation of skeletal-muscle-specific, inducible Akt1 transgenic mice, which can reversibly grow functional typeII muscle fibres by switching Akt1 signalling on and off, has enabled identification of novel muscle-secreted factors.101 They include follistatin-related protein 1, which seems to have cardioprotective effects,102,103 and FGF-21.104 Follistatin-related protein 1 promotes endothelial cell function and revascularization in ischemic tissue through a mechanism dependent on nitric oxide synthase.102 Studies in humans support the notion that FGF-21 is a myokine that is upregulated by insulin.105 Other musclecell-derived proteins include BDNF,106 calprotectin,107 erythropoietin 108 and IL-4, which enhances muscle regeneration by stimulating the fusion of myoblasts withmyotubes.109 A role for myokines in musclebone interactions has been suggested on the basis that two well-known osteogenic factors, IGF-1 and FGF-2, are abundant in homogenized muscle tissue and secreted from cultured myotubes invitro.110 The receptors for these growth factors are localized to the periosteum at the muscle bone interface,111 which suggests that IGF-1 and FGF-2 might be involved in musclebone crosstalk. In the past few years, irisin was discovered as a myokine that drives brown-fat-like development of white adipose tissue. PGC-1 expression in muscle was shown to increase the expression of FNDC5, which encodes a membrane protein that is cleaved and secreted as irisin.112 Mice were injected with FNDC5-expressing adenoviral particles, whereby irisin levels increased by threefold to fourfold, resulting in the induction of a programme of development of brown-fat-like cells of white adipose tissue and a concomitant increase in energy expenditure. Basal plasma levels of irisin increased in response to 10weeks of regular exercise in humans, which suggests that irisin has a role in training adaptation to exercise.112

LIF
LIF was identified in 1988 as a protein secreted from ascites tumour cells.92 This myokine belongs to the IL-6 cytokine superfamily, which consists of structurally and functionally related proteins named neuropoietins (or gp130 cytokines).93 LIF has multiple biological functions, being a stimulus for platelet formation, proliferation of haematopoietic cells, bone formation, neural survival and formation, and acute phase production by hepatocytes.94 Moreover, LIF induces satellite cell proliferation, which is considered essential for proper muscle hypertrophy andregeneration.95 LIF mRNA expression is induced in human skeletal muscle following resistance exercise, and LIF protein is secreted from electrically stimulated human cultured myotubes. 96 In addition, chemical inhibition of the signalling molecules PI3K and mTor and small interfering RNA (siRNA) knockdown (silencing) of Akt1 were independently sufficient to downregulate LIF. Moreover, LIF stimulated proliferation of human myoblasts and induced expression of jun-B and c-Myc in human myotubes. By contrast, siRNA knockdown of the LIF receptor resulted in a reduction of proliferation. These findings suggest that LIF is a contraction-induced myokine that exerts its effects in an autocrine and/or paracrine fashion to promote satellite cell proliferation.

Bioinformatics and proteomic studies


Up to 10% of encoded human genes have the capacity to express proteins that could potentially be secreted from cells. Such secreted factors may be involved in the cellcell communication that is required for homeostasis in a complex organism.22 A number of research groups have contributed to the identification of the muscle cell secretome. Bortoluzzi etal. screened 6,255 products of genes expressed in normal human skeletal muscle.17 They reported that the resulting putative skeletal muscle secretome consisted of 319 proteins, including 78 still uncharacterized proteins. Yoon etal. studied differentiated L6 rat skeletal muscle cells and identified a total of 254 proteins, among which 153 were classified as secretory proteins.18 In a study by Henningsen etal., a quantitative proteomics platform was used to search for factors secreted during the differentiation of murine C2C12 skeletal muscle cells. In total, 635 secreted proteins were identified.19 The members of the CC chemokine family of proteins showed a highly distinct pattern of secretion during differentiation.113 Norheim and co-investigators demonstrated that a total of 236 proteins were detected
ADVANCE ONLINE PUBLICATION | 5

IL15
IL-15 belongs to the IL-2 superfamily and is expressed in human skeletal muscle. In addition to its anabolic effects on skeletal muscle, IL-15 may have a role in lipid metabolism.97 IL-15 decreases lipid deposition in preadipocytes and decreases the mass of white adipose tissue.98,99 Consequently, a negative association has been found in humans between plasma IL-15 levels and total adipose tissue mass, trunk adipose tissue mass and percent adipose tissue mass.100 Physical inactivity leads to loss of muscle mass and accumulation of visceral fat,5 and some evidence points to IL-15 being involved in the regulation of abdominal adiposity. In support, we demonstrated a decrease in visceral fat mass, but not subcutaneous fat mass, when IL-15 was overexpressed in murine muscle.100 Although, IL-15 has been suggested to play a part in muscleadipose tissue crosstalk, secretion of IL-15 from muscle cells has not been described and it is, therefore, premature to classify IL-15 as a true myokine.
NATURE REVIEWS | ENDOCRINOLOGY

2012 Macmillan Publishers Limited. All rights reserved

REVIEWS
Physical inactivity

Myokines: clinical aspects


Many of the myokines identified exert their effects within the muscle itself. Thus, myostatin, LIF, IL-4, IL-6, IL-7 and IL-15 are involved in the regulation of muscle hypertrophy and myogenesis. BDNF and IL-6 are involved in AMPK-mediated fat oxidation and IL-8 (or CXCL-1) might be involved in mediating training-induced angiogenesis. However, IL-6 also appears to mediate systemic effects, including effects on the liver, adipose tissue and the immune system. Follistatin-related protein 1 has been identified to play a role in promoting endothelial cell function and revascularization under conditions of ischaemic stress, and IGF-1 and FGF-2 appear to be involved in musclebone crosstalk. The myokine field is new and, to date, most of the human studies have focused on the biological role of IL-6. The finding that muscle-derived IL-6 has several beneficial metabolic effects suggests that it has a role in the association between a physically inactive lifestyle and an increased risk of chronic diseases. Sadagurski and colleagues have demonstrated that transgenic mice with sustained elevated circulating levels of human IL-6 display enhanced central leptin action and improved nutrient homeostasis that leads to protection from dietinduced obesity.115 In addition, Wunderlich etal.116 have shown that IL-6 signalling is required for normal liver metabolism in mice. Of note, ciliary neurotrophic factor is a member of the IL-6 family of cytokines and also improves metabolic homeostasis in mice when insulin resistance is induced either from a high-fat diet 70 or lipid infusion.117 Interestingly, a ciliary neurotrophic factor variant, axokine, was in clinical trials for the treatment of T2DM, but failed to be approved owing to a lack of a neutralizing effect of the antibody. 118 Nonetheless, the finding that exercise has multiple beneficial effects, which may be mediated by myokines, suggests that there might be therapeutic potential in myokine research. Apart from the effects of myokines on peripheral insulin sensitivity via the activation of AMPK, evidence is emerging that myokines might also play a major part in pancreatic -cell metabolism and insulin secretion. Bouzakri etal. showed that human myotubes express and release a different panel of myokines depending on their insulin sensitivity, with each panel exerting differential effects on cells. These preliminary data suggest a new route of communication between skeletal muscle and cells, which is modulated by insulin resistance.119 Moreover, Ellingsgaard and colleagues showed that whereas exercise increased glucose tolerance in normal mice, this phenomenon did not occur in mice with global IL-6 deficiency.120 Using several elegant models, the researchers were able to show that the exercise-induced GLP-1 response was dependent upon muscle-derived IL-6. Hence, IL-6 mediates crosstalk between two different insulin-sensitive tissues, the gut and pancreatic islets, in order to adapt to changes in insulin demand by increasing GLP-1 secretion. Epidemiological studies suggest an increased risk of breast cancer in women with a sedentary lifestyle, whereas regular physical activity protects against the
www.nature.com/nrendo

Loss of muscle mass and abdominal adiposity

Macrophage in ltration of visceral adipose tissue

Chronic systemic in ammation

Insulin resistance, atherosclerosis, tumour growth, impaired bone formation

Type 2 diabetes mellitus, cardiovascular diseases, cancer, osteoporosis

Figure 3 | Links between physical inactivity and disease development. Loss of muscle mass and accumulation of visceral adipose tissue are general consequences of physical inactivity. Abdominal adiposity stimulates macrophage infiltration of adipose tissue, whereby a network of inflammatory pathways is activated. Inflammation promotes development of insulin resistance, atherosclerosis, neurodegeneration, tumour growth and impaired bone formation and, consequently, the development of a myriad of chronic diseases. During physical activity, muscles release myokines, which stimulate muscle growth and hypertrophy, increase fat oxidation, enhance insulin sensitivity and induce anti inflammatory actions.
Physical activity

Myokines

Muscle hypertrophy (myostatin, LIF, IL-4, IL-6, IL-7, IL-15) Adipose tissue oxidation (IL-6, BDNF) Insulin sensitivity (IL-6) Osteogenesis (IGF-1, FGF-2) Anti-in ammation (IL-6) Anti-tumour defence (unidenti ed secreted factor(s)) Pancreas function (unidenti ed secreted factor(s)) Browning of fat (Irisin)

Decreased risk of chronic diseases and premature mortality

Figure 4 | The finding that muscle produces and releases myokines provides a conceptual basis for understanding some of the molecular mechanisms that link physical activity to protection against premature mortality.

by proteome analysis in medium conditioned by cultured human myotubes.114 Reverse transcription PCR analyses showed that 15 of the secreted muscle proteins had markedly enhanced mRNA expression in the vastus lateralis and/or trapezius muscles after 11weeks of strength training among healthy volunteers.
6 | ADVANCE ONLINE PUBLICATION

2012 Macmillan Publishers Limited. All rights reserved

REVIEWS
development of breast cancer.121 Evidence exists that one or several myokines might mediate some of the inhibitory effects of exercise on mammary cancer cell proliferation and a possible candidate is oncostatin M, a member of the IL-6 superfamily.122 In summary, physical inactivity and muscle disuse lead to loss of muscle mass and accumulation of visceral adipose tissue and consequently to the activation of a network of inflammatory pathways, which promote development of insulin resistance, atherosclerosis, neurodegeneration and tumour growth and, thereby, promote the development of a cluster of chronic diseases (Figure3).21 By contrast, the finding that muscles produce and release myokines provides a molecular basis for understanding how physical activity could protect against premature mortality (Figure4). muscle as a secretory organ provides a conceptual basis for understanding how muscles communicate with other organs such as adipose tissue, liver, pancreas, bone and brain. Physical inactivity or muscle disuse potentially leads to an altered or impaired myokine response and/ or resistance to the effects of myokines, which explains why lack of physical activity increases the risk of a whole network of diseases, including cardiovascular diseases, T2DM, cancer and osteoporosis.
Review criteria
A search for original articles published between 1970 and January 2012 and focusing on skeletal muscle as a secretory organ was performed in MEDLINE and PubMed. The search terms used were muscle, exercise, physical activity endocrine, cytokine, myokine, inflammation and insulin resistance. All articles identified were Englishlanguage, fulltext papers. Reference lists of identified articles were also searched for further papers.

Conclusions
Given that muscle is the largest organ in the body, the identification of the muscle secretome could set a new agenda for the scientific community. To view skeletal
1. Cook, K.S. etal. Adipsin: a circulating serine protease homolog secreted by adipose tissue and sciatic nerve. Science 237, 402405 (1987). 2. Zhang, Y. etal. Positional cloning of the mouse obese gene and its human homologue. Nature 372, 425432 (1994). 3. Scherer, P Adipose tissue: from lipid storage .E. compartment to endocrine organ. Diabetes 55, 15371545 (2006). 4. Shetty, S., Kusminski, C.M. & Scherer, P .E. Adiponectin in health and disease: evaluation of adiponectintargeted drug development strategies. Trends Pharmacol. Sci. 30, 234239 (2009). 5. Olsen, R.H., KroghMadsen, R., Thomsen, C., Booth, F.W. & Pedersen, B.K. Metabolic responses to reduced daily steps in healthy nonexercising men. JAMA 299, 12611263 (2008). 6. KroghMadsen, R. etal. A 2wk reduction of ambulatory activity attenuates peripheral insulin sensitivity. J. Appl. Physiol. 108, 10341040 (2010). 7. Booth, F.W., Chakravarthy, M.V., Gordon, S.E. & Spangenburg, E.E. Waging war on physical inactivity: using modern molecular ammunition against an ancient enemy. J. Appl. Physiol. 93, 330 (2002). 8. Tuomilehto, J. etal. Prevention of type2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N.Engl. J. Med. 344, 13431350 (2001). 9. Nocon, M. etal. Association of physical activity with allcause and cardiovascular mortality: asystematic review and metaanalysis. Eur. J. Cardiovasc. Prev. Rehabil. 15, 239246 (2008). 10. Wolin, K.Y., Yan, Y., Colditz, G.A. & Lee, I.M. Physical activity and colon cancer prevention: ametaanalysis. Br. J. Cancer 100, 611616 (2009). 11. Monninkhof, E.M. etal. Physical activity and breast cancer: a systematic review. Epidemiology 18, 137157 (2007). 12. Borer, K.T. Physical activity in the prevention and amelioration of osteoporosis in women: interaction of mechanical, hormonal and dietary factors. Sports Med. 35, 779830 (2005).

13. Goldstein, M.S. Humoral nature of the hypoglycemic factor of muscular work. Diabetes 10, 232234 (1961). 14. Pedersen, B.K. etal. Searching for the exercise factor: is IL6 a candidate? J. Muscle Res. Cell Motil. 24, 113119 (2003). 15. Kjaer, M. etal. Hormonal and metabolic responses to electrically induced cycling during epidural anesthesia in humans. J. Appl. Physiol. 80, 21562162 (1996). 16. Mohr, T. etal. Longterm adaptation to electrically induced cycle training in severe spinal cord injured individuals. Spinal Cord 35, 116 (1997). 17. Bortoluzzi, S., Scannapieco, P Cestaro, A., ., Danieli, G.A. & Schiaffino, S. Computational reconstruction of the human skeletal muscle secretome. Proteins 62, 776792 (2006). 18. Yoon, J.H. etal. Comparative proteomic analysis of the insulininduced L6 myotube secretome. Proteomics 9, 5160 (2009). 19. Henningsen, J., Rigbolt, K.T., Blagoev, B., Pedersen, B.K. & Kratchmarova, I. Dynamics of the skeletal muscle secretome during myoblast differentiation. Mol. Cell. Proteomics 9, 24822496 (2010). 20. Long, A., Donelson, R. & Fung, T. Does it matter which exercise? A randomized control trial of exercise for low back pain. Spine (Phila. Pa. 1976) 29, 25932602 (2004). 21. Pedersen, B.K. The diseasome of physical inactivityand the role of myokines in musclefat cross talk. J. Physiol. 587, 55595568 (2009). 22. Walsh, K. Adipokines, myokines and cardiovascular disease. Circ. J. 73, 1318 (2009). 23. Pedersen, B.K. & Febbraio, M.A. Muscle as an endocrine organ: focus on musclederived interleukin6. Physiol. Rev. 88, 13791406 (2008). 24. Pedersen, B.K., Akerstrm, T.C., Nielsen, A.R. & Fischer, C.P Role of myokines in exercise and . metabolism. J. Appl. Physiol. 103, 10931098 (2007). 25. Pedersen, B.K. The antiinflammatory effect of exercise: its role in diabetes and cardiovascular disease control. Essays Biochem. 42, 105117 (2006).

26. McPherron, A.C., Lawler, A.M. & Lee, S.J. Regulation of skeletal muscle mass in mice by a new TGFbeta superfamily member. Nature 387, 8390 (1997). 27. Rodgers, B.D. & Garikipati, D.K. Clinical, agricultural, and evolutionary biology of myostatin: a comparative review. Endocr. Rev. 29, 513534 (2008). 28. Allen, D.L. etal. Myostatin, activin receptor IIb, and follistatinlike3 gene expression are altered in adipose tissue and skeletal muscle of obese mice. Am. J. Physiol. Endocrinol. Metab. 294, E918E927 (2008). 29. Feldman, B.J., Streeper, R.S., Farese, R.V. Jr & Yamamoto, K.R. Myostatin modulates adipogenesis to generate adipocytes with favorable metabolic effects. Proc. Natl Acad. Sci. USA 103, 1567515680 (2006). 30. Guo, T. etal. Myostatin inhibition in muscle, but not adipose tissue, decreases fat mass and improves insulin sensitivity. PLoS ONE 4, e4937 (2009). 31. Zhao, B., Wall, R.J. & Yang, J. Transgenic expression of myostatin propeptide prevents dietinduced obesity and insulin resistance. Biochem. Biophys. Res. Commun. 337, 248255 (2005). 32. Lin, J. etal. Myostatin knockout in mice increases myogenesis and decreases adipogenesis. Biochem. Biophys. Res. Commun. 291, 701706 (2002). 33. McPherron, A.C. & Lee, S.J. Suppression of body fat accumulation in myostatindeficient mice. J. Clin. Invest. 109, 595601 (2002). 34. Allen, D.L., Hittel, D.S. & McPherron, A.C. Expression and function of myostatin in obesity, diabetes, and exercise adaptation. Med. Sci. Sports Exerc. 43, 18281835 (2011). 35. Hittel, D.S., Berggren, J.R., Shearer, J., Boyle,K. & Houmard, J.A. Increased secretion and expression of myostatin in skeletal muscle from extremely obese women. Diabetes 58, 3038 (2009). 36. Hansen, J. etal. Exercise induces a marked increase in plasma follistatin: evidence that follistatin is a contractioninduced hepatokine. Endocrinology 152, 164171 (2011).

NATURE REVIEWS | ENDOCRINOLOGY


2012 Macmillan Publishers Limited. All rights reserved

ADVANCE ONLINE PUBLICATION | 7

REVIEWS
37. Fischer, C.P Interleukin6 in acute exercise and . training: what is the biological relevance? Exerc. Immunol. Rev. 12, 633 (2006). 38. De Rossi, M., Bernasconi, P Baggi, F., de Waal ., Malefyt, R. & Mantegazza, R. Cytokines and chemokines are both expressed by human myoblasts: possible relevance for the immune pathogenesis of muscle inflammation. Int. Immunol. 12, 13291335 (2000). 39. Bartoccioni, E., Michaelis, D. & Hohlfeld, R. Constitutive and cytokineinduced production of interleukin6 by human myoblasts. Immunol. Lett. 42, 135138 (1994). 40. Keller, C., Hellsten, Y., Steensberg, A. & Pedersen, B.K. Differential regulation of IL6 and TNFalpha via calcineurin in human skeletal muscle cells. Cytokine 36, 141147 (2006). 41. Serrano, A.L., BaezaRaja, B., Perdiguero, E., Jard, M. & MuozCnoves, P Interleukin6 is an . essential regulator of satellite cellmediated skeletal muscle hypertrophy. Cell Metab. 7, 3344 (2008). 42. Haugen, F. etal. IL7 is expressed and secreted by human skeletal muscle cells. Am. J. Physiol. Cell Physiol. 298, C807C816 (2010). 43. Green, C.J., Pedersen, M., Pedersen, B.K. & Scheele, C. Elevated NFB activation is conserved in human myocytes cultured from obese type2 diabetic patients and attenuated by AMPactivated protein kinase. Diabetes 60, 28102819 (2011). 44. Nieman, D.C. etal. Influence of mode and carbohydrate on the cytokine response to heavy exertion. Med. Sci. Sports Exerc. 30, 671678 (1998). 45. NehlsenCannarella, S.L. etal. Carbohydrate and the cytokine response to 2.5 h of running. J.Appl. Physiol. 82, 16621667 (1997). 46. Ullum, H. etal. Bicycle exercise enhances plasma IL6 but does not change IL1 alpha, IL1 beta, IL6, or TNFalpha premRNA in BMNC. J.Appl. Physiol. 77, 9397 (1994). 47. Starkie, R.L., Angus, D.J., Rolland, J., Hargreaves, M. & Febbraio, M. Effect of prolonged, submaximal exercise and carbohydrate ingestion on monocyte intracellular cytokine production in humans. J. Physiol. 528, 647655 (2000). 48. Starkie, R.L., Rolland, J., Angus, D.J., Anderson,M.J. & Febbraio, M.A. Circulating monocyes are not the source of elevations in plasma IL6 and TNFalpha levels after prolonged running. Am. J. Physiol. Cell Physiol. 280, C769C774 (2001). 49. Febbraio, M.A. etal. Hepatosplanchnic clearance of interleukin6 in humans during exercise. Am. J. Physiol. Endocrinol. Metab. 285, E397E402 (2003). 50. Keller, C. etal. Transcriptional activation of the IL6 gene in human contracting skeletal muscle: influence of muscle glycogen content. FASEB J. 15, 27482750 (2001). 51. Steensberg, A. etal. IL6 and TNFalpha expression in, and release from, contracting human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 283, E1272E1278 (2002). 52. Hiscock, N., Chan, M.H., Bisucci, T., Darby, I.A. & Febbraio, M.A. Skeletal myocytes are a source of interleukin6 mRNA expression and protein release during contraction: evidence of fiber type specificity. FASEB J. 18, 992994 (2004). 53. Rosendal, L. etal. Increase in interstitial interleukin6 of human skeletal muscle with repetitive lowforce exercise. J. Appl. Physiol. 98, 477481 (2005). 54. Steensberg, A. etal. Production of interleukin6 in contracting human skeletal muscles can account for the exerciseinduced increase in plasma interleukin6. J. Physiol. 529, 237242 (2000). Keller, C. etal. Effect of exercise, training, and glycogen availability on IL6 receptor expression in human skeletal muscle. J. Appl. Physiol. 99, 20752079 (2005). Steensberg, A. etal. Interleukin6 production in contracting human skeletal muscle is influenced by preexercise muscle glycogen content. J.Physiol. 537, 633639 (2001). Pedersen, B.K. Muscular IL6 and its role as an energy sensor. Med. Sci. Sports Exerc. 44, 392396 (2012). Ruderman, N.B. etal. Interleukin6 regulation of AMPactivated protein kinase. Potential role in the systemic response to exercise and prevention of the metabolic syndrome. Diabetes 55 (Suppl. 2), S48S54 (2006). Pedersen, B.K. etal. The metabolic role of IL6 produced during exercise: is IL6 an exercise factor? Proc. Nutr. Soc. 63, 263267 (2004). Hoene, M. & Weigert, C. The role of interleukin6 in insulin resistance, body fat distribution and energy balance. Obes. Rev. 9, 2029 (2008). Febbraio, M.A. etal. Glucose ingestion attenuates interleukin6 release from contracting skeletal muscle in humans. J. Physiol. 549, 607612 (2003). Phillips, S.M. etal. Effects of training duration on substrate turnover and oxidation during exercise. J. Appl. Physiol. 81, 21822191 (1996). Fischer, C.P etal. Endurance training reduces . the contractioninduced interleukin6 mRNA expression in human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 287, E1189E1194 (2004). Carey, A.L. etal. Interleukin6 increases insulin stimulated glucose disposal in humans and glucose uptake and fatty acid oxidation invitro via AMPactivated protein kinase. Diabetes 55, 26882697 (2006). Bruce, C.R. & Dyck, D.J. Cytokine regulation of skeletal muscle fatty acid metabolism: effect of interleukin6 and tumor necrosis factoralpha. Am. J. Physiol. Endocrinol. Metab. 287, E616E621 (2004). Petersen, E.W. etal. Acute IL6 treatment increases fatty acid turnover in elderly humans invivo and in tissue culture invitro. Am. J. Physiol. 288, E155E162 (2005). van Hall, G. etal. Interleukin6 stimulates lipolysis and fat oxidation in humans. J. Clin. Endocrinol. Metab. 88, 30053010 (2003). Kahn, B.B., Alquier, T., Carling, D. & Hardie, D.G. AMPactivated protein kinase: ancient energy gauge provides clues to modern understanding of metabolism. Cell Metab. 1, 1525 (2005). Minokoshi, Y. etal. Leptin stimulates fattyacid oxidation by activating AMPactivated protein kinase. Nature 415, 339343 (2002). Watt, M.J. etal. CNTF reverses obesityinduced insulin resistance by activating skeletal muscle AMPK. Nat. Med. 12, 541548 (2006). Steinberg, G.R., Rush, J.W. & Dyck, D.J. AMPK expression and phosphorylation are increased in rodent muscle after chronic leptin treatment. Am. J. Physiol. Endocrinol. Metab. 284, E648E654 (2003). Steinberg, G.R., Watt, M.J. & Febbraio, M.A. Cytokine Regulation of AMPK signalling. Front. Biosci. 14, 19021916 (2009). Wallenius, V. etal. Interleukin6deficient mice develop matureonset obesity. Nat. Med. 8, 7579 (2002). Febbraio, M.A., Hiscock, N., Sacchetti, M., Fischer, C.P & Pedersen, B.K. Interleukin6 is a . novel factor mediating glucose homeostasis during skeletal muscle contraction. Diabetes 53, 16431648 (2004). Steensberg, A. etal. Acute interleukin6 administration does not impair muscle glucose uptake or wholebody glucose disposal in healthy humans. J. Physiol. 548, 631638 (2003). Lyngs, D., Simonsen, L. & Blow, J. Interleukin6 production in human subcutaneous abdominal adipose tissue: the effect of exercise. J. Physiol. 543, 373378 (2002). Wolsk, E., Mygind, H., Grndahl, T.S., Pedersen,B.K. & van Hall, G. IL6 selectively stimulates fat metabolism in human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 299, E832E840 (2010). Nielsen, S. & Pedersen, B.K. Skeletal muscle as an immunogenic organ. Curr. Opin. Pharmacol. 8, 346351 (2008). Petersen, A.M. & Pedersen, B.K. The anti inflammatory effect of exercise. J. Appl. Physiol. 98, 11541162 (2005). Schindler, R. etal. Correlations and interactions in the production of interleukin6 (IL6), IL1, and tumor necrosis factor (TNF) in human blood mononuclear cells: IL6 suppresses IL1 and TNF. Blood 75, 4047 (1990). Mizuhara, H. etal. Tcell activationassociated hepatic injury: mediation by tumor necrosis factors and protection by interleukin 6. J. Exp. Med. 179, 15291537 (1994). Starkie, R., Ostrowski, S.R., Jauffred, S., Febbraio, M. & Pedersen, B.K. Exercise and IL6 infusion inhibit endotoxininduced TNFalpha production in humans. FASEB J. 17, 884886 (2003). Steensberg, A., Fischer, C.P., Keller, C., Mller,K. & Pedersen, B.K. IL6 enhances plasma IL1ra, IL10, and cortisol in humans. Am. J. Physiol. Endocrinol. Metab. 285, E433E437 (2003). Rubio, N. & SanzRodriguez, F. Induction of the CXCL1 (KC) chemokine in mouse astrocytes by infection with the murine encephalomyelitis virus of Theiler. Virology 358, 98108 (2007). Lira, S.A. etal. Expression of the chemokine N51/KC in the thymus and epidermis of transgenic mice results in marked infiltration of a single class of inflammatory cells. J. Exp. Med. 180, 20392048 (1994). Keane, M.P etal. The CXC chemokines, IL8 . and IP10, regulate angiogenic activity in idiopathic pulmonary fibrosis. J. Immunol. 159, 14371443 (1997). Belperio, J.A. etal. CXC chemokines in angiogenesis. J. Leukoc. Biol. 68, 18 (2000). Addison, C.L. etal. The CXC chemokine receptor2, CXCR2, is the putative receptor for ELR+ CXC chemokineinduced angiogenic activity. J. Immunol. 165, 52695277 (2000). Tseng, Y.L., Wu, M.H., Yang, H.C., Wang, C.Y. & Lin, C.F. Autocrine IL6 regulates GROalpha production in thymic epithelial cells. Cytokine 51, 195201 (2010). Pedersen, L. etal. Exerciseinduced liver CXCL1 expression is linked to muscle derived interleukin6 expression. J. Physiol. 589, 14091420 (2011). Nieman, D.C. etal. Muscle cytokine mRNA changes after 2.5 h of cycling: influence of carbohydrate. Med. Sci. Sports Exerc. 37, 12831290 (2005). Hilton, D.J., Nicola, N.A. & Metcalf, D. Purification of a murine leukemia inhibitory factor from Krebs ascites cells. Anal. Biochem. 173, 359367 (1988). Heinrich, P .C., Behrmann, I., MllerNewen, G., Schaper, F. & Graeve, L. Interleukin6type

55.

75.

56.

76.

57.

77.

58.

78.

59.

79.

60.

80.

61.

81.

62.

82.

63.

83.

64.

84.

65.

85.

66.

86.

67.

87. 88.

68.

69.

89.

70.

90.

71.

91.

72.

92.

73.

74.

93.

8 | ADVANCE ONLINE PUBLICATION


2012 Macmillan Publishers Limited. All rights reserved

www.nature.com/nrendo

REVIEWS
cytokine signalling through the gp130/Jak/STAT pathway. Biochem. J. 334, 297314 (1998). 94. Metcalf, D. The unsolved enigmas of leukemia inhibitory factor. Stem Cells 21, 514 (2003). 95. Broholm, C. & Pedersen, B.K. Leukaemia inhibitory factoran exerciseinduced myokine. Exerc. Immunol. Rev. 16, 7785 (2010). 96. Broholm, C. etal. Exercise induces expression of leukaemia inhibitory factor in human skeletal muscle. J. Physiol. 586, 21952201 (2008). 97. Nielsen, A.R. & Pedersen, B.K. The biological roles of exerciseinduced cytokines: IL6, IL8, and IL15. Appl. Physiol. Nutr. Metab. 32, 833839 (2007). 98. Carb, N. etal. Interleukin15 mediates reciprocal regulation of adipose and muscle mass: a potential role in body weight control. Biochim. Biophys. Acta 1526, 1724 (2001). 99. Quinn, L.S., StraitBodey, L., Anderson, B.G., Argils, J.M. & Havel, P Interleukin15 .J. stimulates adiponectin secretion by 3T3L1 adipocytes: evidence for a skeletal muscletofat signaling pathway. Cell Biol. Int. 29, 449457 (2005). 100. Nielsen, A.R. etal. Association between IL15 and obesity: IL15 as a potential regulator of fat mass. J. Clin. Endocrinol. Metab. 93, 44864493 (2008). 101. Izumiya, Y. etal. Fast/Glycolytic muscle fiber growth reduces fat mass and improves metabolic parameters in obese mice. Cell Metab. 7, 159172 (2008). 102. Ouchi, N. etal. Follistatinlike 1, a secreted muscle protein, promotes endothelial cell function and revascularization in ischemic tissue through a nitricoxide synthasedependent mechanism. J. Biol. Chem. 283, 3280232811 (2008). 103. Oshima, Y. etal. Follistatinlike 1 is an Akt regulated cardioprotective factor that is secreted by the heart. Circulation 117, 30993108 (2008). 104. Izumiya, Y. etal. FGF21 is an Aktregulated myokine. FEBS Lett. 582, 38053810 (2008). 105. Hojman, P etal. Fibroblast growth factor21 is . induced in human skeletal muscles by hyperinsulinemia. Diabetes 58, 27972801 (2009). 106. Pedersen, B.K. etal. Role of exerciseinduced brainderived neurotrophic factor production in the regulation of energy homeostasis in mammals. Exp. Physiol. 94, 11531160 (2009). 107. Mortensen, O.H. etal. Calprotectin is released from human skeletal muscle tissue during exercise. J. Physiol. 586, 35513562 (2008). 108. Hojman, P etal. Erythropoietin overexpression . protects against dietinduced obesity in mice through increased fat oxidation in muscles. PLoSONE 4, e5894 (2009). 109. Horsley, V., Jansen, K.M., Mills, S.T. & Pavlath,G.K. IL4 acts as a myoblast recruitment factor during mammalian muscle growth. Cell 113, 483494 (2003). 110. Hamrick, M.W. A role for myokines in muscle bone interactions. Exerc. Sport Sci. Rev. 39, 4347 (2011). 111. Hamrick, M.W., McNeil, P & Patterson, S.L. .L. Role of musclederived growth factors in bone formation. J. Musculoskelet. Neuronal Interact. 10, 6470 (2010). 112. Bostrm, P etal. A PGC1 dependent myokine . that drives brownfatlike development of white fat and thermogenesis. Nature 481, 463468 (2012). 113. Henningsen, J., Pedersen, B.K. & Kratchmarova,I. Quantitative analysis of the secretion of the MCP family of chemokines by muscle cells. Mol. Biosyst. 7, 311321 (2011). 114. Norheim, F. etal. Proteomic identification of secreted proteins from human skeletal muscle cells and expression in response to strength training. Am. J. Physiol. Endocrinol. Metab. 301, E1013E1021 (2011). 115.Sadagurski, M. etal. Human IL6 enhances leptin action in mice. Diabetologia 53, 525535 (2010). 116.Wunderlich, F.T. etal. Interleukin6 signaling in liverparenchymal cells suppresses hepatic inflammation and improves systemic insulin action. Cell Metab. 12, 237249 (2010). 117.Watt, M.J., Hevener, A., Lancaster, G.I. & Febbraio, M.A. Ciliary neurotrophic factor prevents acute lipidinduced insulin resistance by attenuating ceramide accumulation and phosphorylation of cJun Nterminal kinase in peripheral tissues. Endocrinology 147, 20772085 (2006). 118.Ettinger, M.P etal. Recombinant variant of ciliary . neurotrophic factor for weight loss in obese adults: a randomized, doseranging study. JAMA 289, 18261832 (2003). 119.Bouzakri, K. etal. Bimodal effect on pancreatic [beta]cells of secretory products from normal or insulinresistant human skeletal muscle. Diabetes 60, 11111121 (2011). 120.Ellingsgaard, H. etal. Interleukin6 enhances insulin secretion by increasing glucagonlike peptide1 secretion from L cells and alpha cells. Nat. Med. 17, 14811489 (2011). 121.Food, Nutrition, Physical Activity and the Prevention of Cancer. World Cancer Research Fund and American Institute of Cancer Research. Ref. Type: Report (2007). 122.Hojman, P. etal. Exerciseinduced muscle derived cytokines inhibit mammary cancer cell growth. Am. J. Physiol. Endocrinol. Metab. 301, E504E510 (2011). Acknowledgements B. K. Pedersen is supported by a grant from the Danish National Research Foundation (#0251255). M. A. Febbraio is supported by grants from the National Health and Medical Research Council (NHMRC), The Diabetes Australia Research Trust and the Victorian Government Operational Infrastructure Support Program. Author contributions Both authors contributed equally to all aspects of thearticle.

NATURE REVIEWS | ENDOCRINOLOGY


2012 Macmillan Publishers Limited. All rights reserved

ADVANCE ONLINE PUBLICATION | 9