The involvement of transient receptor potential canonical type 1 in skeletal muscle regrowth after unloading‐induced atrophy

Abstract

Key points

• Decreased mechanical loading results in skeletal muscle atrophy. The transient receptor potential canonical type 1 (TRPC1) protein is implicated in this process. Investigation of the regulation of TRPC1 in vivo has rarely been reported. In the present study, we employ the mouse hindlimb unloading and reloading model to examine the involvement of TRPC1 in the regulation of muscle atrophy and regrowth, respectively.

• We establish the physiological relevance of the concept that manipulation of TRPC1 could interfere with muscle regrowth processes following an atrophy‐inducing event. Specifically, we show that suppressing TRPC1 expression during reloading impairs the recovery of the muscle mass and slow myosin heavy chain profile. Calcineurin appears to be part of the signalling pathway involved in the regulation of TRPC1 expression during muscle regrowth.

• These results provide new insights concerning the function of TRPC1. Interventions targeting TRPC1 or its downstream or upstream pathways could be useful for promoting muscle regeneration.

Abstract

Decreased mechanical loading, such as bed rest, results in skeletal muscle atrophy. The functional consequences of decreased mechanical loading include a loss of muscle mass and decreased muscle strength, particularly in anti‐gravity muscles. The purpose of this investigation was to clarify the regulatory role of the transient receptor potential canonical type 1 (TRPC1) protein during muscle atrophy and regrowth. Mice were subjected to 14 days of hindlimb unloading followed by 3, 7, 14 and 28 days of reloading. Weight‐bearing mice were used as controls. TRPC1 expression in the soleus muscle decreased significantly and persisted at 7 days of reloading. Small interfering RNA (siRNA)‐mediated downregulation of TRPC1 in weight‐bearing soleus muscles resulted in a reduced muscle mass and a reduced myofibre cross‐sectional area (CSA). Microinjecting siRNA into soleus muscles in vivo after 7 days of reloading provided further evidence for the role of TRPC1 in regulating muscle regrowth. Myofibre CSA, as well as the percentage of slow myosin heavy chain‐positive myofibres, was significantly lower in TRPC1‐siRNA‐expressing muscles than in control muscles after 14 days of reloading. Additionally, inhibition of calcineurin (CaN) activity downregulated TRPC1 expression in both weight‐bearing and reloaded muscles, suggesting a possible association between CaN and TRPC1 during skeletal muscle regrowth.

Key points

• Decreased mechanical loading results in skeletal muscle atrophy. The transient receptor potential canonical type 1 (TRPC1) protein is implicated in this process. Investigation of the regulation of TRPC1 in vivo has rarely been reported. In the present study, we employ the mouse hindlimb unloading and reloading model to examine the involvement of TRPC1 in the regulation of muscle atrophy and regrowth, respectively.

• We establish the physiological relevance of the concept that manipulation of TRPC1 could interfere with muscle regrowth processes following an atrophy‐inducing event. Specifically, we show that suppressing TRPC1 expression during reloading impairs the recovery of the muscle mass and slow myosin heavy chain profile. Calcineurin appears to be part of the signalling pathway involved in the regulation of TRPC1 expression during muscle regrowth.

• These results provide new insights concerning the function of TRPC1. Interventions targeting TRPC1 or its downstream or upstream pathways could be useful for promoting muscle regeneration.

Abbreviations

14U

14 days of unloading

3R

7R

14R or 28R

3, 7, 14, or 28 days of reloading

Ca²⁺

calcium ions

CaN

calcineurin

[Ca²⁺]_i

intracellular Ca²⁺

CSA

cross‐sectional area

DAPI

4′,6‐diamidino‐2‐phenylindole

GAPDH

glyceraldehyde‐3‐phosphate dehydrogenase

MHC

myosin heavy chain

NFAT

nuclear factor of activated T cells

Orai1

release‐activated calcium channel protein 1

PI3K‐Akt

phosphatidylinositol 3′‐kinase

RCAN1

regulator of calcineurin 1

SOCE

store‐operated Ca²⁺ entry

siRNA

small interfering RNA

STIM1

stromal interacting molecule 1

TRITC

tetramethylrhodamine

TRPC1

transient receptor potential canonical type 1

WB

weight‐bearing

WGA

wheat germ agglutinin

Introduction

Skeletal muscles exhibit high plasticity in response to mechanical stimuli. Decreases in mechanical loading, such as bedrest or inactivity, alter the balance between protein synthesis and degradation, resulting in muscle atrophy (Rennie et al. 2004; Favier et al. 2008). Muscle atrophy is characterized by decreases in muscle mass, myofibre cross‐sectional area (CSA), contractile strength and slow‐to‐fast fibre‐type transformation (Fitts et al. 2001; Adams et al. 2003). Subsequent mechanical reloading induces muscle regrowth and fibre‐type reversion. The potential to recover muscle mass after atrophy is of great clinical importance because impaired/delayed muscle regrowth can have severe consequences on functional performance (Sargeant et al. 1977; Kortebein et al. 2008).

Although it is clear that several catabolic and anabolic pathways are involved in regulating skeletal muscle mass and fibre size (Glass, 2005; Schiaffino et al. 2013), intracellular calcium ([Ca²⁺ _i]) can activate signalling pathways in the regulation of this process. Ca²⁺ signalling plays an important role in skeletal muscle growth and development (Berchtold et al. 2000). Previous studies have demonstrated that store‐operated Ca²⁺ entry (SOCE) participates in the Ca²⁺‐dependent transcriptional signalling that is required for myogenesis to occur (Darbellay et al. 2009), and mice lacking SOCE exhibit reduced muscle growth and decreased perinatal survival (Stiber et al. 2008; Vig et al. 2008).

Transient receptor potential canonical (TRPC) channels are non‐selective Ca²⁺ permeable channels that can be activated by various physical/chemical stimuli (Christensen & Corey, 2007; Nilius et al. 2007), including mechanical loading (Gomis et al. 2008; Inoue et al. 2009). TRPC1, the most abundant isoform in skeletal muscle, participates in SOCE in association with release‐activated calcium channel protein 1 (Orai1) and stromal interacting molecule 1 (STIM1) (Kim et al. 2009; Salido et al. 2011; Srikanth & Gwack, 2012). We have demonstrated previously that Ca²⁺ influx via TRPC1 activity is abnormal in dystrophic mdx mice (Vandebrouck et al. 2002; Yeung et al. 2005; Gervasio et al. 2008), and there is evidence that TRPC1‐mediated Ca²⁺ entry regulates skeletal myoblast migration and differentiation (Louis et al. 2008). Additionally, the expression of TRPC1 was markedly upregulated during myogenesis, and treatment with TRPC1‐siRNA reduced both SOCE and the expression of myogenic differentiation markers, which subsequently suppressed skeletal myogenesis (Formigli et al. 2009). Moreover, muscles from TRPC1 ^−/− mice display smaller myofibre CSAs, and this is accompanied by a decreased expression of myogenic factors, such as MyoD, Myf5, myogenin and myosin heavy chain (MHC) (Zanou et al. 2010). Importantly, the findings from a recent study (Benavides Damm et al. 2013) showed that myoblasts under conditions of simulated microgravity exhibit reduced TRPC1 expression concomitant with slowing of cell proliferation and delayed differentiation. The same study also demonstrated that blocking the TRPC1 channel arrests proliferation in a dose‐dependent manner (Benavides Damm et al. 2013). Collectively, these data strongly suggest that TRPC1 is required to respond to changes in mechanical load during skeletal myogenesis.

The study of the regulation of TRPC1 in vivo has rarely been reported; most studies largely comprise reports of in vitro phenomena. In previous studies, we showed that TRPC1 is dynamically expressed during myogenesis in C2C12 mammalian myoblasts; the expression of TRPC1 significantly increases during myogenesis and is detectable in differentiated myocytes and myotubes (Cheung et al. 2011). We also showed in vivo that TRPC1 expression and localization to the sarcolemma were greatly reduced after 14 days of hindlimb unloading. Furthermore, TRPC1 expression was not upregulated until the later stages of reloading (Zhang et al. 2014). Thus, our working hypothesis is that downregulation of TRPC1 may impair the recovery of muscle mass during regrowth following atrophy. In the present study, we employ the hindlimb unloading and reloading model to examine the involvement of TRPC1 in the regulation of muscle atrophy and regrowth, respectively. Given that muscle disuse preferentially affects anti‐gravity muscles, we used slow‐twitch soleus muscles for our experiments. To establish the physiological relevance of the present study, we examined whether in vivo manipulation of TRPC1 interfered with the muscle atrophy and regrowth processes. We also tested the potential link between TRPC1 and calcineurin (CaN) activity in the regulation of reloading‐associated muscle regrowth. The identification of TRPC1 as a regulator of unloading‐induced muscle wasting suggests that therapeutic interventions targeting TRPC1 or its downstream or upstream pathways would prevent or alleviate this debilitating condition.

Methods

Animals

Adult male mice (BALB/c; 8–10 weeks old) were obtained from The Chinese University of Hong Kong. All animal care and experimental procedures were approved by The Hong Kong Polytechnic University Animal Ethics Committee (ASESC no. 11/26). Mice were housed in cages, supplied with food and water ad libitum and maintained under a 12 : 12 h light/dark cycle at 20–22°C. All of the animals were housed for 1 week to acclimate to the laboratory conditions before commencing experimentation. Mice were randomly divided into five groups: (1) the hindlimb unloading for 14 day group (14U) and (2–5) the hindlimb unloading for 14 days followed by 3 days (3R), 7 days (7R), 14 days (14R) and 28 days (28R) of reloading groups. Weight‐bearing (WB), age‐matched groups were used as controls (n = 6 per group).

Hindlimb unloading and reloading procedures

Hindlimb unloading was mediated by tail suspension and was performed as described previously (Guo et al. 2012; Zhang et al. 2014). Briefly, orthopaedic traction tape was applied to the animal from the end of the tail to around two‐thirds of the tail proximally. The animal was suspended by a swivel harness attached at the top of the cage. The body of the animal was maintained at a 30º elevation such that only the forelimbs were able to maintain contact with the cage floor. The animal could move and freely access food and water within the cage during this procedure. The height of the animal was checked daily and adjusted if necessary. For the reloading groups, the animal was released from the unloading device and allowed to resume normal weight bearing.

Body weight and muscle wet mass

The body weight was recorded daily to document the body mass induced by muscle disuse, atrophy and regrowth. After the intervention, the mouse was killed by cervical dislocation, and the soleus muscles were isolated and weighed using a digital platform balance. The wet muscle weight was normalized to body weight (mg g^–1).

Section preparation and immunohistochemistry

Soleus muscles were cryoembedded in OCT compound (Tissue‐Tek, Vogel, Germany) before being snap frozen in pre‐chilled isopentane with liquid nitrogen. Cross‐sections of 5 μm thickness were cut from the mid‐belly of each muscle. The sections were fixed in ice‐cold 4% paraformaldehyde or acetone for 10 min and permeabilized with PBS containing 0.2% Triton X‐100 for 10 min. Sections were then blocked with 5% normal horse serum in PBS for 30 min. For immunohistochemistry against nuclear factor of activated T cells (NFAT), sections were blocked with the Fab fragment of goat anti‐mouse IgG (H+L) (dilution 1:100; Invitrogen, Carlsbad, CA, USA) at 4°C overnight. After blocking, sections were incubated with primary antibodies (rabbit monoclonal anti‐TRPC1, dilution 1:100, Santa Cruz Biotechnology, Santa Cruz, CA, USA; mouse monoclonal anti‐NFAT1, dilution 1:50, Santa Cruz Biotechnology; goat monoclonal anti‐dystrophin, dilution 1:100, Santa Cruz Biotechnology; mouse monoclonal anti‐MHCI, dilution 1:50, Developmental Studies Hybridoma Bank, Iowa City, IA, USA; mouse monoclonal anti‐MHCIIa, dilution 1:600, Santa Cruz Biotechnology) diluted in 5% normal horse serum at 4°C overnight. Subsequently, sections were further incubated with fluorescent dye‐conjugated secondary antibodies (Invitrogen) for 1 h at room temperature. Tetramethylrhodamine (TRITC)‐conjugated wheat germ agglutinin (WGA) was used for sarcolemmal localization experiments. The Alexa Fluor 568 F(ab′)₂ fragment of goat anti‐mouse IgG (H+L) was applied to anti‐NFAT1 staining. The washed sections were then mounted with 4′,6‐diamidino‐2‐phenylindole (DAPI) mounting medium (Vector Laboratories Inc., Burlingame, CA, USA) for nuclear counterstaining. Images were acquired with a fluorescence microscope (20× objective, Eclipse 80i; Nikon, Tokyo, Japan) and captured using Spot Advanced software (Diagnostic Instruments, Inc., Sterling Heights, MI, USA). Myofibre CSA was determined by dividing the total muscle area by the number of muscle fibres, the values of which were obtained by analysing the microscope‐acquired images in ImageJ (NIH, Bethesda, MD, USA). The number of MHCI‐positive fibres was counted on entire muscle sections and expressed per total number of muscle fibres. The percentage of dephosphorylated NFAT staining was expressed as the total number of NFAT‐positive nuclei per total number of muscle cells.

Quantitative RT‐PCR

Total RNA from soleus muscles was extracted and purified using the ReliaPrep RNA Tissue Miniprep System in accordance with the manufacturer's instructions (Promega Corporation, Madison, WI, USA). The concentration and purity of the RNA were evaluated by measuring absorbance at 230, 260 and 280 nm using a U‐0080D spectrophotometer (Hitachi High‐Technologies Corporation, Tokyo, Japan). A total of 0.5 μg of RNA from each sample was converted into cDNA in 20 μl reaction mixtures using the iScript cDNA synthesis kit (Bio‐Rad Laboratories, Hercules, CA, USA). Taqman‐MGB probes (Applied Biosystems, Foster City, CA, USA) were used for PCR amplification. For each sample, 1.2 μl of cDNA was amplified in 20 μl reaction mixtures containing 1 × ssoFast™ Probes Supermix (Bio‐Rad Laboratories) using the CFX Connect Real‐Time PCR Detection System (Bio‐Rad Laboratories). Glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) was used as an endogenous control. Relative changes in target gene mRNA expression were determined using the Livak (2^−ΔΔCt) method of analysis (Livak & Schmittgen, 2001). The mRNA expression levels at each time point were normalized to each corresponding control (i.e. negative control oligos in siRNA‐electroporated experiments or vehicle control in FK506‐treated experiments) in which the relative fold change was adjusted to 1.0 to facilitate comparison.

Western blot analysis

Soleus muscles were lysed at 4°C in radioimmunoprecipitation assay buffer (Cell Signaling Technology, Beverly, MA, USA) containing a protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN, USA), 1 mm 4‐(2‐aminoethyl)‐benzenesulphonyl fluoride (Merck, Darmstadt, Germany) and 2 mm dithiothreitol. Lysate concentrations were measured in duplicate using the Bradford Assay (Bio‐Rad Laboratories). Between 50 and 100 μg of total muscle proteins were run on SDS‐PAGE gels and electrotransferred to nitrocellulose membranes (Perkin Elmer Life Sciences, Foster City, CA, USA). Non‐specific binding was blocked by 5% non‐fat milk in Tris‐buffered saline containing 0.05% Tween 20 (Bio‐Rad Laboratories) for 1 h. Primary antibody incubations (mouse monoclonal anti‐TRPC1, dilution 1:1000; Santa Cruz Biotechnology; mouse monoclonal anti‐CaN, dilution 1:200; Sigma, St Louis, MO, USA) were performed followed by secondary antibody immunoreactions (IR Dye 800CW goat anti‐mouse IgG (H + L), dilution 1:15,000; LI‐COR Biosciences, Lincoln, NE, USA). GAPDH (mouse monoclonal anti‐GAPDH, dilution 1:5000; Cell Signaling Technology) was used as the endogenous control. Immunoreactive bands were detected using the Odyssey® Infrared Imaging System and quantified with Odyssey® image analysis software (LI‐COR Biosciences). The expression levels of TRPC1 and CaN in each sample were normalized to their respective GAPDH levels. The fold change of the controls was adjusted to value 1 to facilitate comparison.

Microinjection of siRNA and electroporation

WB mice were randomly divided into siRNA‐ or negative control oligos‐electroporated groups. The sequence 5′‐GCAUCGUAUUUCACAUUCU‐3′ (Invitrogen) was selected to target TRPC1 mRNA based on previous studies (Sabourin et al. 2009; Sabourin et al. 2012). The soleus muscle of one limb was injected with siRNA targeting TRPC1, whereas the contralateral limb was treated with negative control siRNA (Invitrogen), which does not target any known genes. Injections were performed under a stereomicroscope (64× magnification; Leica Microsystems MZ16, Wetzlar, Germany). Mice were deeply anaesthetized with 4% isoflurane in O₂, and the hindlimbs were shaved and placed in a prone position. For each mouse, an incision was made through the fascia along the posterolateral edge of the gastrocnemius muscle, and the soleus muscle was carefully isolated by blunt dissection without disturbing the nearby blood vessels, nerves and muscles. A 34‐gauge Hamilton syringe (Hamilton Company, Reno, NV, USA) was inserted near the distal myotendinous junction of the soleus muscle, and 1.5 μl of PBS containing 1.1 ng of siRNA was slowly injected evenly along the longitudinal axis of the soleus muscle at the same time as the syringe was slowly withdrawn. After injection, the two‐needle array electrodes (BTX‐Harvard Apparatus Inc., Holliston, MA, USA) were horizontally inserted into both ends of the soleus muscle. The muscle was electroporated by applying six pulses (50 ms each in 200 ms intervals) of 50 V at a frequency of 1 Hz using the BTX Electro Square Porator (ECM 830 Electroporation System; BTX‐Harvard Apparatus Inc.). After electroporation, the incision was surgically closed, and the mouse was returned to its cage and carefully monitored until it had fully recovered from anaesthesia. At days 4 and 7 post‐electroporation (n = 6 per group), mice were killed and the soleus muscles were harvested. Similar electroporations were also performed on mice that experienced 7 days of reloading (7R), and muscles were harvested at days 4 and 7 post‐electroporation (n = 12 per group).

Drug administration in vivo

Drugs were administered on the first day of reloading (1R). A total of 2 mg kg^–1 day^–1 FK506 diluted in 30% ethanol and 70% saline (Sigma) was injected i.p. twice daily (Bodine et al. 2001). Cyclosporine A (Sandimmune, Novartis, Switzerland) dissolved in 67% cremphor EL (Merck) and 33% ethanol was administered at a dose of 30 mg kg^–1 day^–1 i.p. twice daily (Pandorf et al. 2009; Banzet et al. 2012). Injections of FK506 and cyclosporine A vehicles were used as controls. Soleus muscles from drug‐ or vehicle‐injected mice were harvested at days 3, 7 and 14 of reloading (n = 3 per group). The dosages of FK506 and cyclosporine A effectively suppressed CaN activity without any side effects on overall body health and muscle growth.

Statistical analysis

Data were analysed using Prism 4.0 software (GraphPad, La Jolla, CA, USA). Comparisons between groups were analysed using one‐ or two‐way ANOVA, followed by post hoc Bonferroni tests. P < 0.05 was considered statistically significant. All values are expressed as the mean ± SEM.

Results

Effects of unloading and reloading on TRPC1 expression in mouse soleus muscles

The 14 day unloading protocol reduced soleus muscle mass (Fig. 1 A) and myofibre CSA (Fig. 1 B) compared to WB age‐matched controls (P < 0.001, n = 6). There was also a significant reduction in MHCI mRNA expression to 25.6 ± 4.09% of control (P < 0.001) (Fig. 1 C). To investigate whether the number of MHCI‐positive myofibres was reduced, image analysis and quantification of immunohistochemical staining were performed. These analyses revealed that mice subjected to 14 days of unloading (14U) exhibited a decrease in the percentage of MHCI‐positive myofibres compared to controls (P < 0.001, n = 6) (Fig. 1 D). The subsequent 3 and 7 days of reloading (3R and 7R, respectively) led to a gradual recovery of muscle mass equal to control levels by 14R, although myofibre CSA and MHCI did not return to control levels until 28R. Furthermore, western blot analysis showed that TRPC1 levels were significantly reduced by 14U (P < 0.001). TRPC1 expression remained low at 3R and 7R (all P < 0.001, n = 6) and did not return to control levels until 28R (Fig. 1 E). Collectively, these observations are consistent with previous studies (Dumont & Frenette, 2010; Washington et al. 2011; Zhang et al. 2014) and establish the process by which skeletal muscles adapt to changes in mechanical stimuli.

Figure 1. Effect of unloading and reloading in mouse soleus muscles .

Animals were randomized into five experimental groups: hindlimb unloading for 14 days (14U) and those with 14 days of unloading followed by 3, 7, 14 or 28 days of reloading (3R, 7R, 14R, 28R). Age‐matched WB animals were used for controls (WB control). A, muscle mass normalized to body weight (mg g^–1) during the experimental period. Animals in all groups were similar at the beginning of the experiment. B, muscle fibre CSA throughout the experimental period. C, mRNA expression of slow MHC (MHCI) normalized to the WB control group. D, percentage of MHCI‐positive fibres per total number of muscle fibres. Inset: immunostaining for MHCI (blue) and MHCIIa (green) in soleus muscle sections. Muscles were counterstained with TRITC‐WGA to visualize the plasma membrane. Scale bar = 50 μm. E, changes in TRPC1 protein expression were normalized to GAPDH and expressed as a percentage of the WB control group. Values represent the mean ± SEM (n = 6 per group). *P < 0.05, **P < 0.01, ***P < 0.001 compared to WB controls.

Effects of TRPC1 silencing in vivo

In an attempt to define the direct role of TRPC1 during muscle disuse, atrophy and regrowth, we selectively knocked down the expression of TRPC1 using RNAi. Microinjection of TRPC1‐siRNA followed by electroporation into WB soleus muscles (n = 6) led to a significant downregulation of endogenous TRPC1 both at the RNA and the protein level by day 4 (mRNA: 41.5 ± 5.20%; protein: 69.3 ± 3.73%; P < 0.001 for both) and day 7 (mRNA: 45.4 ± 1.76%, protein: 54.1 ± 5.12%; P < 0.001 for both) post‐transfection (Fig. 2 A and B). Concomitantly muscle mass (Fig. 2 C, Ea) and myofibre CSA (Fig. 2 D and Eb‐c) were significantly reduced in TRPC1‐knockdown muscles compared to negative controls (control siRNA‐injected muscles). For example, at day 7 post‐transfection, muscle mass and CSA were reduced to ∼52.1% and ∼53.5%, respectively, of control (P < 0.001 for both).

Figure 2. Effects of TRPC1 silencing in vivo .

siRNA targeting TRPC1 (TRPC1‐siRNA) or negative‐control siRNA (control) was introduced into soleus muscles by microinjection followed by electroporation. mRNA (A) and protein expression (B) of TRPC1 were assessed at days 2, 4 and 7 post‐electroporation. Comparisons were made with negative control oligos on the same day of post‐electroporation and the fold change of the negative control was adjusted to 1. Phenotypic changes were also assessed. C, muscle mass normalized to body weight. D, analysis of muscle fibre CSA in haematoxylin and eosin‐stained sections. E, gross morphological changes of the soleus muscle at day 7 post‐electroporation (Ea). Scale bar = 2 mm. TRITC‐WGA‐stained transverse sections of control (Eb) and TRPC1‐siRNA‐expressing (Ec) muscles at day 7 post‐electroporation. Scale bar = 200 μm. Values represent the mean ± SEM (n = 6 per group). ***P < 0.001 compared to controls.

Effects of TRPC1 silencing during recovery from muscle disuse and atrophy

We investigated whether TRPC1 is involved in reloading‐induced muscle regrowth. To test this, we suppressed the recovery of TRPC1 expression during the reloading process by injecting TRPC1‐siRNA into soleus muscles at 7R (n = 12), which is when TRPC1 protein expression was the lowest over the entire unloading–reloading period. In TRPC1‐siRNA‐targeted 7R soleus muscles, TRPC1 mRNA and protein expression levels were decreased to 34.1 ± 2.30% (P < 0.001) and 68.3 ± 1.51% (P < 0.001), respectively, at day 4 post‐transfection (i.e. 11 days of reloading). At day 7 post‐transfection, which represents 14 days of reloading (14R), TRPC1 remained downregulated both at the RNA and protein level (mRNA: 33.5 ± 5.10%, protein: 52.8 ± 2.10%; P < 0.001 to both, Fig. 3 A and B).

Figure 3. Effects of TRPC1 silencing during recovery from muscle disuse and atrophy .

After 7 days of reloading (7R), TRPC1‐siRNA was introduced into the soleus muscles. This was compared to negative control siRNA (control). mRNA (A) and protein expression (B) of TRPC1 was assessed at days 4 and 7 post‐transfection (i.e. days 11 and 14 of reloading; 11R and 14R respectively). C, analysis of muscle fibre CSA in haematoxylin and eosin‐stained sections. D, percentage change of mRNA expression of MHCI relative to negative control siRNA. E, immunostaining for MHCI (blue) and MHCIIa (green) in soleus muscle sections at day 7 post‐transfection (i.e. 14 days of reloading, 14R). Muscles were counterstained with TRITC‐WGA (red) to visualize the plasma membrane. Negative control (Ea) was compared with TRPC1‐siRNA (Eb). Scale bar = 50 μm. Values represent the mean ± SEM (n = 12 per group). ***P < 0.001 compared to 7R.

The resumption of WB activity was accompanied by an increase in myofibre CSA and upregulation of MHCI (Fig. 1 C and D). These variables provide a direct measure of the recovery process after unloading. The myofibre CSA recovery was significantly impaired in the TRPC1‐siRNA‐expressing muscles at 11 and 14 days of reloading compared to the siRNA negative control (P < 0.001 for all) (Fig. 3 C). Figure 3 D shows the mRNA expression profile of MHCI. Silencing TRPC1 led to a large and significant reduction in slow MHCI isoform expression. For example, at day 7 post‐transfection, which represents reloading and recovery for 14 days (14R), MHCI mRNA expression levels were 36.4 ± 5.55% (P < 0.001) compared to untreated 14R muscles. Similarly, the number of MHCI‐positive myofibres was significantly lower at day 7 post‐transfection (∼61% of control, P < 0.01) (Fig. 3 E).

Effects of unloading and reloading on the expression of CaN and NFAT in mouse soleus muscles

The above results indicate that TRPC1 plays a modulatory role in reloading‐mediated muscle regrowth and regulation of MHCI. Previous studies have shown that TRPC subunits are involved in CaN‐NFAT activation in cardiac muscles (Bush et al. 2006; Nakayama et al. 2006; Pigozzi et al. 2006; Morales et al. 2007) and have highlighted the distinct role of the CaN‐dependent pathway in modulating muscle regeneration (Mitchell et al. 2002; Sugiura et al. 2005). These findings prompted us to investigate the regulatory role of TRPC1 in CaN‐NFAT signalling during muscle regrowth following muscle disuse and atrophy.

We first examined changes in CaN and NFAT1 expression during unloading and reloading. The expression levels of CaN were upregulated following 14 days of mechanical unloading, peaked at day 7 of reloading (P < 0.001) and remained upregulated at 14R (P < 0.001; n = 4) (Fig. 4 A). We then examined NFAT nuclear localization as an indicator of CaN activity using immunohistochemistry. We counted the number of nuclei located inside dystrophin‐positive plasma membranes. Following 14U, there was an increase trend in NFAT immunoreactive nuclei, although this did not reach significance (P > 0.05). The percentage of NFAT immunoreactive nuclei was significantly increased at 7R compared to WB controls (P < 0.001) and 14U (P < 0.001) (Fig. 4 B and C).

Figure 4. Effects of unloading and reloading on the expression of CaN and NFAT in mouse soleus muscles .

A, western blot analysis for CaN protein expression during the unloading and reloading process normalized to GAPDH and the WB control group. ***P < 0.001 compared to WB controls; ^# P < 0.01 compared to 14U. B, percentages of NFAT1‐positive myonuclei per total number of cells. C, merged immunohistochemical images of NFAT1‐positive (green) myonuclei (blue, DAPI counterstaining) and anti‐dystrophin (red, membrane marker) in soleus muscle cross sections. Scale bar = 20 μm. Values represent the mean ± SEM (n = 4 per group). ***P < 0.001 compared to WB controls; ^### P < 0.001 compared to 7R.

TRPC1 expression is dependent on CaN activity

Because the increases in CaN expression and NFAT nuclear localization precede the upregulation of TRPC1 expression during the reloading process (compare 7 vs. 14 days of reloading), we considered the possibility that TRPC1 function may be regulated by the CaN‐NFAT pathway. To test this, we examined the effects of the CaN inhibitors, FK506 and cyclosporine A, on TRPC1 expression during reloading (n = 3). Using the expression of regulator of calcineurin 1 (RCAN1) transcripts as a readout for successful CaN inhibition (Koulmann et al. 2008; Pandorf et al. 2009; Banzet et al. 2012), we found that FK506 treatment significantly decreased RCAN1 mRNA expression at 3R, 7R and 14R (P < 0.001 for all) (Fig. 5 A). In addition, the mRNA expression of MHCI was also significantly reduced (7R: 13.5 ± 5.84%, 14R: 46.4 ± 10.7%; P < 0.001 for both) (Fig. 5 B). Furthermore, FK506‐mediated inhibition of CaN delayed the recovery of myofibre CSA during reloading; the mean CSA at 7R and 14R was 79.8% and 82.1%, respectively, compared to untreated controls (P < 0.001 for both) (Fig. 5 C). Importantly, the expression of TRPC1 transcript was significantly reduced (Fig. 5 D) after FK506 treatment during the recovery process at 7R (37.4 ± 11.0% of control; P < 0.001) and 14R (50.0 ± 11.1% of control; P < 0.01). We further examined TRPC1 protein levels after CaN inhibition and found that FK506 treatment reduced the expression of TRPC1 to 70.0 ± 3.71% relative to controls at 14R (P < 0.01) (Fig. 5 E). Furthermore, in situ detection of TRPC1 revealed that FK506 treatment reduced the TRPC1 immunoreactivity in soleus muscles at 14R (Fig. 6), confirming the results of the western blot analysis. For all of the experiments using FK506 as an inhibitor of CaN, parallel experiments were also conducted using cyclosporine A, another CaN inhibitor. Similar results were observed (data not shown), suggesting that inhibition of CaN resulted in decreased expression of TRPC1, and that TRPC1 silencing consequently impedes the process of muscle regrowth after mechanical unloading.

Figure 5. FK506‐mediated inhibition of CaN activity decreased TRPC1 expression at days 3, 7 and 14  of reloading (3R, 7R and 14R, respectively) .

Data were normalized to vehicle‐injected controls. A, RCAN1 mRNA expression. B, percentage change of mRNA expression of MHCI relative to negative control siRNA after FK506 treatment. C, changes in myofibre CSA. mRNA (D) and protein expression (E) of TRPC1 after CaN inhibition. Data were normalized to GAPDH and expressed as a percentage of the vehicle control group. Comparisons were made with the vehicle control group on the same day of reloading and the fold change of the vehicle control was adjusted to 1. Values represent the mean ± SEM (n = 3 per group). **P < 0.01, ***P < 0.001 compared to vehicle controls.

Figure 6. Fluorescence images of soleus muscles immunostained for TRPC1 expression .

Soleus muscle cross sections labelled with anti‐TRPC1 antibody (green). Cell membranes were counterstained with TRITC‐WGA (red). Scale bar = 50 μm.

Discussion

In the present study, we show that the expression of TRPC1, a component of SOCE, is sensitive to mechanical unloading and reloading in skeletal muscle. We provide evidence indicating that manipulation of TRPC1 using siRNA hampers the recovery of muscle mass, myofibre CSA and fibre‐type transitions, suggesting the requirement of TRPC1 in skeletal muscle regrowth following muscle atrophy. We observe sequential activation of CaN and TRPC1 upon different stages of reloading‐induced muscle regrowth. Furthermore, inhibition of CaN signalling downregulated TRPC1 expression in both WB and reloaded muscles, suggesting that CaN appears to serve, in part, in the regulation of TRPC1 expression during the process of muscle regrowth.

The mechanosensitivity of TRPC1 was first revealed in frog oocytes (Maroto et al. 2005) with channel characteristics similar to those of stretch‐activated channels described in mammalian skeletal muscle. Previous studies have investigated the potential involvement of TRPC1 in the stretch‐induced muscle damage model (Gervasio et al. 2008; Zhang et al. 2012) and a role for TRPC1 in mechanotransduction induced by gravitational mechanical unloading was reported in two recent studies (Benavides Damm et al. 2013; Zhang et al. 2014). Using C2C12 myoblasts under simulated microgravity conditions, Benavides Damm et al. (2013) reported that a reduction in TRPC1 expression accompanied the retardation of cell proliferation. Similarly, in a classical hindlimb unloading model, we showed that TRPC1 was dynamically expressed in anti‐gravity soleus muscles during mechanical unloading and reloading (Zhang et al. 2014). Both the in vitro and in vivo models suggest that impairment of TRPC1‐mediated calcium signalling plays a pivotal role in unloading‐induced muscle atrophy.

Because TRPC1 expression changes with classical phenotypic alterations (i.e. changes in muscle mass, CSA and force), TRPC1 could be involved in the process of muscle atrophy and regrowth. In the present study, we confirm this hypothesis by downregulating TRPC1 expression using RNAi in vivo. Gene silencing experiments indicate that suppressing TRPC1 at the gene level, whether under WB conditions or during mechanical reloading, results in significant reductions in muscle mass and myofibre CSA. These findings were consistent with previous data demonstrating that muscles in TRPC1^−/− mice exhibited smaller myofibre CSA, reduced myofibrillar protein content and less force per unit area (Zanou et al. 2010). The only notable discrepancy between the two datasets is that, in the present study, the proportion of MHCI‐positive myofibres was significantly reduced in TRPC1‐knockdown soleus muscles, whereas this was not the case in the soleus muscles of TRPC1^−/− mice (Zanou et al. 2010). Although the present study focused on reloading‐induced muscle regrowth, for which TRPC1 is hypothesized to play a role, the primary focus of the study by Zanou et al. (2010) was to examine fibre‐type proportions in TRPC1^−/− mice vs. WB models, which probably accounted for this discrepancy. It is also worth noting that the siRNA‐knockdown model, although not completely removing TRPC1 expression, is preferable to the TRPC1^−/− model in the present study because (i) siRNA knockdown can be introduced at a specific time point of the reloading experiment and (ii) in vivo electroporation of siRNA allows knockdown to the targeted soleus muscle without interfering adjacent muscles. We consider this approach to be more physiologically relevant for understanding the function of TRPC1 in the muscle regrowth process.

Other studies provide a mechanistic view of how TRPC1 contributes to muscle regrowth. First, a recent report showed that TRPC1 is highly expressed in satellite cells, the predominant source for adult muscle regeneration (Liu & Schneider, 2014). It was shown that TRPC1‐mediated elevation of [Ca²⁺]_i in satellite cells can be triggered by the addition of fibroblast growth factor 2. This is followed by nuclear translocation of NFATc2 and NFATc3, which subsequently leads to an increase in MyoD‐positive nuclei, a marker for satellite cell activation (Liu & Schneider, 2014). Second, primary myoblasts isolated from TRPC1^−/− mice show a reduced expression of myogenic factors, including MyoD, Myf5, myogenin and MHC, indicating that mice lacking TRPC1 exhibit impaired myogenic differentiation possibly as a result of a dysregulated PI3K/Akt/mammalian target of rapamycin/p70S6K pathway (Zanou et al. 2012). Furthermore, TRPC1^−/− myoblasts exhibit diminished transient Ca²⁺ influx resulting in a reduced calpain activity that significantly delays myotube fusion (Louis et al. 2008). Taken together, it is possible that TRPC1 plays a role in the regulation of muscle regrowth following disuse. Further experiments are clearly needed to determine this regulatory mechanism.

Dynamic expression of CaN during unloading and reloading has also been reported previously. Similar to Childs et al. (2003), we observed no change in the level of CaN at the end of unloading. However, Oishi et al. (2008) reported a 32% reduction of CaN levels at the end of 14 days of unloading compared to controls. Even though there is sufficient evidence to suggest that CaN is required for muscle growth at specific stages during the regrowth process, the role of CaN in the maintenance of muscle mass in the atrophic process is not so clear. The interaction of CaN activation and PGC‐1α via MEF2 and NFAT is assumed to be the primary signalling process in attenuating muscle atrophy (Long et al. 2007; Banzet et al. 2012). It has also been suggested that CaN protects muscle atrophy by increasing the proportion of slow fibres (Lin et al. 2002). One other mechanism whereby CaN prevents muscle loss via activation of miR‐23a, thus inhibiting atrogin‐1 and MuRF1, has also been proposed (Hudson & Price, 2013).

After 14 days of unloading, we observed a temporal increase in CaN, and the level was significantly upregulated at day 7  of recovery. This temporal pattern was similar to that reported previously (Sugiura et al. 2005; Oishi et al. 2008). It has been shown that CaN regulates muscle remodelling from muscle disuse and atrophy mainly via regulation of fibre‐type transition (Sugiura et al. 2005) and myofibre size (Oishi et al. 2008) in the later stages of regrowth. Consistent with these reports, the use of the CaN inhibitors, FK506 and cyclosporine A, demonstrated that the suppression of CaN activity remarkably alters fibre‐type transition and myofibre CSA. Furthermore, we also observed that upregulation of CaN preceded that of TRPC1 during hindlimb reloading, raising the possibility that CaN may be involved in regulating TRPC1 expression. Accordingly, we performed CaN inhibition during recovery and observed a reduction in TRPC1 protein expression at 14 days after reloading. TRPC1 is abundantly expressed in blood vessels and it is also known that endothelial cells are highly sensitive to microgravity and undergo morphological, biochemical and functional changes (Maier et al. 2015). Thus, the expression of TRPC1 in muscles and in blood vessels would be reflected in the protein analysis. A recent study also provided evidence suggesting that CaN contributes to STIM1 phosphorylation in regulating SOC entry in endothelial cells (Vasauskas et al. 2014). Perhaps a more important finding is that immunohistochemical staining (Fig. 6) demonstrates a decreased expression of TRPC1 in FK506‐treated muscle fibres compared to vehicle control.

CaN has been shown to regulate the expression, or activities, of other TRPC receptor subtypes, among which, TRPC3 and TRPC6 have been studied extensively (Rosenberg et al. 2004; Kuwahara et al. 2006; Nakayama et al. 2006). Constitutive expression of active forms of CaN leads to NFATc1 translocation from the cytoplasm to the nucleus, and nuclear NFATc1 binds to and increases the activity of the TRPC3 promoter, resulting in the upregulation of TRPC3 expression in skeletal muscle (Rosenberg et al. 2004). Transgenic mice overexpressing TRPC3, on the other hand, exhibit cardiomyopathy and abundant SOCE with increased CaN‐NFAT activation in cardiomyocytes (Nakayama et al. 2006). Transgenic mice harbouring an α‐MHC‐CaN transgene demonstrate profound cardiac hypertrophy, and it was shown that the hypertrophic phenotype is mediated by TRPC6, which is directly regulated by CaN‐NFAT signalling (Kuwahara et al. 2006). Overexpressing TRPC6 in the heart further activates the CaN‐NFAT pathway, indicating the existence of a positive regulatory circuit between TRPC6 and CaN‐NFAT signalling (Kuwahara et al. 2006). Calcineurin was shown to modulate TRPC1 expression in smooth muscle cells (Morales et al. 2007). Furthermore, phenotypic features of cardiac hypertrophy subjected to transverse aortic constriction were present in wild‐type mice but not in TRPC1^−/− mice, and CaN‐NFAT activity in TRPC1^−/− mice was less active than in wild‐type mice, suggesting that CaN‐NFAT signalling‐mediated cardiac hypertrophic remodelling also involves TRPC1 (Seth et al. 2009). Collectively, these data highlight the close relationship between TRPC1 and CaN‐NFAT signalling during skeletal (and cardiac) muscle remodelling. Expanding on these findings, our data have identified the possible link between CaN and TRPC1 during reloading‐induced skeletal muscle regrowth. Despite the regulatory role of CaN on TRPC1, we do not exclude the possibility that a positive feedback mechanism may exist in which the influx of TRPC1 may further activate CaN signalling, although this notion requires further confirmation by additional experiments.

TRPC channels exist as either homomeric or multimeric complexes of different TRPC isoforms in different cell types. Although TRPC1 and TRPC4 have been shown to form complexes with scaffolding proteins in skeletal myoblasts and myotubes (Sabourin et al. 2009), TRPC4 expression in adult skeletal muscle is barely detectable (Flockerzi et al. 2005). Additionally, TRPC3 may only play a very minor role in reloading‐induced muscle regrowth considering its static level of expression during periods of unloading and reloading in our previous study (Zhang et al. 2014). TRPC6 is probably a potential candidate for forming heteromeric channels with TRPC1, although the role of TRPC6 in skeletal muscle has not been examined thus far. Dynamic associations of TRPC1, Orai1 and STIM1 are involved in activation of SOCE (Huang et al. 2006; Yuan et al. 2007; Cheng et al. 2008). Although the precise role of TRPC1/Orai1/STIM1‐mediated SOCE in adult skeletal muscle is not completely understood, myoblast differentiation was shown to require TRPC1 activity and STIM1‐ and Orai1‐mediated SOCE (Louis et al. 2008; Kiviluoto et al. 2011; Zanou et al. 2012). Moreover, both muscle‐specific STIM1^−/− mice and dominant‐negative Orai1 mice exhibit reductions in muscle mass and myofibre CSA (Li et al. 2012; Wei‐Lapierre et al. 2013), similar to the phenotypes observed in TRPC1‐siRNA‐expressing soleus muscles reported in the present study. Taken together, these data indicate the importance of STIM1/Orai1/TRPC1‐mediated SOCE in adult muscle regrowth. Upon store depletion, the Ca²⁺‐binding STIM1 residing in the endoplasmic reticulum moves towards the plasma membrane and activates Orai1 SOCE channels (Yuan et al. 2009). STIM1‐dependent SOCE was shown to be important in mediating cardiac and skeletal muscle hypertrophy, both of which also require CaN‐NFAT signalling (Li et al. 2012; Luo et al. 2012). In the present study, we demonstrate that the dynamic changes in TRPC1 expression during mechanical unloading and reloading are possibly regulated by CaN, which is in line with the literature.

In summary, in the present study, we have described a novel role for TRPC1 in reloading‐induced muscle regrowth and remodelling after mechanical reloading in vivo. Additionally, in this context, CaN appears to be part of the signalling pathway involved in mediating the expression of TRPC1. Further studies will be required to address and identify the downstream cellular and molecular pathways that mediate mechanically sensitive muscle regrowth.

Additional information

Competing interests

The authors declare that they have no competing interests.

Funding

This study was supported by the Hong Kong Research Grants Council General Research Fund (PolyU 5636/13 M to EWY) and the Hong Kong Jockey Club Joint PhD Scheme (RTQ5 to EWY).

Author contributions

EWY, SSY and K‐KC conceived and designed the experiments. LX and K‐KC performed the experiments. EWY, SSY and K‐KC analyzed and interpreted the data. EWY wrote the manuscript with contribution by all authors.