Skeletal muscle metabolism is highly dependent on mitochondrial function, with impaired mitochondrial biogenesis associated with the development of metabolic diseases such as insulin resistance and type 2 diabetes. Mitochondria display substantial plasticity in skeletal muscle, and are highly sensitive to levels of physical activity. It is thought that physical activity promotes mitochondrial biogenesis in skeletal muscle through increased expression of genes encoded in both the nuclear and the mitochondrial genome; however, how this process is co-ordinated at the cellular level is poorly understood. Nuclear receptors (NRs) are key signalling proteins capable of integrating environmental factors and mitochondrial function, thereby providing a potential link between exercise and mitochondrial biogenesis. The aim of this review is to highlight the function of NRs in skeletal muscle mitochondrial biogenesis and discuss the therapeutic potential of NRs for the management and treatment of chronic metabolic disease.
Non-communicable diseases (NCDs) are considered a leading cause of global mortality , with the prevalence of metabolic NCDs rising exponentially in the past decade [2,3]. Impairments in skeletal muscle substrate use, termed ‘metabolic flexibility’, and blunted insulin sensitivity have emerged as central factors in the development of the metabolic syndrome [4,5]. These metabolic processes rely, to a great extent, on the extensive mitochondrial reticulum within skeletal muscle, which acts as a site of substrate convergence, ultimately leading to the generation of ATP. Accordingly, diseases such as insulin resistance and type 2 diabetes have been associated with reduced mitochondrial content and gene expression , indicating that alterations in mitochondrial function may play a critical role in the development and progression of metabolic NCDs .
Although the question of whether mitochondrial dysfunction causes insulin resistance is still hotly debated [7–9], it is clear that physical inactivity leads to reduced mitochondrial function and impaired skeletal muscle metabolism, and therefore represents a central risk factor for numerous diseases . Physical activity is innately linked to optimal mitochondrial function, due to skeletal muscle contraction regulating a wide spectrum of signal pathways that promote the expression of both nuclear- and mitochondrially encoded genes controlling energy metabolism . The mechanism by which exercise regulates mitochondrial gene expression involves a number of transcription factors (TFs) and transcriptional co-regulators . These families of proteins are able to directly or indirectly interact with DNA at specific regions, regulating gene transcription. Among the TFs, nuclear receptors (NRs) are attractive therapeutic targets because their activity can be modulated by exercise and endogenous or synthetic ligands [11,12]. The NR family of TFs is made up of 48 members, but at present only a few have been shown to directly regulate mitochondrial physiology. Therefore, the aim of this review is to highlight the regulatory role of NRs in skeletal muscle mitochondrial function and discuss their potential as therapeutic targets for the treatment of metabolic NCDs.
The protein structure of this family of TFs contains three characteristic domains: the activation function 1 (AF-1), DNA-binding domain (DBD) and ligand-binding domain (LBD; covering the AF-2); of these the first two are the most conserved . Despite these conserved structures, only 24 NRs can act as genuine ligand receptors, whereas the remaining ones are classified as orphan NRs . This protein class exhibits all the structural and functional characteristics of conventional NRs, including a putative LBD; however, no endogenous ligands have been identified . An important mechanistic feature of NRs is the interaction with transcriptional co-activators (CoAs) and co-repressors (CoRs). Thus, in the absence of a ligand or the presence of an antagonist, NRs recruit CoRs, whereas ligands induce the recruitment of CoAs and epigenetic modifications, such as histone acetylation and DNA demethylation, which ultimately activate gene transcription (Figure 1). The CoA, peroxisome proliferator activated receptor-γ co-activator-1α (PGC-1α), has been extensively shown to be induced by exercise and to enhance skeletal muscle mitochondrial function, whereas the CoRs, nuclear receptor co-repressor 1 (NCoR1) and receptor-interacting protein 140 (RIP140), counteract these effects [14–16]. Further discussion about CoA/R regulation and the effect on mitochondrial biogenesis is beyond the aim of this review. Therefore, we instead refer the reader to the excellent reviews on this topic [14–16].
Ligand-dependent regulation of NR activity
Thyroid hormone receptors
Thyroid hormone (triiodothyronine or T3) is well known to modulate skeletal muscle biology, regulating processes such as development and regeneration . T3 is the main ligand of the thyroid hormone receptor (TR) family of NRs, which is composed of TRα (NR1A1) and TRβ (NR1A2) . TR expression is extensively regulated at the transcriptional and post-transcriptional levels, generating different transcript and isoform variants, among which p43 has been identified as a mitochondrial TR . This implies an important function of TRs in mitochondrial physiology, because they can regulate the expression of both nuclear- and mitochondrially encoded genes.
The role of TRs in skeletal muscle metabolism has been revealed via T4 (thyroxine, a precursor of T3) administration and thyroidectomy, which have been found to increase and decrease mitochondrial respiration, respectively . It is interesting that mitochondrial size and number are increased in both experimental conditions mentioned above, although thyroidectomy appears to induce structural abnormalities linked to lower mitochondrial function . Similarly, T3 administration increases the expression and activity of mitochondrial enzymes (e.g. citrate synthase) in skeletal muscle, concomitant with the up-regulation of PGC-1α [20–22]. T3 also enhances maximal mitochondrial oxidative capacity, although these adaptations seem to be predominantly found in oxidative muscle . The function of PGC-1α in T3-mediated mitochondrial biogenesis has not been fully elucidated, but the fact that PGC-1α increases TRβ transcriptional activity supports a role for PGC-1α in the T3 mode of action . It is of interest that administration of a TRβ-specific agonist significantly prevents body and liver fat accumulation and alleviates the development of insulin resistance in rats fed a high-fat diet (HFD) . Similar effects have been reported in mice, although it is thought that an HFD attenuates the beneficial effects of this agonist [25,26]. Although the molecular mechanisms mediating these effects remain elusive, it seems that this agonist primarily targets brown adipose tissue [24,25].
As mentioned above, p43 is specifically localized in the mitochondrial matrix and can directly regulate the transcription of genes encoded in the mtDNA . In skeletal muscle, p43 over-expression efficiently increases the expression and activity of mitochondrially encoded genes controlling oxidative phosphorylation . Moreover, skeletal muscles from p43 muscle-specific transgenic (mTg) mice were found to exhibit higher mitochondrial number and oxidative capacity, consistent with higher PGC-1α and mitochondrial transcription factor A (TFAM) expression . Surprisingly, the effects of p43 over-expression on skeletal muscle mitochondrial biogenesis disappear during ageing, whereas skeletal muscle wasting is increased in p43 mTg mice . These results have been further validated through use of whole-body p43 knockout (KO) mice, which display lower skeletal muscle oxidative capacity and higher muscle mass . Paradoxically, however, despite lower mitochondrial function, p43 KO mice exhibit higher exercise performance and lower respiratory exchange ratio (RER) during low-intensity exercise . Thus, it seems that, although TRs exert positive effects on skeletal muscle mitochondrial function, their potential as therapeutic targets needs further validation.
Peroxisome proliferator-activated receptors
The peroxisome proliferator-activated receptor (PPAR) family of NRs has been extensively shown to regulate energy metabolism, with fatty acids and their derivate metabolites acting as the main endogenous ligands . Such ligands can activate PPARα (NR1C1), PPARδ (NR1C2; also known as PPARβ) and PPARγ (NR1C3), which represent the three members of this family . Although PPARs can directly bind the DNA at PPAR-response elements (PPREs), the retinoid X receptor (RXR) is a key partner mediating this process . Consequently, either PPARs or RXR ligands can activate the PPAR/RXR heterodimer, a process that is characterized by an exchange of CoRs by CoAs . The function of PPARs in skeletal muscle was initially delineated via pharmacological approaches, showing that activation of PPARα induced the up-regulation of uncoupling protein 3 (UCP3) and pyruvate dehydrogenase kinase 4 (PDK4) [33,34]. This transcriptional effect was found to be coupled to higher fatty acid oxidation (FAO) and a lower TAG content in skeletal muscle . These data suggest a therapeutic potential of PPARα in the context of metabolic NCDs. Accordingly, administration of a dual PPARα/γ agonist has been shown to improve insulin sensitivity in different rodent models of obesity and diabetes [36,37].
Similarly, selective activation of PPARα via fenofibrate administration can promote similar effects on metabolic health . Muoio et al.  found, however, that genetic deletion of PPARα did not impair basal levels of FAO in skeletal muscle cells. Actually, under basal conditions, PPARα KO mice contain normal levels of skeletal muscle UCP3 and PDK4 mRNA, whereas the positive effects of exercise and fasting are not impaired . Despite the beneficial effects of a PPARα agonist on metabolic health , it has also been suggested that activation of this NR, specifically in skeletal muscle, may induce detrimental effects. PPARα mTg mice have been shown to exhibit insulin resistance both under basal conditions and after HFD feeding . Furthermore, although whole-body PPARα KO mice are more susceptible to HFD-induced obesity, these animals are protected against the development of insulin resistance . Therefore, it appears that selective PPARα activation in skeletal muscle can lead to negative effects on metabolic health.
The studies discussed above lead to the identification of PPARδ as the main isoform expressed in skeletal muscle, whereas PPARγ was found to have no regulatory role in skeletal muscle FAO [39,41]. In accordance with this, stimulation of cultured muscle cells by a selective PPARδ agonist increases the expression of genes regulating β-oxidation, thus leading to higher levels of FAO [39,42,43]. This effect on gene transcription is further enhanced by co-administration of PPARδ and RXR agonists, with PGC-1α acting as a key CoA . Importantly, the PPARδ agonist GW501516 has been reported to reduce gains in body weight induced by HFD feeding, whereas lipid accumulation in muscle and liver is reduced . The effects attained by this pharmacological approach are thought to mediate the improvement in insulin sensitivity observed in mice fed an HFD [42,45]. In genetic mouse models of obesity and diabetes (ob/ob and db/db mice), GW501516 produces similar improvement in glucose homoeostasis in a dose-dependent manner [42,43,46], although administration of GW501516 or the dual PPARα/δ agonist GFT505 significantly improved blood lipid profiles and insulin tolerance in obese humans [47,48]. In line with these pharmacological studies, mTg mice over-expressing a constitutive active form of PPARδ (VP16- PPARδ) are protected against HFD-induced obesity, whereas muscle oxidative metabolism and exercise performance are increased under basal conditions . However, the phenotype of mTg mice over-expressing wild-type PPARδ is considerably milder, because this mouse model does not show changes in oxygen consumption (V̇O2) or RER during treadmill running [49,50].
Moreover, PPARδ muscle-specific knockout (mKO) mice show minor impairments in skeletal muscle oxidative metabolism under basal conditions [51,52]. Finally, although skeletal muscle deletion of PPARδ exacerbates the negative effects of HFD and ageing on whole-body metabolism , it does not affect the positive effects of PGC-1α on skeletal muscle oxidative metabolism [43,52]. These data therefore suggest that, although not essential, PPARδ is an important regulator of skeletal muscle mitochondrial function. Importantly, it should be emphasized that ligands and/or CoRs seem to be critical for the full activation of PPARδ. In fact, increasing plasma free fatty acids efficiently induce skeletal muscle mitochondrial biogenesis, which coincides with a higher recruitment of PPARδ to the CPT1 promoter .
ORPHAN NUCLEAR RECEPTORS
Oestrogen-related receptors (ERRs) are the founding members of this class of NRs . Three genes encoding ERRα (NR3B1), ERRβ (NR3B2) and ERRγ (NR3B3) comprise this family, with ERRα and ERRγ highly expressed in oxidative skeletal muscle . Both ERRα and ERRγ have been shown to be versatile regulators of skeletal muscle physiology, modulating processes such as angiogenesis, regeneration, glycolysis and oxidative metabolism .
The main mechanism by which ERRα promotes mitochondrial biogenesis involves the recruitment of the transcriptional CoA PGC-1α, which is considered a key ‘protein ligand’ . Under basal conditions, in the absence of such a protein ligand, ERRα transcriptional activity appears to be suppressed . PGC-1α can directly interact with ERRα and, thus, increase its transcriptional activity [57,58]. Hence, changes in PGC-1α expression and/or activity appear to be an important physiological regulator of ERRα. Microarray studies in mouse skeletal muscle and C2C12 muscle cells have revealed an enrichment of oestrogen-related response elements (ERREs) among genes induced by PGC-1α over-expression [59,60]. Interestingly, the ERRα promoter also contains ERREs and its expression is increased by the PGC-1α/ERRα complex, as revealed via a reporter gene assay in C2C12 myoblasts . Genetic and pharmacological inhibition of ERRα reduces the positive effects of PGC-1α on gene transcription, mitochondrial content and cellular respiration, further validating the biological relevance of this transcriptional complex [57,60,61]. In vivo studies using whole-body ERRα KO mice have revealed its involvement in whole-body metabolism, mainly through the regulation of brown adipose tissue . Notably, Perry et al.  have recently shown that ERRα KO mice exhibit lower exercise performance and spontaneous activity. These physiological perturbations were associated with higher reliance on carbohydrate metabolism, lower expression of mitochondrial enzymes and higher lactate production after acute exercise . Hence, skeletal muscle ERRα can significantly boost mitochondrial capacity, but its function in whole-body metabolism and muscle physiology requires further examination, possibly via muscle-specific deletion or over-expression mouse models.
PGC-1α has also been shown to act as a protein ligand of ERRγ which through direct interaction can enhance its transcriptional activity . ERRγ expression levels are higher in oxidative compared with glycolytic skeletal muscles, although its expression is induced by exercise [64,65]. The function of this orphan NR was initially investigated by genetic approaches, demonstrating that mTg mice over-expressing VP16-ERRγ or wild-type ERRγ were characterized by a fibre-type switch to a slow oxidative phenotype . In line with this fibre-type switch, ERRγ mTg mice exhibit higher number and size of skeletal muscle mitochondria, whereas microarray analysis revealed an up-regulation of mitochondrial enzymes . Functionally, this phenotype was reflected in improved exercise performance, higher V̇O2peak and lower exercise RER . Besides confirming these results, subsequent studies showed enhanced oxidative capacity and parallel reductions in glycolytic rate in C2C12 muscle cells over-expressing ERRγ . Genetic deletion of skeletal muscle ERRγ reduces mitochondrial content and oxidative capacity, increasing oxidative stress and impairing muscle cell differentiation . Therefore, together with ERRα, ERRγ represents an attractive target to enhance skeletal muscle oxidative metabolism.
Nuclear receptor subfamily 4, group A
These orphan NRs have been classified as early response genes due to their sensitivity to cellular stress . In fact, stimuli such as β-adrenergic activation, motor nerve stimulation and exercise increase the expression of the NR subfamily 4, group A (NR4A) members Nur77 (NR4A1), Nurr1 (NR4A2) and Nor1 (NR4A3) in skeletal muscle [68–72]. This suggests that NR4As might be physiological regulators of skeletal muscle remodelling after metabolic and/or mechanical stress. Skeletal muscle NR4As have also been suggested to play a role in the aetiology of metabolic diseases. Conditions associated with lower mitochondrial function, such as denervation, limb immobilization and insulin resistance, significantly decrease skeletal muscle Nur77 and/or Nor1 transcript levels [71,73–75]. It appears therefore that changes in NR4A expression levels might modulate skeletal muscle plasticity in both health and disease.
The involvement of Nur77 in muscle metabolism was initially observed in C2C12 muscle cells, which exhibited impaired lipid and glucose metabolism after Nur77 knockdown (KD) . The opposite effects were induced by Nur77 over-expression, which increased the expression of genes controlling glucose metabolism and, consequently, enhanced glucose transport . The generation of Nur77 mTg mice has further uncovered its positive effects on skeletal muscle biology, with these animals showing a switch toward a slow oxidative phenotype, improved mitochondrial oxidative capacity and ex vivo muscle fatigue resistance . In the context of diet-induced obesity, Nur77 KO mice exhibit higher susceptibility to pathological adaptations such as skeletal muscle insulin resistance and hepatic steatosis . It is of interest, however, that it has been reported that the negative effects of HFD on glucose tolerance are not prevented in Nur77 mTg mice , indicating that its therapeutic potential requires further investigation.
Although the function of Nurr1 in skeletal muscle has not been explored, the third member of the NR4A subfamily, Nor1 appears to have an important function. Similar to Nur77, siRNA-mediated KD of Nor1 impairs oxidative metabolism in C2C12 cells, mainly affecting FAO . Interestingly, VP16-Nor1 mTg mice show an oxidative phenotype, which is characterized by higher expression and activity of mitochondrial enzymes [80,81]. Further support for a regulatory role in mitochondrial biogenesis is given by increased mitochondrial numbers and mtDNA levels in skeletal muscle of VP16-Nor1 mTg mice . Importantly, this molecular remodelling of skeletal muscle phenotype is accompanied by higher V̇O2 and exercise performance . Besides the regulation of oxidative metabolism, Nor1 has also been implicated in the control of glucose homoeostasis, with 6-mercaptopurine increasing both basal and insulin-mediated glucose uptake in a Nor1-dependent manner  via enhanced glucose transporter type 4 (GLUT4) translocation to the plasma membrane . These data imply a central function of Nor1 in skeletal muscle mitochondrial function and glucose metabolism, although its therapeutic potential for metabolic NCDs remains unclear, given that, although Nor1 can exert a significant anti-obesogenic effect, it does not protect against obesity-induced glucose and insulin intolerance .
Chicken ovalbumin, upstream promoter-transcription factors
The chicken ovalbumin, upstream promoter-transcription factor (COUP-TF)-I (NR2F1) and COUP-TFII (NR2F2) form this family of orphan NRs and, although both are expressed in skeletal muscle tissue, COUP-TFII represents the predominant isoform . COUP-TFII expression has been shown to increase during C2C12 muscle cell proliferation, although it decreases during differentiation . In line with this phasic regulation, COUP-TFII can directly repress skeletal muscle differentiation . With regard to the metabolic function of COUP-TFs, it was initially reported that siRNA-mediated KD of COUP-TFI/II impairs β-oxidation in C2C12 myotubes, whereas the expression of genes controlling fatty acid metabolism is significantly decreased . Despite lower oxidative capacity for fatty acids, COUP-TF KD increased ATP levels, potentially, due to lower UCP1 levels . These studies suggest a potential function of COUP-TFs in skeletal muscle metabolism, but future studies are required to define its relevance and therapeutic potential.
The Rev-erb family of orphan NRs comprises two different isoforms: Rev-erb-α (NR1D1) and Rev-erb-β (NR1D2); both have been implicated in the control of energy metabolism and circadian rhythm . Rev-erb-α deletion has been shown to alter skeletal muscle fibre-type composition . Similar to other NRs, Rev-erb-α is mainly expressed in oxidative muscle, where it exerts a positive effect on exercise performance and maximal oxidative capacity . In line with this physiological function, genetic deletion of Rev-erb-α strongly decreases skeletal muscle mitochondrial content and function, whereas its over-expression induces the opposite effect . It is of interest that both pharmacological and genetic disruption of AMP-activated protein kinase (AMPK) signalling has been reported to impair the positive effects of Rev-erb-α on mitochondrial function . Moreover, both Rev-erb-α and PGC-1α induce a reciprocal up-regulation when over-expressed in skeletal muscle cells [88,89], suggesting a possible Rev-erb-α/PGC-1α axis in the control of skeletal muscle metabolism. Furthermore, the Rev-erb-α/β agonist SR9009 can increase energy expenditure and reduce fat mass in mice, through increased expression of genes controlling glucose and oxidative metabolism in skeletal muscle . Furthermore, treatment of obese mice with SR9009 significantly reduces fat mass content and improves blood lipid profiles, although its effects on insulin sensitivity remain to be defined . Collectively these data suggest the potential for Rev-erb-α/β as a pharmacological target for the treatment of metabolic NCDs, but further studies are required to validate their clinical relevance.
EXERCISE AS A MODULATOR OF NR ACTIVITY
It is now evident that modulating the activity of NRs and/or their transcriptional co-regulators represents an efficient way to improve skeletal muscle mitochondrial function and, potentially, metabolic health. Exercise training has been extensively shown to improve oxidative metabolism at the physiological and molecular levels, with skeletal muscle as a key target tissue . In this context, it is evident that the NRs described above are attractive therapeutic targets to improve mitochondrial function. However, a limited number of NRs have been studied in the context of exercise, of which the PPARs, ERRs and NR4As are the best characterized.
Exercise and PPARs
Pharmacological studies have revealed PPARs as promising targets for the treatment of metabolic NCDs. However, such studies have not defined the specific function of skeletal muscle PPARs as mediators of these therapeutic effects. Another stimulus that seems to specifically regulate skeletal muscle PPARs is exercise, which can promote an increase in PPARδ mRNA levels irrespective of blood glucose and free fatty acid levels . It is of interest that, although moderate-intensity exercise did not affect PPARα and PPARγ expression levels in human skeletal muscle , maximal exercise promotes a significant increase in all PPARs . Thus, it seems that PPAR expression in skeletal muscle is regulated in an exercise intensity-dependent manner. This effect seems to be mediated by epigenetic modifications because the PPARδ promoter shows lower levels of DNA methylation after acute exercise, which consequently also promotes gene transcription in an intensity-dependent process . Additional stimuli such as ligands appear to be critical to fully activate PPARδ and are likely to have a synergistic effect with exercise on PPARδ biological activity. In agreement, Narkar et al.  demonstrated that the PPARδ agonist GW1516 potentiates the effects of exercise on skeletal muscle mitochondrial biogenesis and exercise performance. Importantly, it appears that such potentiation can also be achieved via non-pharmacological approaches. Depletion of skeletal muscle glycogen has recently been demonstrated to significantly enhance the effects of exercise on PPARδ recruitment to the CPT-1 promoter, concomitant with higher PGC-1α and AMPKα nuclear abundance . These effects were further validated via reporter gene assays in cultured muscle cells, showing that low glycogen or glucose levels enhance PPARδ activity . The study mentioned above is particularly relevant, because it directly validates the effects of exercise on PPARδ transcriptional activity, beyond its effects on mRNA or protein levels.
Exercise and ERRs
Although ERRs are key regulators of skeletal muscle mitochondrial function, the mechanism by which exercise modulates their transcriptional activity remains elusive. Nevertheless, it has been reported that acute exercise induces an increase in ERRα mRNA that parallels changes in PGC-1α mRNA in human skeletal muscle [95,96]. Importantly, this synchronized induction of ERRα and PGC-1α mRNA has also been found at the protein level, with exercise increasing protein abundance in human skeletal muscle over a period of 14 days of training . Similar results have been observed in mice subject to chronic exercise, where co-ordinated up-regulation of ERRα, ERRγ and PGC-1α is observed in both skeletal muscle and the hippocampus [64,98]. It is interesting that the ERR/PGC-1α transcriptional axis has recently been found to control the expression of PGC-1- and ERR-induced regulator in muscle 1 (Perm1), which on its induction drives an oxidative muscle phenotype .
Exercise and NR4As
It has been reported that rats exhibiting high running capacity contain elevated protein levels of Nur77 and Nor1 in both oxidative and glycolytic skeletal muscle . Importantly, 6 weeks of exercise training was sufficient to increase Nur77 protein levels in skeletal muscle of rats with low running capacity . In human skeletal muscle it has been shown that NR4As (mainly Nur77 and Nor1) are among the highest up-regulated genes in response to acute endurance and resistance exercise [69,72]. This effect has been observed in both the exercising and non-exercising leg, suggesting the key function of systemic factors released in response to exercise . Similar results have been observed in rat skeletal muscle after a single bout of swimming, treadmill running, electrical stimulation or AMPK activation, with Nur77 and Nor1 mRNA showing a significant increase immediately after stimulation . Considering that skeletal muscle Nur77 and Nor1 were decreased by limb immobilization, it seems that motor nerve stimulation is a central signal controlling NR4A transcriptional networks .
CONCLUSIONS AND FUTURE PERSPECTIVES
Impaired skeletal muscle mitochondrial function has been implicated in the development of prevalent metabolic NCDs such as insulin resistance and type 2 diabetes . In this context, a variety of small molecules acting as NR agonists have emerged as ‘exercise mimetics’, because they prevent the development of such diseases through stimulation of residual mitochondrial function . Through their prescription, these molecules have revealed NRs as promising therapeutic targets, but it remains unclear whether their effects are mediated by skeletal muscle NRs. Actually, despite improving mitochondrial function under basal conditions, muscle-specific over-expression of some NRs (e.g. Nur77 and Nor1) does not lead to therapeutic effects. More importantly, some of the detrimental effects of ageing and HFD feeding on skeletal muscle metabolism and function are even enhanced by NR over-expression (e.g. p43 and PPARα). This indicates that non-muscle organs such as liver and adipose tissue are most probably mediating some of the beneficial effects of endogenous and synthetic NR activators.
It should be reinforced, however, that exercise appears to be an important stimulus regulating NR activity in skeletal muscle, suggesting that the therapeutic potential of this TF family might be context dependent. Supporting this idea, PGC-1α mTg mice show improved oxidative metabolism, but higher susceptibility to NCDs unless they are concomitantly exercised [102,103]. However, a recent study suggests that PGC-1α over-expression in skeletal muscle does not potentiate the therapeutic effects of exercise and/or caloric restriction . Similar exercise–NR interaction has been reported for PPARδ, which on activation with a specific agonist enhances the effects of exercise training on skeletal muscle and whole-body metabolism . These data suggest that enhanced skeletal muscle mitochondrial function is not sufficient to improve metabolic health, and additional stimuli induced by exercise are critical to unmasking the therapeutic potential of NRs. An efficient approach to treat metabolic NCDs would be combining NR activation and exercise training; this would enhance mitochondrial oxidative capacity and simultaneously increase metabolic flux due to higher substrate utilization (Figure 2). It seems likely that such strategies would further increase metabolic flexibility and insulin sensitivity.
Therapeutic strategies modulating nuclear receptors and mitochondrial function
Collectively, these data support NRs as important modulators of skeletal muscle mitochondrial function. However, it is evident that there is an urgent need to define the optimal strategies and contexts in which NRs can be manipulated to improve human health. It will be also important to uncover the molecular mechanism by which exercise regulates NR activity in skeletal muscle. To date, most studies have focused almost exclusively on changes in NR mRNA or protein levels, whereas NR activity is mainly determined by their cellular localization, promoter recruitment and interaction with transcriptional CoRs. Accordingly, chromatin immunoprecipitation and chromatin immunoprecipitation coupled to sequencing may be key to understanding the mechanisms controlling NR function in response to environmental stimuli such as muscle contraction. Finally, it will be relevant to identify natural compounds modulating NR activity, allowing the development of non-pharmacological strategies to potentiate the therapeutic effects of exercise training.
This publication was supported in part by a Wellcome Trust's Sir Henry Wellcome Postdoctoral Fellowship (103094/Z/13/Z) to J.P.-S. and a Biotechnology and Biological Sciences Research Council (BBSRC) new investigator award (BB/L023547/1) to A.P.
activation function 1
AMP-activated protein kinase
oestrogen-related response element
fatty acid oxidation
nuclear receptor co-repressor 1
NR subfamily 4, group A
peroxisome proliferator-activated receptor-γ co-activator-1α
peroxisome proliferator-activated receptor-γ
respiratory exchange ratio
receptor-interacting protein 140
retinoid X receptor
triiodothyronine or thyroid hormone
thyroid hormone receptor