Endocrinology and Metabolism

β-Adrenoceptor signaling in regenerating skeletal muscle after β-agonist administration

Felice Beitzel, Martin N. Sillence, Gordon S. Lynch


Stimulating the β-adrenoceptor (β-AR) signaling pathway can enhance the functional repair of skeletal muscle after injury, but long-term use of β-AR agonists causes β-AR downregulation, which may limit their therapeutic effectiveness. The aim was to examine β-AR signaling during early regeneration in rat fast-twitch [extensor digitorum longus (EDL)] and slow-twitch (soleus) muscles after bupivacaine injury and test the hypothesis that, during regeneration, β-agonist administration does not cause β-AR desensitization. Rats received either the β-AR agonist fenoterol (1.4 mg·kg−1·day−1 ip) or saline for 7 days postinjury. Fenoterol reduced β-AR density in regenerating soleus muscles by 42%. Regenerating EDL muscles showed a threefold increase in β-AR density, and, again, these values were 43% lower with fenoterol treatment. An amplified adenylate cyclase (AC) response to isoproterenol was observed in cell membrane fragments from EDL and soleus muscles 7 days postinjury. Fenoterol attenuated this increase in regenerating EDL muscles but not soleus muscles. β-AR signaling mechanisms were assessed using AC stimulants (NaF, forskolin, and Mn2+). Although β-agonist treatment reduces β-AR density in regenerating muscles, these muscles can produce large cAMP responses relative to healthy (uninjured) muscles. Desensitization of β-AR signaling in regenerating muscles is prevented by altered rates of β-AR synthesis and/or degradation, changes in G protein populations and coupling efficiency, and altered AC activity. These mechanisms have important therapeutic implications for modulating β-AR signaling to enhance muscle repair after injury.

  • muscle regeneration
  • β-adrenoceptor
  • muscle function
  • muscle injury
  • adenylate cyclase
  • G protein

studies in rodents and livestock have shown that chronic administration of β-adrenoceptor (β-AR) agonists can lead to skeletal muscle hypertrophy and an increase in maximum force-producing capacity (5, 10, 15, 41, 50). The increase in muscle mass is the result of protein accretion via an increase in protein synthesis and a decrease in degradation, effects mediated via β-AR activation, adenylate cyclase (AC) induction, and a consequent increase in intracellular cAMP concentration (15, 17, 23, 37). There is potential to exploit the anabolic effects of β-agonists therapeutically, to promote muscle regeneration after injury, since β-AR-mediated mechanisms are postulated to play a physiological role in successful muscle regeneration (34). Evidence comes from the fact that the β-agonist fenoterol can hasten functional recovery of regenerating rat skeletal muscle after myotoxic injury. Previous studies have shown that daily fenoterol administration enhances the force-producing capacity of regenerating muscles by 19% compared with control by 14 days postinjury (5). The improvement in contractile function with fenoterol treatment was associated with increases in protein content and the cross-sectional area of regenerating muscle fibers (5). Additionally, fast-twitch muscles in the rat have a higher density of β-AR when they are regenerating, suggesting that endogenous catecholamine stimulation could play a role in the regenerative process. However, a local increase in catecholamine activity, mediated by an increase in tissue sensitivity, would exploit the anabolic pathways linked to β-AR, without systemic complications.

In addition to changes in β-AR density, alterations in β-AR signaling could influence muscle regenerative processes and specific responses to β-AR stimulation after β-agonist administration. In adult muscle, the hypertrophic response to β-agonists may be limited by β-AR downregulation (a reduction in the density of receptors at the cell membrane) and desensitization (uncoupling of the receptors from their response elements for cell signaling; see Refs. 26, 28, and 48). Downregulation occurs after prolonged exposure to high-dose β-agonist administration, which renders β-AR more susceptible to β-AR kinase (β-ARK), an enzyme that phosphorylates agonist-bound receptors and enhances their affinity for β-arrestins. Further interaction between the β-AR and G proteins is prevented, and endocytosis of the receptors is initiated, which leads to fewer receptors on the cell surface (18, 26, 28, 31). Studies in rat hindlimb muscles have demonstrated a reduction in β-AR density with 7 days exposure to various β-agonists, including clenbuterol, isoproterenol, and terbutaline (14, 46). Additionally, exposure of L6 myoblast membranes to isoproterenol can result in downregulation of β-AR within 4 and 24 h of internalization of the β-AR from the cell membrane (21).

β-AR desensitization may be homologous or heterologous. As with downregulation, homologous desensitization also occurs when the β-arrestins bind to the phosphorylated β-AR and prevent further G protein interaction (22, 28, 33). However, unlike downregulation, homologous desensitization is reversible. In heterologous desensitization, elements downstream of the β-AR, such as G proteins or AC itself, may be phosphorylated, or their expression and function altered to favor inhibition. This impairs signal efficacy and reduces AC activity, not only for β-AR but for any receptor that shares this second messenger signaling pathway (16, 28, 52).

The attenuation of β-AR signaling is important for cellular homeostasis, since continued overstimulation of the pathway can lead to cellular damage, changes in cell differentiation, and apoptosis (7, 13, 49). However, from a therapeutic standpoint, it has been suggested that β-AR downregulation and desensitization may limit the effectiveness of chronic β-agonist administration (11).

Several studies have examined the β-AR signaling pathway during fetal or neonatal development. Experiments using heart, lung, brown fat, and liver suggest that immature, developing tissue is resistant to desensitization by chronic β-agonist exposure (3, 4, 9, 29, 36). These observations indicate that the downregulation and desensitization responses may not be inherent but are acquired during development. Thus it is possible that changes in β-AR signaling and an altered response to β-agonists may occur in tissue that is regenerating after injury. If so, this could have important implications for the therapeutic use of β-agonists to enhance muscle regeneration and promote functional restoration. Thus the aim was to examine the effect of β-agonist administration on the β-AR population and the downstream signaling pathway during early regeneration, in both fast-twitch and slow-twitch muscles, and test the hypothesis that, during regeneration, chronic β-agonist administration does not cause β-AR desensitization.



The Animal Experimentation Ethics Committee of The University of Melbourne approved all experiments, which were conducted in accordance with the guidelines for Care and Use of Experimental Animals, as described by the National Health and Medical Research Council of Australia. Young adult (275–300 g) male Fischer 344 rats (total n = 116; Animal Resource Centre, Canning Vale, WA, Australia) were housed in the Biological Research Facility at The University of Melbourne. They were maintained under a 12:12-h light-dark cycle (light 0600–1800), with access to standard rat chow and water ad libitum.

Experimental muscle injury.

All rats were anesthetized deeply (100 mg/kg ketamine and 10 mg/kg ip xylazine) such that they were unresponsive to tactile stimuli. The extensor digitorum longus (EDL) and soleus muscles of the right hindlimb were surgically exposed, with care taken to avoid damaging the nerve and blood supplies, and then injected to maximal holding capacity with 0.5% bupivacaine hydrochloride (Marcain, Astra, North Ryde, NSW, Australia), using techniques described previously (5, 19), to ensure degeneration of all muscle fibers (Fig. 1). The EDL and soleus muscles of the left hindlimb served as the uninjured controls. After surgery, the rats were assigned at random to receive daily treatment with either fenoterol, a full agonist of both β2-AR and β1-AR (1.4 mg·kg−1·day−1 ip dissolved in 1 ml isotonic saline; Sigma-Aldrich, Castle Hill, NSW, Australia; see Refs. 6 and 41), or an equal volume of saline for 2, 5, or 7 days. Previous studies using this protocol have shown that degeneration is maximal 2 days after bupivacaine, which is followed by spontaneous regeneration, such that immature but functional muscle fibers are present by 7 days postinjury (5, 20 and Fig. 1).

Fig. 1.

Hematoxylin & eosin-stained cross sections of uninjured (top left) and regenerating skeletal muscle at 2 (top right), 5 (bottom left), or 7 (bottom right) days following maximal injection with bupivacaine hydrochloride. Note the maximal bupivacaine-induced degeneration at 2 days postinjury. By 5 days postinjury, small myotubes are evident, which mature to form the new muscle fibers seen at 7 days postinjury.

At the end of the treatment period, the rats were anesthetized deeply (100 mg/kg ketamine and xylazine 10 mg/kg ip xylazine), and the EDL and soleus muscles were surgically excised and trimmed of tendons and any adhering nonmuscle tissue. Muscles used for β-AR density measurements, Western blotting, and PCR were frozen rapidly in isopentane cooled in liquid nitrogen and stored at −80°C for subsequent processing and analysis. Muscles used for AC activity assays were placed in ice-cold buffer for immediate preparation of cell membrane extracts from the fresh tissue. After the surgical procedures, the thoracic region was exposed, and the rats were killed by cardiac excision while still anesthetized.

Preparation of cell membranes.

Procedures for the preparation of cell membrane fragments have been described previously (5, 41, 45, 47). Briefly, to obtain enough material for the assays, the EDL and soleus muscles were pooled (in pairs for radioligand-binding assays and in threes for cAMP) according to treatment group and injury state. The muscles were then homogenized in 2 ml ice-cold buffer [buffer A, in mM: 50 Tris (pH 7.0) 250 sucrose, and 1 EGTA; pH 7.4 at 4°C], and cell membrane fragments were prepared by centrifugation at 4°C. The centrifugation protocol ensures isolation of the sarcolemmal membrane fraction only. Therefore, the β-AR examined are those present in the sarcolemma that are available for stimulation by both endogenous catecholamines and exogenously administered β-agonists. The cell membrane pellets were resuspended in ice-cold buffer [for ligand binding 1 ml buffer B (in mM): 50 Tris (pH 7.6), 10 MgCl2·(6H2O), and 150 NaCl, pH 7.4 at 4°C, and, for cAMP assays, 550 μl buffer C (in mM): 50 Tris (pH 7.4), 5 MgCl2·(6H2O), and 1 EGTA, pH 7.4 at 4°C] using a hand-held glass homogenizer. The protein concentration of the membrane suspension was determined using a Bradford protein assay (Bio-Rad, Richmond, CA). Cell membrane samples for ligand binding assay were then stored at −80°C, whereas the preparations for AC assays were used immediately.

Radioligand binding assays.

Radioligand binding assays were performed to measure the density of β-AR in muscles from rats (n = 60; muscles pooled in pairs) at 2, 5, and 7 days postinjury using methods described previously (5). Single-point saturation assays were performed because of the limited amount of protein that could be obtained from the small muscles and to ensure that the sample protein concentration was within the concentration range of 0.05–0.3 mg/ml, in which the binding of the radioligand to the β-AR sites has been shown to be linear. Measurement of β-AR density in muscle membranes detects both β1- and β2-AR present. However, the β-AR detected cannot be distinguished as β1- or β2-AR because of the limited amount of membrane protein that can be obtained from small muscles, since this procedure requires greater amounts of membrane. Therefore, the results for receptor density are expressed as total β-AR. Briefly, the incubations were performed in 12 × 75-mm tubes containing 400 μl of the cell membrane suspension with 50 μl of the radioligand [125I]iodocyanopindolol (ICYP, 135 pM) and 50 μl of either buffer B (to determine the total binding of ICYP to β-AR) or the nonselective β-AR antagonist dl-propranolol (2 μM; to determine nonspecific binding of ICYP to the membrane). The receptor-bound radioligand was then collected by filtering the preparation through glass-fiber filter paper (Whatman GF-C filter paper, Maidstone, UK) and by rinsing each tube using a cell harvester (Brandel M-48R cell harvester; Biomedical Research and Development Laboratories, Gaithersburg, MD). Radioactivity remaining on the filters was determined using a gamma counter (1470 Wizard-automatic gamma counter; Wallac, Turku, Finland) at a counting efficiency of 78%. Results were expressed as γ-radiation counts per minute, and a conversion was applied to determine the concentration of β-AR (mol/mg of protein) based on the specific activity of the radioligand.

RNA analysis.

Total RNA was isolated from small portions (∼10 mg) taken from the midbelly region of the same EDL and soleus muscles at 2, 5, and 7 days postinjury. Samples were prepared using a commercially available kit and according to the manufacturer's instructions (kit no. 75742; Qiagen). Both β1- and β2-AR mRNA were examined because, although β2-AR have been identified as the predominant subtype in skeletal muscle, some studies have shown that skeletal muscles may also contain ∼7–10% β1-AR, with this value slightly higher in slow-twitch muscles (25, 27). Levels of β1-AR and β2-AR mRNA were analyzed by semiquantitative RT-PCR based on techniques described previously, using 18S as a loading control (12), also using a commercially available kit (kit no. 74704; Qiagen, Valencia, CA).

Western blot analysis.

Western immunoblot analysis was performed to measure Gsα splice variants in intact EDL and soleus muscles from rats (n = 8) at 7 days postinjury, utilizing the methodology of Schertzer and colleagues (42, 43). Primary antibodies were Gsα (K-20) and sc-823. A G protein-tagged fusion protein was used as positive control (Gsα-sc-4225 WB; Santa Cruz Biotechnology).

AC activity assays.

AC activity was measured in muscles taken from rats (n = 48; pooled in triplicate) at 7 days postinjury based on methods described previously (45). Production of cAMP was measured in response to various stimulants that act at different points on the AC signaling pathway. The response to direct β-adrenergic stimulation was determined using the nonselective β-agonist isoproterenol (30). Maximal G protein activation was determined using NaF, which recruits all G proteins present (8, 53). Forskolin and MnCl2 were used to directly activate AC. Forskolin binds to a conserved region on the AC molecule and acts to “glue together” the two domains of the catalytic core, and Gsα association with AC enhances the effect of forskolin to increase AC activity (24, 44, 51). In contrast, Mn2+ activates AC by attaching to the Mg2+ binding site in the AC active site and interferes with G protein association such that Mn2+-stimulated AC activity is reduced (24, 32, 52). Use of various stimulants enables the identification of differences in the intracellular signaling pathway between uninjured and injured muscles, fast- vs. slow-twitch muscles, and whether aspects of the signaling pathway are altered by fenoterol treatment. Briefly, fresh l-isoproterenol (4 mM) and forskolin solutions (0.4 mM) were prepared, and an incubation buffer was made immediately before the incubation [in mM: 50 Tris (pH 7.7), 10 MgCl2·(6H20), 2 EGTA, 2 isobutyl methylxanthine, 0.4 mg/ml bacitracin, 6 mg/ml BSA 6, 40 phosphocreatine, 2 ATP, 0.2 GTP, and 0.6 mg/ml creatine phosphokinase, pH 7.4 at 30°C]. A small volume of incubation buffer was used to prepare NaF incubation buffer (20 mM) and Mn2+ incubation buffer (20 mM). All incubations were performed in triplicate. The final extraction of cAMP from samples was performed at 4°C by centrifugation with ethanol as described previously (45), and the samples were then assayed for cAMP concentration using a commercial assay kit (cAMP Biotrak EIA System, nonacetylation procedure; Amersham, GE Healthcare, Castle Hill, NSW, Australia; RPN 225 or 2255).

Statistical analysis.

All values are expressed as means with bars denoting SE. Experimental groups from each time point were compared using a two-way ANOVA with main effects sought for treatment (saline vs. fenoterol) and injury (injured vs. uninjured). When significant differences were found, means were compared by using Fisher's least-significant difference post hoc multiple-comparison procedure. In all cases, differences between groups were considered significant when P < 0.05.


β-AR density.

Ligand binding assays showed that β-AR density was ∼2.5-fold greater in healthy slow-twitch soleus than in fast-twitch EDL muscles (Fig. 2). The comparatively low density of β-AR in uninjured EDL was not reduced significantly by fenoterol treatment at any time tested (Fig. 2A). In regenerating EDL muscles, there was a marked increase in β-AR density (∼80%) that was evident at 2, 5, and 7 days postinjury. Fenoterol decreased the density of β-AR in regenerating EDL muscles at 5 and 7 days postinjury compared with saline-treated regenerating muscles; however, β-AR levels remained higher than in the contralateral fenoterol-treated uninjured muscles at those time points (Fig. 2A).

Fig. 2.

β-Adrenoceptor (β-AR) density of uninjured and regenerating extensor digitorum longus (EDL; A) and soleus (B) muscles from rats treated with saline or fenoterol at 2, 5, or 7 days postinjury. In the EDL muscles (A), note the marked increase in β-AR density during regeneration, and downregulation of β-AR density in regenerating muscles occurred with fenoterol treatment. In the soleus muscles (B), downregulation of β-AR density occurred with fenoterol treatment, with a greater effect in injured than uninjured muscles; n = 4–5 experiments. P < 0.05, different from saline-uninjured (control) muscles (*), difference between fenoterol-uninjured and fenoterol-injured muscles (†), and difference between saline-injured and fenoterol-injured muscles (#).

In uninjured soleus muscles, β-AR density was not affected significantly after 2 and 5 days but was reduced by 20% compared with control levels after 7 days of fenoterol administration (Fig. 2B). In contrast to the EDL, in regenerating soleus muscles, β-AR density was reduced by 47% compared with control muscles at 2 days but was restored to control levels by 5 days postinjury. However, similar to EDL muscles, fenoterol treatment prevented the restoration of β-AR in regenerating soleus muscles such that levels were reduced compared with saline-treated regenerating muscles at 5 and 7 days postinjury. Thus, with fenoterol treatment, β-AR levels were not restored to those found in the contralateral uninjured muscles (Fig. 2B).

β-AR mRNA expression.

Although fenoterol treatment did not cause downregulation of β-AR in uninjured EDL muscles, after 5 days of fenoterol administration, there was a 51% and a 27% reduction in mRNA for β1-AR and β2-AR, respectively, that was restored to control levels by 7 days (Fig. 3, A and B). In regenerating EDL muscles, the marked increase in β-AR density was accompanied by a 28 and a 37% increase in β2-AR mRNA compared with control muscles at 5 and 7 days postinjury, respectively, whereas there was no change in β1-AR mRNA levels at any time during muscle regeneration (Fig. 3, A and B). Although fenoterol attenuated the increase in β-AR density in EDL muscles after injury, it did not prevent the increase in mRNA levels for β2-AR at 5 days postinjury. β1-AR mRNA was unchanged in regenerating muscles with fenoterol treatment compared with saline-treated regenerating muscles (Fig. 3, A and B).

Fig. 3.

Expression of β2-AR and β1-AR mRNA in uninjured and injured EDL (A and B) and soleus (C and D) muscles from rats treated with saline or fenoterol at 2, 5, or 7 days postinjury. In uninjured EDL muscles, 5 days of fenoterol treatment caused a reduction in β2-AR mRNA (A) and β1-AR mRNA (B). However, β-AR mRNA levels in regenerating EDL muscles were not effected by fenoterol. In the soleus, β2-AR mRNA (C) was reduced with fenoterol in both uninjured and regenerating muscles, and β1-AR mRNA (D) levels were not affected by fenoterol treatment. (n = 4). P < 0.05, different from saline-uninjured (control) muscles (*), difference between fenoterol-uninjured and fenoterol-injured muscles (†), and difference between saline-injured and fenoterol-injured muscles (#).

In uninjured soleus muscles, β1-AR mRNA was not affected by fenoterol treatment; however, β2-AR mRNA was reduced (by 39 and 51%) after 5 and 7 days of administration, respectively. In regenerating soleus muscles, β2-AR mRNA was reduced to 52%, β1-mRNA increased by 90% compared with control muscles at 5 days, and mRNA levels for both β-AR subtypes were returned to control values by 7 days postinjury (Fig. 3, C and D). With fenoterol treatment in regenerating soleus muscles, β2-AR mRNA was reduced at 2, 5, and 7 days postinjury compared with control levels and was in fact reduced by 43% compared with levels in saline-treated injured muscles at 7 days postinjury (Fig. 3C). As for regenerating EDL muscles, β1-AR mRNA was unchanged in regenerating soleus muscles after fenoterol treatment compared with saline-treated regenerating muscles (Fig. 3D).

G proteins.

Similar to the observed differences in β-AR density between fast- and slow- twitch muscles, the amount of total Gsα in control soleus muscles was ∼2-fold greater than in control EDL muscles because of the ∼2-fold increases in the amount of the Gsα splice variants, GsαS and GsαL, in the control soleus muscles (Fig. 4). Fenoterol treatment did not alter the amount of either splice variant of Gsα in uninjured EDL or soleus muscles (Fig. 4, C and D).

Fig. 4.

Gsα proteins in regenerating EDL and soleus muscles from rats treated with saline or fenoterol at 7 days postinjury. Total Gsα (A) was determined by Western blotting (B). The sum of both the 45-kDa short (GsαS; C) and the 52-kDa long (GsαL; D) splice variants for EDL and soleus muscles. Note that fenoterol treatment did not alter the amount of either splice variant of Gsα in both uninjured and regenerating EDL and soleus muscles, however, the decrease in total Gsα in regenerating soleus muscles was only significant in control rats. (n = 4). P < 0.05, different from saline-uninjured (control) muscles (*) and difference between fenoterol-uninjured and fenoterol-injured muscles (†). UE, uninjured EDL; IE, injured EDL; US, uninjured soleus; IS, injured soleus.

In regenerating EDL muscles, GsαS and GsαL increased by 55 and 222%, respectively, so there was a 64% increase in total Gsα in regenerating EDL muscles. Fenoterol treatment did not alter the amount of either splice variant of Gsα in regenerating EDL muscles compared with saline treated regenerating muscles (Fig. 4, C and D).

In regenerating soleus muscles, levels of GsαS decreased by 49%, whereas levels of GsαL increased by 116% compared with control (Fig. 4, C and D). However, despite the marked increase in GsαL because of the reduction in the predominant splice variant GsαS, there was a net 40% reduction of total Gsα in regenerating soleus muscles compared with control (Fig. 4A). Although fenoterol treatment did not alter the amount or proportions of either isoform of Gsα in regenerating soleus muscles, the reduction of total Gsα was only significant in saline-treated rats, not in fenoterol-treated rats (Fig. 4A).

AC activity.

Fenoterol treatment did not alter basal cAMP production or isoproterenol-, NaF-, Mn2+-, or forskolin-stimulated cAMP production in uninjured EDL muscles (Fig. 5). In regenerating EDL muscles, basal cAMP production was not different from controls and was not affected by fenoterol treatment (data not shown). β-AR agonist (isoproterenol)-stimulated cAMP production was increased by 77% in regenerating EDL muscles; however, fenoterol treatment impaired the increase in isoproterenol-stimulated cAMP production (Fig. 5B). NaF-stimulated cAMP production during regeneration was not different from control but, with fenoterol treatment, was increased by 88% in regenerating muscles compared with control levels (Fig. 5C). Mn2+-stimulated cAMP production in regenerating muscles was not different from controls and was not affected by fenoterol treatment (Fig. 5D). Forskolin-stimulated cAMP production was increased by 64% in regenerating EDL muscles compared with controls but was also not affected by fenoterol treatment (Fig. 5E).

Fig. 5.

Basal (A) and isoproterenol (B)-, NaF (C)-, Mn2+ (D)-, and forskolin (E)-stimulated cAMP production of uninjured and regenerating EDL and soleus muscles from rats treated with saline or fenoterol at 7 days postinjury. Note the marked increase in isoproterenol-stimulated AC activity during regeneration compared with control muscles, which was reduced with fenoterol treatment in regenerating EDL but not regenerating soleus muscles (A). NaF-stimulated cAMP production was enhanced during regeneration in the EDL with fenoterol treatment (B). Fenoterol did not affect basal, Mn2+-, or forskolin- stimulated cAMP production during regeneration. (n = 5–8). P < 0.05, different from saline-uninjured (control) muscles (*) and difference between fenoterol-uninjured and fenoterol-injured muscles (†).

In uninjured soleus muscles, fenoterol treatment did not alter basal cAMP production or the response to any of the stimulants tested (Fig. 5). In the soleus muscles, basal cAMP production during regeneration was increased by 77% compared with control, which was attenuated with fenoterol treatment (Fig. 5A). Isoproterenol-stimulated cAMP production during regeneration was 54% greater than control but, unlike basal AC activity, was not affected by fenoterol administration and remained 50% greater than control levels (Fig. 5B). Regenerating soleus muscles also showed marked increases in NaF-, Mn2+-, and forskolin-stimulated cAMP production by 135, 58, and 117%, respectively, compared with control. Fenoterol treatment did not attenuate these increases, which remained at 81, 40, and 145% greater than control levels for NaF-, Mn2+-, and forskolin-stimulated cAMP production, respectively (Fig. 5, C–E).


This study identified unique characteristics of β-adrenergic signaling in regenerating fast- and slow-twitch skeletal muscles that highlight this signaling pathway as a potential therapeutic target for promoting muscle repair after injury. In healthy adult muscle, β-AR responses to β-agonist stimulation are limited by downregulation and by desensitization of AC signaling (2, 26, 28). We found the capacity of fenoterol to downregulate β-AR was influenced by the muscle phenotype, the basal level of β-AR expression, and injury status. Greater downregulation of β-AR was evident in regenerating muscles than in healthy uninjured muscles. The classical mechanism of β-AR downregulation involves internalization of the receptor, and, consistent with this, the prolonged administration of a high-dose β-agonist would make β-AR more susceptible to phosphorylation by β-ARK and downregulation from the cell surface (26, 28). However, we suggest that, although downregulation imposes limitations on β-AR signaling in adult tissue, this does not necessarily occur in regenerating muscle, such as in soleus muscle where maximum signal transduction was not reduced by fenoterol treatment.

With chronic β-agonist administration, β-AR synthesis can also be impaired. In regenerating EDL muscles, the decrease of β-AR density at the membrane at 5 and 7 days postinjury was not reflected by a similar reduction in β1-AR or β2-AR mRNA at these time points. This suggests that the reduction in β-AR density at the membrane after fenoterol treatment was due primarily to their downregulation at the cell surface or in the posttranscriptional regulation of β-AR.

In contrast, in the soleus muscles, the β2-AR mRNA levels were reduced in both uninjured and injured muscles with 5 and 7 days of fenoterol treatment, respectively. These changes in mRNA levels did not match the time course of β-AR density. These observations suggest that there is an even greater reduction in the rate of receptor degradation. In particular, fenoterol impaired the restoration of β2-AR mRNA levels at 7 days postinjury, suggesting that treatment with the β-agonist prevented the reduction in receptor degradation that would have otherwise occurred during regeneration. It should be noted that, during the inflammatory process that accompanies the muscle degeneration/regeneration process, it is possible that the isolated membrane fragment could contain macrophages and/or neutrophils that may themselves have disturbed β-AR function. It has been shown previously in studies on airway smooth muscle that mediators released from inflammatory cells such as reactive oxygen species and fatty acid metabolites may directly or indirectly induce β-AR dysfunction, which may have consequences for their immune function, mediator release, and effects on surrounding tissues (38). Thus, although the clear differences between the responses of fast and slow muscles in the present study suggest that the relative contribution of disturbed β-AR function of macrophages and/or neutrophils is relatively small, nonetheless they should be taken into account when interpreting the findings.

Our results suggest that the proportion of β1-AR in soleus muscles is likely to be higher than normal for a brief period during regeneration and higher still following fenoterol treatment because of a downregulation of β2-AR mRNA at 7 days postinjury. However, the relative contribution of β1-AR to muscle regeneration is unclear. Specifically, although fenoterol did not prevent the increase in β1-AR expression at day 5, it reduced the total β-AR density markedly. This suggests that, even after upregulation, the β1-AR still represents a minor population relative to the β2-AR. Nevertheless, the role of β1-AR in maintaining β-AR signaling during muscle repair is worthy of further investigation.

In regenerating EDL muscles at 7 days postinjury, the downregulation of β-AR with fenoterol resulted in diminished β-AR signaling, signified by a reduction in the isoproterenol-stimulated cAMP response. The maintained NaF response suggests that β-AR downregulation caused a reduction in coupling between G proteins and the β-AR; thus, the β-AR-mediated cAMP response was reduced with fenoterol treatment at this time point. However, fenoterol did not impair elements of the downstream signaling pathway, since the responses to the downstream stimulants, NaF, Mn2+, and forskolin were not impaired, and amounts and proportions of G proteins were unaffected. In fact, this response was elevated in regenerating EDL muscles, suggesting that, at the level of postreceptor signaling, a heterologous mechanism occurs in regenerating muscles to enhance cAMP responses, despite fenoterol treatment.

A reduction in the cAMP response to isoproterenol was not observed in soleus muscles from fenoterol-treated rats at 7 days postinjury. This indicates that any downregulation caused by fenoterol was compensated for by alterations in downstream signaling such that the cAMP response was maintained. The coupling between G proteins and β-AR was not impaired by fenoterol administration, the expression of β1-AR remained increased at 5 days postinjury, and there was no significant reduction of total Gsα in regenerating muscles with 7 days of treatment. Additionally, the enhanced cAMP production observed in regenerating muscles at 7 days postinjury was maintained, since the responses of signaling elements downstream of receptor were unaffected by fenoterol and cAMP responses remained elevated.

Studies using similar AC activity measurements in developing rat fetal or neonatal heart and liver have demonstrated that, unlike in the adult, after chronic β-agonist administration, the ability of β-AR stimulation to elicit an increase in cAMP is not desensitized. Instead, signaling is maintained or even enhanced through heterologous sensitization of elements downstream of the receptor, at the levels of G protein coupling and AC activity (3, 4, 48, 53). For example, Auman and colleagues (3) showed that β-AR/AC responses in the liver and heart were maintained in neonatal rats after chronic treatment with the β2-agonist terbutaline despite β-AR downregulation. In some cases, the neonatal β-AR signaling pathway exhibited sensitization through mechanisms downstream of the receptor and possible deficiencies in the ability of immature cells to uncouple receptors from response elements. The observation that such processes enable β-AR function to be maintained during heart and liver development highlights the potential value of the β-AR signaling pathway as a useful therapeutic target for promoting growth and repair after muscle injury.

Our findings support the notion that the functional response to β-agonists depends on the β-AR-to-G protein-to-AC ratio rather than the absolute numbers of receptors and that the amount and isoforms of G proteins and AC are important (40). We have identified that these aspects of the signaling pathway vary depending on tissue type or state of growth, development, or repair.

Many studies have suggested that the number of β-AR present exceeds that required to maximally activate the effector, since G proteins exist in great stoichiometric excess (1, 39). It has also been argued that maximal stimulation of cAMP production can be increased with an elevation in the total levels of AC (35). With an excess of receptors, a decrease in receptor density may decrease the sensitivity to an agonist but would only decrease the maximal effect if the drug was not a full agonist (4). Therefore, with the downregulation after 5 days of fenoterol treatment in regenerating muscles, an initial reduction in the potency of fenoterol may occur, although its ability to maximally stimulate AC activity would be unaffected. A further reduction of β-AR density would lead to a proportional decrease in maximal stimulation because of a lack of spare receptors. Thus, in the EDL muscles, the reduction in β-AR density would be expected to first decrease the potency of fenoterol to stimulate AC and subsequently to decrease the maximal effect of fenoterol. During regeneration of soleus muscles, there was no decrease in response with the decreased β-AR, suggesting that the β-AR density exceeds that required to maximally activate AC.

The findings suggest that cAMP output is elevated in regenerating muscles with fenoterol treatment because they have greater β-AR stimulation than in untreated muscles. The alterations in β-AR density and second messenger coupling are likely to alter tissue sensitivity and responsiveness to β-agonist administration. Whereas the anabolic effect of a weak partial agonist drug may be attenuated over time (because of β-AR downregulation), the response of a regenerating muscle to a full agonist is likely to be unaffected, even when the drug is given in high doses. Differences have been identified between liver and heart in the ability of their developing cells to desensitize following chronic β-AR administration (4), and we have identified an enhanced responsiveness of the signaling pathway, shown by responses to NaF, forskolin, and Mn2+ may vary depending on tissue type or state of growth, development or repair. Thus it would be interesting to examine whether increased responsiveness of β-AR signaling plays an important role in repair of other tissues, or attributes of signaling may be shared by other GPCRs, not only in skeletal muscle.

Our findings demonstrate that, despite the marked β-AR downregulation within 5 days of β-agonist administration, desensitization is prevented in regenerating skeletal muscle by alterations in the G protein population and the coupling efficiency and AC, which not only improve signaling and promote the physiological responses required for successful regeneration but have important implications when considering tissue sensitivity and responsiveness to β-AR agonist therapies for promoting muscle repair.


This research was supported under the Australian Research Council's Discovery-Project funding scheme (project nos. DP0665071 and DP0772781) and the National Health and Medical Research Council of Australia (project nos. 350439 and 454561).


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract