Background

Previous studies have demonstrated that chronic administration of β-adrenoceptor (β-AR) agonists (particularly β2-AR agonists) can increase myofibrillar protein content and thus induce skeletal muscle hypertrophy in mammals [1–3]. This β2-AR induced hypertrophy is believed to be a result of decreased proteolysis coupled with increased protein synthesis [4–9]. The ubiquitin-proteasome signaling [8, 10, 11], Ca2+-dependent proteolysis [12] and/or calpain-mediated proteolysis [6, 13, 14] have all been proposed to play a role, however the molecular and cellular pathways altered following β-AR agonist administration remain poorly understood [15].

In addition to hypertrophy, acute exposure of skeletal muscle (and cells) to β-AR agonists has been found to modulate oxidative metabolism, energy expenditure, lipolysis [16–21], glucose transport [22], and glucose oxidation [20]. Skeletal muscle accounts for a large proportion of the body's energy demand and thus plays a pivotal role in insulin sensitivity, blood lipid profile and energy balance [23, 24].

Underscoring the importance of β-AR signaling in regulating metabolism, transgenic mice lacking all three β-ARs are susceptible to diet-induced obesity. These animals lack any diet- and cold-induced thermogenic response, indicating that β-ARs play a major role in energy expenditure [25]. Similar to the molecular mechanisms underlying skeletal muscle hypertrophy, our understanding of the pathways regulating the metabolic response to β-AR stimulation have yet to be fully elucidated.

In the context of β-AR signaling and skeletal muscle, we (and others ([26]) have previously demonstrated that acute β-AR signaling markedly and transiently increased the expression of the NR4A subgroup of orphan nuclear receptors (Nur77, Nurr1 and Nor-1) in skeletal muscle tissue and cultures [27, 28]. The induction of the NR4A subgroup was associated with the modulation of critical metabolic genes and cellular metabolism [28, 29] ([26–29]. Interestingly, in terms of hypertrophy, knockdown of Nor-1 expression, in vitro, resulted in a >65 fold increase in the expression of myostatin, a key negative regulator of muscle hypertrophy. Such studies suggest that the NR4A subgroup may mediate some effects of β-AR signaling in skeletal muscle.

To examine skeletal muscle gene expression following acute β-AR stimulation, we examined global gene expression in mice using Illumina BeadArrays. Spurlock et al. [30] previously examined global gene expression in skeletal muscle 1 and 10 days after administration of the β2-AR agonist clenbuterol. Spurlock et al. identified genes associated with skeletal muscle growth/hypertrophy, including multiple genes associated with proliferation, differentiation, and the recruitment of satellite cells into muscle fibers. Furthermore, they found an increase in the expression of transcriptional and translational initiators responsible for increasing protein synthesis.

We also performed a comprehensive expression analysis of both the acute (1-24 hours) and chronic (1-28 days) effects of β2-AR agonist administration. We identified changes in the expression of mRNAs encoding genes associated with skeletal muscle hypertrophy, myoblast differentiation, metabolism, circadian rhythm, transcription, histones, and oxidative stress that occur within 4 hours and alter signaling pathways responsible for the long-term phenotypic footprint of b2-AR activation.

Results

Acute systemic administration of the β2-adreneroreceptor agonist, formoterol, induces widespread changes in gene expression in skeletal muscle

The entire data set is available via Gene Expression Omnibus (accession number GSE15793). Expression profiling was performed on 16 mice in total using 46K Illumina Sentrix BeadArray chips. Skeletal muscle samples from 16 independent animals were removed at 1 and 4 hours following a single i.p. injection of the β2-AR agonist, formoterol, or saline (vehicle control). Each timepoint consisted of eight animals with four saline and four formoterol treated animals. The tibialis anterior muscle was chosen for all analyses as it contains predominantly type II fibers, and (in rodents) is known to exhibit marked increases in protein content and lean mass (hypertrophy) in response to β-AR agonist administration [31–35].

Using a p value cutoff of p < 0.05 (see methods for full statistical analysis) and a fold change cutoff of 1.85, at one hour following formoterol administration, 23 probes were significantly altered and 112 probes were significantly altered at four hours. Significant annotated genes from both timepoints are shown in table 1. Significant non-annotated genes (Riken cDNAs and hypothetical proteins) are included in Table 2.

Table 1 Significant differentially expressed genes in skeletal muscle after acute systemic administration of β2-AR agonist
Full size table
Table 2 Significant differentially expressed non-annotated genes in skeletal muscle after acute systemic administration of β2-AR agonist
Full size table

Functional categorization of genes differentially regulated by β2-AR activation

Genes presented in table 1 were grouped according to their potential relevant function in skeletal muscle. The potential relevant function is based on the authors' opinion gained from a combination of Illumina Gene Ontology classifications, Ingenuity Pathway Analysis http://www.ingenuity.com, Online Mendelian Inheritance in Man (OMIM; http://www.ncbi.nlm.nih.gov/sites/entrez?db=omim), and AceView http://www.ncbi.nlm.nih.gov/IEB/Research/Acembly/ searches. The Illumina BeadArray expression analysis revealed significant changes in the expression of genes in several functional categories at 1 and 4 hours following formoterol administration, including genes involved in skeletal muscle hypertrophy/growth, myoblast differentiation, metabolism, circadian rhythm, transcription, histones, oxidative stress, angiogenesis, solute carriers, apoptosis, cell cycle, cancer, DNA repair, and the ubiquitin-proteasome system.

Validation of differential gene expression by quantitative RT-PCR

The expression of 16 genes (Stat3, Idb1, Smad1, Smad3, Hk2, Pdk4, Sorbs1, Pgc1α, Lipin1α, FoxO1, Ucp3, Nfil3, Dbp, Nurr1, Crem, and Cebpb) that were identified as differentially expressed by Illumina beadarray analysis (Table 1) and associated with the regulation of skeletal muscle mass, circadian rhythm and metabolism were validated and examined in greater detail following acute (1-24 h) and chronic (1-28 days) formoterol administration via quantitative RT-PCR (qRT-PCR; Figures 1, 2, 3, 4 and 5). All qRT-PCR analyses were performed on an independent/different set of formoterol treated mice (n = 5 per timepoint) than the group used in the Illumina BeadArray study. All 16 gene analyzed by qRT-PCR on independent animals closely mirrored the Illumina changes at both timepoints, highlighting the robust nature of the Illumina platform.

Figure 1
figure 1

Acute systemic administration of formoterol alters the expression of genes associated with muscle growth and differentiation at multiple timepoints. Quantitative RT-PCR was used to assay the expression of A. Stat3, B. Idb1, C. Smad1, D. Acvr2b, E. Smad3, and F.Myostatin mRNAs in tibialis anterior over acute timepoints. Muscles were removed at 1, 4, 8 and 24 hours following a single intraperitoneal injection of formoterol or saline vehicle (NT = no treatment). Results were normalized against 36B4 at each timepoint. Statistical significance was assessed using a one-way ANOVA with Bonferroni's post-test where p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***). Unmarked data points are non-significant.

Full size image
Figure 2
figure 2

Chronic systemic administration of formoterol alters the expression of genes associated with skeletal muscle hypertrophy and myogenesis at multiple timepoints. Quantitative RT-PCR was used to assay the expression of A. Stat3, B. Idb1, C. Smad1, D. Acvr2b, E. Smad3, and F.Myostatin mRNAs in tibialis anterior over chronic timepoints. Muscles were removed at 1, 7 and 28 days following daily intraperitoneal injection of formoterol or saline vehicle (NT = no treatment). Results were normalized against 36B4 at each timepoint. Data are expressed as mean ± SEM (n = 5). Statistical significance was assessed using a one-way ANOVA with Bonferroni's post-test where p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***). Unmarked data points are non-significant. G. Protein levels of Myostatin precursor (pro-Myostatin), Smad3, phosphorylated Smad3, and Gapdh were visualized by Western blotting performed on tibialis anterior muscle following 28 days of formoterol/saline administration in four animals for each treatment. H. Diagrammatic representation of acute and chronic gene expression changes related to skeletal muscle hypertrophy and myogenesis in response to formoterol.

Full size image
Figure 3
figure 3

Acute systemic administration of formoterol alters the expression of genes associated with metabolism. Quantitative RT-PCR was used to assay the expression of A. Hk2, B. Pdk4, C. Sorbs1, D. Pgc1α, E.Lipin1α, F. FoxO1, and G.Ucp3 mRNAs in tibialis anterior over acute timepoints. Muscles were removed at 1, 4, 8 and 24 hours following a single intraperitoneal injection of formoterol or saline vehicle (NT = no treatment). Results were normalized against 36B4 at each timepoint. Statistical significance was assessed using a one-way ANOVA with Bonferroni's post-test where p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***). Unmarked data points are non-significant. G. Diagrammatic representation of acute gene expression changes related to metabolic in skeletal muscle in response to formoterol.

Full size image
Figure 4
figure 4

Chronic systemic administration of formoterol alters the expression of genes associated with metabolism. Quantitative RT-PCR was used to assay the expression of A. Hk2, B. Pdk4, C. Sorbs1, D. Pgc1α, E. Lipin1α, F. FoxO1, and G.Ucp3 mRNAs in tibialis anterior over chronic timepoints. Muscles were removed at 1, 7 and 28 days following daily intraperitoneal injection of formoterol or saline vehicle (NT = no treatment). Results were normalized against 36B4 at each timepoint. Statistical significance was assessed using a one-way ANOVA with Bonferroni's post-test where p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***). Unmarked data points are non-significant.

Full size image
Figure 5
figure 5

Acute and chronic systemic administration of formoterol alters the expression of genes associated with circadian rhythm and transcriptional regulation. Quantitative RT-PCR was used to assay the expression of A. Nfil3, B. Dbp, C. Nurr1, D. Creb, and E.Cebpb mRNAs in tibialis anterior over acute timepoints. Muscles were removed at 1, 4, 8 and 24 hours following a single intraperitoneal injection of formoterol or saline vehicle (NT = no treatment). For chronic timepoints, the expression of F. Nfil3, G. Dbp, H. Nurr1, I. Creb, and J.Cebpb mRNAs were measured in tibialis anterior removed at 1, 7 and 28 days following daily intraperitoneal injection of formoterol or saline vehicle (NT = no treatment). All results were normalized against 36B4 at each timepoint. Data are expressed as mean ± SEM (n = 5). Statistical significance was assessed using a one-way ANOVA with Bonferroni's post-test where p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***). Unmarked data points are non-significant.

Full size image

Formoterol administration alters the expression of genes associated with skeletal muscle hypertrophy and differentiation: attenuation of myostatin signaling

Several differentially expressed genes associated with the regulation of muscle differentiation and mass (identified from the Illumina BeadArray) were examined over an acute and chronic time course of formoterol administration using qRT-PCR. In addition we also examined the expression of myostatin and the myostatin receptor, activin receptor IIB (Acvr2b) that are critical modulators of muscle mass. We examined these genes using qRT-PCR as Acvr2b was down-regulated at 4 hour in the Illumina BeadArray, however it did not pass statistical analysis (data not shown). Tibialis anterior muscle was isolated from groups (n = 5) of male mice, treated with either the specific β2-AR agonist treatment or saline (vehicle control) and assayed at 0, 1, 4, 8 and 24 h post treatment. Significant changes in expression at one or more timepoints were observed in the mRNAs encoding signal transducer and activator of transcription 3 (Stat3; Figure 1A), inhibitor of DNA binding 1 (Idb1; Figure 1B), small mothers against decapentaplegic homolog 1 (Smad1; Figure 1C), Acvr2b (Figure 1D), and small mothers against decapentaplegic homolog 3 (Smad3; Figure 1E). We did not observe any significant changes in myostatin expression, after acute β2-AR agonist treatment (Figure 1F)

To examine chronic changes induced by formoterol administration, qRT-PCR was used to examine expression of genes from chronically treated mice (after 1, 7 and 28 days of agonist treatment). Similar to acute timepoints, tibialis anterior was isolated from groups (n = 5) of male mice, treated daily with either the specific β2-AR agonist formoterol or saline (vehicle control) and assayed at 0, 1, 7 and 28 days of treatment. Chronic formoterol administration was associated with a significant attenuation in the expression of the mRNAs encoding Idb1 (Figure 2B), Smad3 (Figure 2E) and myostatin (Figure 2F) after 7 or 28 days. No significant changes were observed following chronic formoterol administration in the expression of the mRNAs encoding Stat3 (Figure 2A), Smad1 (Figure 2C), and Acvr2b (Figure 2D), despite significant repression following acute formoterol administration.

To examine the effect of chronic β2-AR agonist treatment on critical regulators of the myostatin signaling pathway we examined the levels of the Myostatin precursor (pro-Myostatin), Smad3, phosphorylated Smad3 relative to Gapdh. We assayed levels by Western blotting analysis of tibialis anterior muscle (contralateral to muscle used for qRT-PCR analysis) following 28 days of formoterol/saline administration in four animals for each treatment (Figure 2G). Consistent with the qRT-PCR data, at the protein level, pro-Myostatin appears subtley (but consistently) suppressed following 28 days of formoterol administration. In concordance, the levels of Smad3 phosphorylation following the chronic formoterol administration are also reduced (in 3 out of 4 mice), while total Smad3 appears unchanged.

In summary, β2-adrenergic stimulation mediates changes in the expression of several genes associated with myostatin signaling, and the regulation of muscle mass (Figure 2H).

Formoterol administration alters the expression of genes associated with metabolism and circadian rhythm

Many genes that regulate and/or are directly involved in metabolism are regulated in a circadian manner. The Illumina BeadArray study identified differential expression of several genes involved in these pathways. Consequently, we utilized qRT-PCR to validate the differential expression of these genes after acute or chronic administration of fomoterol (vs vehicle) in tibialis anterior as detailed above. Significant expression changes at one or more timepoints were observed in several genes involved in metabolism, including, hexokinase 2 (Hk2; Figure 3A), pyruvate dehydrogenase kinase 4 (Pdk4; Figure 3B), sorbin and SH3 domain containing 1 (Sorbs1; Figure 3C), PPARγ coactivator 1 alpha (Pgc1α; Figure 3D), Lipin1α (Figure 3E); forkhead box O1 (FoxO1; Figure 3F), and uncoupling protein 3 (Ucp3; Figure 3G). In the context of crosstalk between β-AR signaling and Nor-1 (NR4A3) signaling in skeletal muscle, we have previously identified and examined the induction of Pdk4, Pgc1α, FoxO1, and Lipin1α over following acute β-AR activation [28, 36].

Chronic formoterol administration significantly altered the expression of FoxO1 (Figure 4F) and Ucp3 (Figure 4G) at 7 and 28 days respectively, while Hk2 (Figure 4A), Pdk4 (Figure 4B), Sorbs1 (Figure 4C), Pgc1α (Figure 4D), and Lipin1α (Figure 4E) were not significantly altered.

The expression of two peripheral tissue regulators of circadian rhythm, albumin D-box binding protein (Dbp), and nuclear factor interleukin 3 regulated (Nfil3) were significantly dysregulated by both acute and chronic formoterol administration (Figures 5A, B, F and 5G).

In summary, formoterol administration mediated the significant modulates of several metabolic genes (for example Pgc1α, Lipin1α, Pdk4, FoxO1, Hk2, Ucp3, Sorbs1 etc) associated with the transient induction of oxidative metabolism, particularly following acute stimulation of β-AR's. Interestingly, the expression of these genes was normalized 24 h post treatment, and remained at control levels throughout the 28 days of formoterol administration. In addition, acute and chronic β2-AR agonist treatment significantly regulates the expression of two critical regulators of circadian cycling.

Altered transcriptional regulation following formoterol administration

We have previously demonstrated that β-AR agonists markedly increased the expression of the NR4A subgroup (Nur77, Nurr1 and Nor-1) of nuclear receptor transcription factors in skeletal muscle [27–29]. From the Illumina BeadArray, ten transcription factors (not placed in other categories) were induced by formoterol 1 and 4 h post-administration (Table 1). At one or more timepoints, acute administration of formoterol significantly induced nuclear receptor related 1 protein (Nurr1; Figure 5C), cAMP responsive element modulator (Crem; Figure 5D), and CCAAT/enhancer binding protein β (Cebpb; Figure 5E). In contrast, Crem was the only transcription factor to remain elevated throughout the 28 day formoterol administration period (Figure 5I compared to Figures 5H and 5J).

Discussion

A variety of studies have demonstrated that acute (and chronic) β-AR stimulation in skeletal muscle induces hypertrophy, and modulates oxidative metabolism, mitochondrial parameters, energy expenditure, lipolysis [1, 16–20], glucose transport [37], and glucose oxidation [20].

To examine gene expression associated with these effects, we utilized Illumina BeadArray gene expression profiling to examine global gene expression in skeletal muscle in response to acute systemic administration (1 and 4 hours) of a specific β2-AR agonist (formoterol). In this study we have revealed that β2-AR agonist treatment altered the expression of several genes associated with myostatin signaling, a previously unreported effect of β-AR signaling in skeletal muscle. This is also the first study to demonstrate a β2-AR agonist regulation of circadian rhythm genes, indicating crosstalk between β-AR signaling and circadian cycling in skeletal muscle

Skeletal muscle hypertrophy and myoblast differentiation

Previous studies have demonstrated that systemic administration of β2-AR agonists induces hypertrophy in both skeletal and cardiac muscle [1–3, 38]. Furthermore, β2-AR agonists have been found to prevent or reverse the muscle wasting and weakness associated with numerous conditions [for review see Ryall & Lynch, 2008 [15]]. However, the cellular and molecular mechanisms underlying these changes have yet to be fully elucidated. The scientific literature is not consistent on whether β-AR agonists increase myofibrilar protein mass during hypertrophy by increased protein biosynthesis, decreased proteolysis, or both.

Reeds et al. [4] suggested that β-AR agonists do not increase global protein biosynthesis, however recent reports suggested that β-AR agonists increase the expression of several contractile proteins [30, 39, 40], myogenin [41], and initiators of protein translation [30]. In addition to effects on protein synthesis a number of studies have attributed increased myofibrilar protein content to an inhibition of myofibrilar proteolysis [4, 8, 42], possibly via inhibition of ubiquitin-proteasome mediated degradation [8, 10, 11], Ca2+-dependent proteolysis [12] or calpain-mediated proteolysis [6, 13, 14].

At the molecular level, the gene ankyrin repeat and SOCS box protein Asb15, known to promote protein synthesis and myoblast differentiation, has been found to be induced by β-AR stimulation [43–46]. Similarly, both Igf1 and Igf2 mRNA expression has been observed to increase following chronic systemic administration of β-AR agonists, suggesting the involvement of these growth factors in hypertrophy [40, 47]. Interestingly, in the current study no significant acute changes were observed in Asb15 or Igf1 (data not shown), however Igf2 expression was down-regulated (non-significantly; data not shown) after four hours, a result that is not consistent with these previous studies [40, 47].

We have also included a more detailed table (Table 3), which compares the Illumina BeadArray information in this study to previously published hypotheses on β-AR agonist-induced hypertrophy.

Table 3 Summary of previous reports describing the known molecular effects of β-AR agonist-induced hypertrophy in skeletal muscle
Full size table

In our study, Illumina BeadArray expression profiling analysis revealed several gene expression changes associated with the regulation of skeletal muscle mass and myoblast differentiation. From the array (and qRT-PCR), we observed alterations in Stat3, and Smad3, Acvr2b three genes directly associated with the regulation of muscle hypertrophy. Both Acvr2b (a key myostatin receptor) and Smad3 are downstream mediators of myostatin, a well-characterized negative regulator of muscle mass. With qRT-PCR, we also observed a subtle, but significant attenuation of the mRNAs encoding myostatin in response to chronic β2-AR agonist treatment, however this was not detected by the acute Illumina analysis. This is concordant with the chronic effects of formoterol administration on skeletal muscle, and could provide a partial mechanistic basis for hypertrophy.

While muscle growth from β-AR agonists is associated with hypertrophy, enhancement of myogenesis could lead to proliferation, differentiation, and/or recruitment of satellite cells into muscle fibers to promote muscle growth. We observed significant changes in Itgb1bp3, Smad1, Smad3, FoxO1 (listed under metabolism in Table 1) and, Idb1, genes believed to play an important role in regulating myogenesis. We observed significant changes in Smad1 expression, and while closely related to Smad3, Smad1 has not been associated with skeletal muscle hypertrophy, however it may have a role in the regulation of myogenesis [48]. A complex of Smad1 and 4 has been shown to transactivate Id1 [49], another myogenic gene altered by formoterol administration. Interestingly, we observed a significant repression of FoxO1, a negative regulator of myogenesis, following 28 days of formoterol administration [50–52], thus suggesting β2-AR agonist enhance myogenesis in the context of chronic treatment. Furthermore, myostatin may also be a transcriptionaly regulated by FoxO1 [53], possibly suggesting coordinate regulation of hypertrophy and myogenesis via multiple mechanisms including Fox01/myostatin signaling. Moreover, we have performed Western (immunoblot) analysis on the tibialis anterior muscle from 28 day vehicle (saline) and β2-AR agonist treatment (formoterol) treated mice (n = 4 mice per treatment). This demonstrated that formoterol treatment suppressed expression of pro-Myostatin, and the levels of phospho-Smad3. These observations are in concordance with the attenuation of the myostatin signaling pathway and therefore the hypertrophic phenotype [54, 55].

In summary, formoterol administration induced significant changes in genes associated with skeletal muscle hypertrophy and myogenesis. A broad overview of these expression changes are provided in Figure 2H, notably highlighting possible crosstalk between β-AR signaling and myostatin/Smad3 signaling pathway.

Metabolism

Many previous studies have implicated β-AR signaling in the control of metabolism. Mice lacking all three β-AR are unable to effectively regulate energy expenditure and thus develop obesity on a high-fat diet [25]. In terms of skeletal muscle, β-AR agonist administration has been found to modulate oxidative metabolism, energy expenditure, lipolysis [1, 16–20], glucose transport [22] glucose oxidation [20] and mitochondrial morphology [56]. In the current study we found that formoterol administration significantly induced the expression of 14 genes associated with metabolism and mitochondrial function greater than 1.85 fold (Table 1). These consisted of genes involved in lipid regulation and metabolism, including Lipin1α, FoxO1, scavenger receptor class B member 1 (Scarb1), phosphomevalonate kinase (Pmvk), plasma membrane associated protein (S3-12), peroxisome proliferator-activated receptor γ (Pparγ), and Ucp3. The largest change observed was the induction of Pgc-1α, a key transcriptional regulator of oxidative metabolism and regulator of skeletal muscle fiber type and therefore lipid metabolism. Genes involved in glucose metabolism/storage and insulin signaling were up-regulated, Pdk4, protein phosphatase 1 regulatory subunit 3C (Ppp1r3c), and Hk2.

Several of these genes have been previously identified in our studies following in vitro and in vivo treatment of muscle with a β2-AR agonist. For example, the induction of the mRNAs encoding Pgc1α, Lipin1α, Pdk4 and FoxO1 in skeletal muscle were identified in our study [28] and that of Miura et al. [36].

Interestingly, the expression levels of Pgc1α, Lipin1 and FoxO1 have also been found to be increased following exercise [57], suggesting β2-AR treatment may imitate some functions and/or effects of exercise.

Circadian Rhythm

Skeletal muscle, like most other tissues, is known to have a peripheral circadian clock, characterized by the expression of peripheral clock genes. It is thought that the regulation of these peripheral circadian clocks is ultimately regulated and synchronized by a central circadian clock located in the suprachiasmatic nucleus of the brain, however the exact mechanism of communication to skeletal muscle remains unclear.

As previous studies have implicated β-AR signaling as a mediator of circadian rhythm [58–62], it is of interest that we observed the dysregulation of three peripheral clock genes by formoterol administration (Table 1), including nuclear factor, interleukin 3 (Nfil3), D site albumin promoter binding protein (Dbp), and cryptochrome 2 (Cry2). The observation, that only a subset of peripheral clock genes were dysregulated by formoterol administration suggests regulation of these genes is occurring at an organ specific level and not via direct actions on the central circadian clock since interference to the central circadian clock should result in changes to the majority of peripheral clock genes. Furthermore, as only a small fraction of known circadian-regulated genes in skeletal muscle are being altered in our study (compared to McCarthy et al. [63]), this suggests that these results are valid and not occurring due to minor timing issues that occur during tissue collection.

Adrenergic regulation of the related clock gene cryptochrome 1 (Cry1) has been previously reported in the rat pineal glands [59, 64]. In addition, Cry1 has been shown to be regulated by resistance exercise, a stimulus known to increase sympathetic activity in skeletal muscle [65]. This further underscores the crosstalk between exercise, and β2-AR signaling in the context of peripheral circadian control.

The central role of circadian rhythm in skeletal muscle metabolism is highlighted by the circadian changes in glucose utilization in muscle, and the findings that disruption of skeletal muscle circadian rhythm occurs in diabetic rats [66]. Furthermore, major changes to peripheral clock genes have been found to be increased by altering the activity of AMP kinase, a key metabolic regulator of skeletal muscle [67]. Interestingly, AMPK modulators have been recently described as exercise mimics [68].

Interestingly, a limited number of genes highlighted in this study have been shown to be regulated in the circadian transcriptome of adult mouse skeletal muscle [63]. Cry2, Dbp, Ucp3, Pdk4, Ubc, Pgc1α, and Usp2 have been shown to be regulated in the circadian transcriptome, possibly suggesting that these metabolic transcripts and Usp2 may be regulated by dysregulated clock genes rather than directly by β-AR signaling. However, some genes such as Pgc1α have a well characterized induction pathway involving β-AR signaling independent of circadian regulation [69].

In conclusion, this is the first paper to show regulation of the peripheral circadian regulators in skeletal muscle by β-AR signaling, possibly implicating β-AR (sympathetic) signaling as a pathway that coordinates communication between central and peripheral circadian clocks in skeletal muscle.

Transcription and histones

Spurlock et al. [30] have previously used microarray technology to examine skeletal muscle gene expression following chronic β-AR administration. In this study the authors found an up-regulation of transcriptional and translational initiators responsible for increasing protein synthesis. In this context, we observed the induction of 11 genes (Table 1) noted as transcription factors (although some transcription factors are in other categories eg. FoxO1). The 11 induced transcriptional regulators were FBJ osteosarcoma oncogene (Fos), kruppel-like factors 2 and 4 (Klf2 and 4), cAMP responsive element modulator (Crem), CAAT/enhancer binding protein beta (Cebpb), nuclear receptor related 1 protein (Nurr1), fos-like antigen 2 (Fosl2), V-maf musculoaponeurotic fibrosarcoma oncogene family protein F (Maff), activating transcription factor 3 (Aft3), T-box 3 (Tbx3), transcript variant 2, and LPS-induced TN factor (Litaf).

Interestingly, concomitant with transcriptional induction, our Illumina analysis revealed that β2-AR stimulation significantly inhibited the expression of two histone proteins (also four more histones were present at >2 fold, but these were removed via multiple testing correction), suggesting that β-AR stimulation promote the formation of euchromatin, thus enhancing transcription. This result, coupled with the induction of transcription factors, indicates β-AR signaling may play an important role in regulating skeletal muscle transcription.

Oxidative stress

In our study, many of the largest inductions occurred in genes associated with the response to oxidative stress. Metallothioneins 1 and 2 (Mt1, 2), sulfiredoxin 1 homolog (Npn3), and uncoupling protein 3 (Ucp3; listed under metabolism) were all significantly induced at four hours after formoterol administration. During exercise, reactive oxygen species (ROS) are generated in skeletal muscle [70, 71], and some oxidative damage occurs [72]. Excess ROS may lead to cellular damage, which has been implicated in with the development of insulin resistance [73], and apoptosis of myoblasts [74]. Given that exercise induces β-AR signaling activity in skeletal muscle [75], this activity may provide a mechanism for increased expression of antioxidant genes to restore oxidative homeostasis. This is supported by the work of Mahoney et al [57], who demonstrated that exercise induced a wide range of antioxidant genes, including several metallothioneins. Furthermore, induction of both Ucp2 and Ucp3 has also been shown previously to be induced by β-AR signaling in exercising skeletal muscle [27, 76–78]. Since Ucp3 and Ucp2 have been implicated in the reduction of ROS [79, 80], it is possible that the induction of Ucp3 and Ucp2 expression in skeletal muscle is one mechanism for reducing ROS production during or after exercise.

Conclusion

This study utilized gene expression profiling to examine global gene expression in skeletal muscle following acute and long-term (chronic) β-AR agonist administration. In summary, systemic administration of formoterol had a profound effect on global gene expression in skeletal muscle. With respect to skeletal muscle hypertrophy, formoterol altered the expression of several genes associated with the attenuation of myostatin signaling, highlighting the role of β-AR signaling in the mechanisms regulating skeletal muscle mass

Interestingly, many changes in gene expression with β-AR signaling are similar to findings from other studies examining exercise-related effects in skeletal muscle. For example, the changes in expression of Pgc1α, FoxO1, Ucp3, the NR4A subgroup and severalmetallothioneins are identical in both systems [57, 65]. This would suggest the induction of β-AR signaling during exercise is a major determinant of the subsequent changes in gene expression changes following exercise.

The findings also demonstrate for the first time crosstalk between β-AR signaling and the peripheral regulators of circadian rhythm in skeletal muscle, possibly implicating β-AR signaling in circadian effects on skeletal muscle. Gene expression changes discovered in the present study may provide insight into the mechanisms underlying β-AR-mediated changes in skeletal muscle hypertrophy and metabolism.

Methods

Animals, β-AR agonist administration and tissue collection

All procedures were approved by the Animal Experimentation Ethics Committees of The University of Melbourne. All procedures conformed to the Guidelines for the Care and Use of Experimental Animals described by the National Health and Medical Research Council of Australia. Male C57BL/10 ScSn (wild type, 6-7 weeks old) were obtained from the Animal Resource Centre (Canning Vale, WA, Australia) and were randomly assigned to either saline control, formoterol (a specific β2-AR agonist) treated or non-treated groups. For BeadArray analysis, 16 mice in total were analyzed with eight mice for both timepoints. At each timepoint, four mice were treated with formoterol and four with saline (n = 4 mice per timepoint per treatment). For quantitative real-time PCR (qRT-PCR) analysis, ten mice were used per timepoint with five formoterol treated and five saline formoterol animals. Five mice were also analyzed as non-treated animals (n = 5 mice per timepoint per treatment). The animals used for quantitative real-time PCR were independent animals to that used for BeadArray analysis. The mice were housed in the Biological Research Facility at The University of Melbourne and maintained on a 12 h-light/12 h-dark cycle, with standard mouse chow and water provided ad libitum.

Treated mice received a single intraperitoneal injection of formoterol (Astra-Zeneca: 100 μg/kg in 0.2 mL saline) and control mice received an equivalent volume of sterile saline. We have demonstrated previously that this is the most efficacious dose for eliciting skeletal muscle hypertrophy in rodents [3]. For chronic treatment timepoints, mice received a single intraperitoneal injection of formoterol (Astra-Zeneca: 100 μg/kg in 0.2 mL saline) daily and control mice received an equivalent volume of sterile saline daily.

Mice were anesthetized at 1, 4, 8, and 24 hours after acute treatment with formoterol and at 7 and 28 days after chronic formoterol administration. Following anesthetization, tibialis anterior muscles were surgically excised. Tissue was also removed from anesthetized untreated mice (n = 5) that did not receive an intraperitoneal injection for qRT-PCR "no treatment" controls. Tissue from "no treatment" was removed at time zero following injections of the treatment mice. All injections were carried out at approximately 3 hours within the light cycle and muscle was removed at relative appropriate intervals during the light cycle. Care was taken to administer formoterol and saline injections as close as possible to negate gene expression changes associated with circadian rhythm. For chronic treatments injections, were carried out at approximately 3 hours within the light cycle and muscle was removed 24 hours following the last injection. Due to concerns about stress hormones in acute timepoints, both saline control and formoterol treated animals were handled and injected in the same manner. Non-treated animals were also included with the qRT-PCR data to provide an indication of any stress related gene expression changes. As the non-treated animals were quickly anesthetized, stress related gene expression changes should be avoided.

All samples were snap-frozen in liquid nitrogen and stored at -70°C. Samples were used for RNA extraction for BeadArray analysis and qRT-PCR. Contralateral tibialis anterior muscle samples were also used for Western blot analysis.

RNA extraction for quantitative real-time PCR

RNA was extracted from skeletal muscle tissue using TRI-Reagent (Sigma Aldrich) according to the manufacturer's protocol. For qRT-PCR only, RNA was treated with 2U of Turbo DNase (Ambion) for 30 minutes. RNA was further purified using a mini-RNeasy kit (QIAGEN) according to manufacturer's instructions and quantified using a NanoDrop ND-1000 spectrophotometer.

cDNA synthesis for quantitative real-time PCR

cDNA was synthesized from 3 μg of total RNA for cell culture experiments and 1 μg for animal muscle experiments (normalized via UV spectroscopy) using Superscript III primed by random hexamers (Geneworks), according to the manufacturer's instructions (Invitrogen).

Quantitative real-time PCR (qRT-PCR) and statistical analysis

Target cDNA levels were compared by qRT-PCR in 25 μl reactions containing either 1× SYBR green (Applied Biosystems) or Taqman PCR master mix (Roche Molecular Systems), 100 nM of each forward and reverse primers for SYBR green or 1× Assay-on-Demand Taqman primers (Applied Biosystems) and the equivalent of 0.3 μL cDNA. Using an ABI Prism 7500 (Applied Biosystems) sequence detection system, PCR was conducted over 45 cycles of 95°C for 15 seconds and 60°C for 1 minute, preceded by an initial 95°C for 10 minutes. Expression levels were normalized to 36B4 as determined from the ratio of delta CT values. All 36B4 probes remained stable during Illumina BeadArray analysis, confirming this gene was appropriate for normalization. Results are expressed as means ± SEM from five biological replicates. Statistical analyses were performed using GraphPad Prism software. All qRT-PCR data were analyzed using a one-way ANOVA with Bonferroni's post-test.

qRT-PCR primers

Primers for qRT-PCR analysis of the mRNA populations using SYBR green have been described in detail for 36B4 [81], Ucp3 ([82], Nurr1 [83], Pdk4 [28], and Pgc1α ([82]. The following SYBR primers were designed using Primer Express (Applied Biosystems, Foster City, CA): Sorbs1 (F:5'-GTG CCA CAG AAC GAT GAT GAG T-3'; R:5'-AAG TAC CAA ACT GCC TCG TCC TT-3'), Id1 (F:5'-GCA GGT GAA CGT CCT GCT CTA-3'; R:5'-TCT CCA CCT TGC TCA CTT TGC-3'), Activin receptor IIB (F:5'-ACG TGG CGG AGA CGA TGT-3'; R:5'-GTG AGG TCG CTC TTC AGC AG TAC-3'), Hk2 (F:5'-TTA GGT CAG TCG GCG TTT CAG-3'; R:5'-TAG GAG GGC AAA TAA ATG TAC AAA CA-3'), Nfil3 (F:5'-CGG TTA CAG CCG CCC TTT-3'; R:5'-GTT GTC CGG CAC AGG GTA AAT-3'), and Stat3 (F:5'-GAG GAG GCA TTT GGA AAG TAC TGT A-3'; R:5'-GTC ACA CAG ATG AAC TTG GTC TTC A-3'). Assay-on-Demand Taqman primer/probe sets were used to assay expression of myostatin, Crem, Crebp, Dbp, FoxO1, and Lipin1α.

BeadArray hybridization and statistical analysis

For BeadArray analysis, 16 mice in total were analyzed with eight mice for both timepoints. At each timepoint, four formoterol treated and four saline formoterol animals. Total skeletal muscle RNA was assessed for integrity using the Agilent Bioanalyzer 2100 and RNA integrity (RIN) scores above 8.3 were present in all samples. 500 ng of RNA was amplified using the Illumina TotalPrep RNA Amplification kit (Ambion) with an in vitro transcription reaction period of 12 hours. Biotinylated, amplified cRNA was assessed for quantity and quality also using the Agilent Bioanalyzer 2100. 1500 ng per array of amplified cRNA was hybridized to Sentrix Mouse-6.v1 BeadChip arrays (Illumina) according to manufacturer directions. Hybridized BeadChip arrays were stained with Amersham fluorolink streptavidin-Cy3 (GE Healthcare). BeadChip arrays were scanned with Illumina BeadStation Scanner and data values with detection scores were compiled using BeadStudio v1.5.1.3 (Illumina) and imported into GeneSpring GX v7.3.1 (Agilent) for data analysis. Mouse Illumina probe set was defined in the GeneSpring Workgroup using the Illumina targetIDs as the unique identifiers and annotated according to array content files supplied by Illumina. Normalized data was produced using GeneSpring GX version 7.3.1 via normalization to control genes, where control genes were represented by all genes with an Illumina detection score equal to one in at least four out of the 16 samples (7,596 control genes in total). All probes except for the 7,596 probes that were determined to have an Illumina detection score equal to one in at least four out of the 16 samples were filtered out to remove probes without adequate expression levels. A parametric Welch's t-test (where variances were not assumed equal) was performed on the 7,596 probes independently for both one and four hour times groups with a p-value cutoff of 0.05. Multiple testing correction (Benjamini and Hochberg False Discovery Rate) was then applied to genes that had passed the parametric Welch's t-test based on the total detected probe-set of 7,596 probes to reduce false positives. About 5.0% of the identified probes would be expected to pass the restriction by chance. Following this statistical filtering, 393 probes were significant at four hours and 43 probes at one hour. Statistical filtered probe-sets were then independently filtered by fold change using a minimum cutoff of 1.85 fold. Following this, 112 probes were present at four hours and 23 genes at one hour. Multiple significant probes for the same gene were removed from final data tables with the probe with the highest fold change being chosen.

Western blot analysis

Whole skeletal muscle lysates were prepared from homogenisation and sonication in lysis buffer [50 mM Tris HCl, 75 mM NaCl, 5 mM EGTA, 1 mM dithiothreitol, 1% Nonidet P-40, Complete protease inhibitors (Roche Diagnostics Australia), and PhosSTOP phosphatase inhibitors (Roche Diagnostics Australia)] and resolved on 10% SDS-PAGE gels under reducing conditions as outlined [84]. Proteins were transferred to an Immobilon-P polyvinyl difluoride membrane (Millipore) and blocked for 1 h with 5% non-fat milk powder in tris-buffered saline with 0.1% Tween 20. Blots were probed overnight with anti-Myostatin (sc-6884; Santa Cruz) at 1:300 and anti-Gapdh (R&D Systems) at 1:10000 in blocking buffer. Blots were also probed overnight with anti-Smad3 (#9523; Cell Signaling) at 1:1000 and anti-phospho-Smad3 (#9520; Cell Signaling) at 1:1000 in 5% fraction V BSA (Sigma-Aldrich) in tris-buffered saline. Secondary anti-rabbit-horseradish peroxidase conjugate (Pierce Biotechnology) in blocking buffer was used at 1:10000 for Gapdh, Smad3, and phospho-Smad3 for 1 h. Secondary anti-goat-horseradish peroxidase conjugate (Pierce Biotechnology) in blocking buffer was used at 1:10000 for Myostatin for 1 h. Horseradish peroxidase localization was detected with Immobilon Western Chemiluminescent HRP Substrate (Millipore) according to the manufacturer's instructions and visualized by X-ray film.