Calcineurin plays a modulatory role in loading-induced regulation of type I myosin heavy chain gene expression in slow skeletal muscle

Abstract

The role of calcineurin (Cn) in skeletal muscle fiber-type expression has been a subject of great interest because of reports indicating that it controls the slow muscle phenotype. To delineate the role of Cn in phenotype remodeling, particularly its role in driving expression of the type I myosin heavy chain (MHC) gene, we used a novel strategy whereby a profound transition from fast to slow fiber type is induced and examined in the absence and presence of cyclosporin A (CsA), a Cn inhibitor. To induce the fast-to-slow transition, we first subjected rats to 7 days of hindlimb suspension (HS) + thyroid hormone [triiodothyronine (T₃)] to suppress nearly all expression of type I MHC mRNA in the soleus muscle. HS + T₃ was then withdrawn, and rats resumed normal ambulation and thyroid state, during which vehicle or CsA (30 mg·kg⁻¹·day⁻¹) was administered for 7 or 14 days. The findings demonstrate that, despite significant inhibition of Cn, pre-mRNA, mRNA, and protein abundance of type I MHC increased markedly during reloading relative to HS + T₃ (P < 0.05). Type I MHC expression was, however, attenuated by CsA compared with vehicle treatment. In addition, type IIa and IIx MHC pre-mRNA, mRNA, and relative protein levels were increased in Cn-treated compared with vehicle-treated rats. These findings indicate that Cn has a modulatory role in MHC transcription, rather than a role as a primary regulator of slow MHC gene expression.

a broad range of contractile properties, primarily attributed to diversity of the motor protein myosin heavy chain (MHC), can exist in different muscle fibers. Four isoforms of MHC (I, IIa, IIx, and IIb), each encoded by a distinct gene, can be expressed in adult skeletal muscle. Intrinsic differences in the ATPase (and shortening velocity) properties of the MHC isoforms have led to the classification of slow and fast fiber types in muscle (5). The characteristics of a “slow” muscle fiber arise from several intersecting variables that together contribute to the slow muscle phenotype. The slow muscle fiber is characterized by a relatively high mitochondrial content, high oxidative enzyme capacity, high levels of myoglobin, presence of sarcoplasmic reticulum Ca²⁺-ATPase (SERCA2a), and abundance of type I MHC, which has low ATPase activity. Type I MHC, however, is the primary driver of the slow contractile phenotype by virtue of the its low ATPase activity. Thus the slow muscle consists of a selective group of proteins/enzymes that results functionally in a relatively nonfatigable muscle suited to sustained contractions, which provide postural support and locomotion/movement spanning a broad range of intensity.

In many cell types, intracellular Ca²⁺ regulates the serine/threonine protein phosphatase calcineurin (Cn). Significant effort has been expended in identifying the signaling pathways involved in activating the slow fiber phenotype in skeletal muscle. Much of this effort has focused on Ca²⁺-activated Cn signaling, which has been shown to result in transcriptional activation of genes associated with slow fiber types. This has been shown to be mediated by the interactions of nuclear factor of activated T cells (NFATc1) and myocyte enhancer factor-2 at gene promoters (10, 48). It was hypothesized that the type of motor nerve activity characteristic of slow muscle activity, which results in a sustained increase in intracellular Ca²⁺, would activate Cn (10). Indeed, endurance-type exercise has been shown to result in large increases in Cn activity in skeletal muscle (16).

Mechanistic studies have demonstrated a role for Cn in the transcriptional regulation of slow fiber-type genes in cell culture. However, in vivo studies have demonstrated a disconnect in the purported function of Cn in skeletal muscle. Data from some reports show that Cn signaling is implicated in a slow muscle gene program in vivo by showing that inhibition of Cn with cyclosporin A (CsA) induces a shift from slow to fast fiber type (4, 6, 10, 12, 24, 39). Terminology referring to “slow-to-fast shifts,” or vice versa, can be somewhat ambiguous, because the terminology does not always include the slow type I MHC and may, instead, include shifts from type IIa to type IIx/IIb MHC or describe other measures of fiber type, such as myofibrillar ATPase. Accordingly, other data have shown that Cn has no effect on type I MHC (8, 12, 13, 15, 21, 22, 24) or on type I fiber type (4, 7). Transgenic gain- or loss-of-function studies have also been utilized to examine the function of Cn in mouse skeletal muscle. In these studies, Cn was shown to have a limited, if any, effect (13, 32, 33) or large effects (29, 32, 33) on type I MHC, depending on the targeted gene and the muscle studied. Some of the above cited studies report data that support a role for Cn in type I MHC expression under some conditions and no role under other conditions (4, 12, 13, 24, 33). Overall, there is a lack of consistent findings. Thus an explicit role for Cn in the activation of the slow contractile protein type I MHC remains controversial.

At least three factors may have contributed to the inconsistent findings regarding the role of Cn in fiber-type remodeling, specifically concerning its role in MHC gene expression. 1) There is a lack of consistent and highly accurate methodologies for measurement of MHC composition. 2) Various models in which MHC may respond differently to Cn signaling (i.e., cell culture, postnatal development, muscle regeneration, functional overloading, and normal muscle loading in adult animal and various animal species) have been used. 3) No in vivo model has been utilized that has resulted in an upregulation of type I MHC that is sufficiently robust to allow one to tease out the contribution of Cn to the transcriptional activation of type I MHC.

Therefore, the focus of this report was to determine whether Cn is a major regulator in the induction of the slow type I MHC in adult rat slow muscle. In this context, we employed a novel strategy using hindlimb suspension (HS) plus exogenous thyroid hormone [triiodothyronine (T₃)] treatment, to abolish transcription and pretranslational events of type I MHC expression, followed by 7 and 14 days of reloading of the muscle with normal ambulation and a euthyroid state. Type I MHC is strongly upregulated by reloading relative to the HS + T₃ treatment procedure. Therefore, we present a regulatory environment where the type I MHC gene, initially repressed, can be analyzed under full transcriptional activation in the context of Cn inhibition.

Therefore, in this study, we tested the hypothesis that Cn plays a central role in the induction of type I MHC gene expression during the reloading phase following its repression. This was accomplished via inhibition with CsA treatment at an optimal dose that does not disrupt the normal health of adult rodents. The central findings of this study suggest that CsA has a modulatory effect on type I MHC transcription, whereas it has a greater impact on shifts to type IIa and IIx MHC in this particular model.

METHODS

Animal procedures.

Female Sprague-Dawley rats (140–150 g) were used for all experiments. Animals were housed in a temperature- and light-controlled environment (i.e., 12:12-h light-dark cycle). Food and water were provided ad libitum, and all procedures were approved by the Institutional Animal Care and Use Committee. Animals were divided into six weight-matched experimental groups (n = 8 per group): 1) normal control (Con), 2) HS + T₃ treatment (HS + T₃), 3) HS + T₃ followed by 7 days of muscle reloading by normal ambulation (7d RL), 4) 7d RL + CsA treatment (7d RL + CsA), 5) HS + T₃ followed by 14 days of muscle reloading by normal ambulation (14d RL), and 6) 14d RL + CsA treatment (14d RL + CsA). CsA treatment began at the time that reloading was initiated. CsA (Sandimmune, Novartis) was diluted daily to 15 mg/ml in 0.9% saline from 50 mg/ml stock and administered at 15 mg/kg ip twice a day, for a total of 30 mg·kg⁻¹·day⁻¹. Vehicle, consisting of an equivalent dose of CsA vehicle (67% cremphor EL and 33% ethanol) diluted in 0.9% saline, was administered intraperitoneally twice per day to 7d RL and 14d RL groups. HS was carried out for 7 days, which was shown in prior experiments to be sufficient to induce measurable alterations in the endogenous MHC gene expression (31). At the end of the experiment, the rats were euthanized, and the muscles were rapidly removed, weighed, and frozen at −80°C for later analysis.

HS protocol.

The HS model employed a tail-traction method using a noninvasive tail-casting procedure described previously (45). The technique utilized a swivel-harness system incorporated into the casting materials, which was attached to a hook at the top of the cage. The hook was adjusted to allow only the front legs of the animal to reach the floor of the cage. Suspended animals used their front legs to move about the cage to obtain food and water.

CsA dose and myocyte-enriched Cn-interacting protein-1 analysis.

We performed a separate experiment to determine the optimal dose of CsA. Three separate groups of rats were treated with vehicle or with 20, 30, or 50 mg·kg⁻¹·day⁻¹ CsA, with each dose divided into two equal intraperitoneal injections per day, spaced ∼10–12 h apart, for 7 days (n = 6–8 rats per group) under normal loading conditions. Myocyte-enriched Cn-interacting protein-1 (MCIP-1, or RCAN-1) is regulated by the Cn-NFAT pathway, and its mRNA levels were used as an indicator of Cn activity (49). This approach to assess Cn activity has been reported by others (21–23). The animals from these experiments were also used to assess the effect of CsA on MHC gene expression during normal loading in the soleus muscle.

MCIP-1 mRNA was significantly decreased by each of the three CsA doses compared with vehicle (Fig. 1). There were no significant differences among the three CsA groups. MCIP-2 mRNA levels were also decreased with each CsA dose, and there were no significant differences among the three CsA doses (data not shown). We also examined the effect of the different CsA doses on the four adult MHC isoforms at the pre-mRNA and mRNA levels, as well as the percent MHC mRNA distribution (data not shown). There were no significant differences in these MHC transcript levels among the three CsA doses. Therefore, the 30 mg·kg⁻¹·day⁻¹ CsA was chosen for the remainder of the experiments to assess the role of Cn in MHC gene expression. Also, 50 mg·kg⁻¹·day⁻¹ CsA resulted in a loss of half of the rats before completion of the reloading phase of the study because of toxicity of this high dose of CsA. In addition to the normal loading state, MCIP-1 mRNA levels were strongly and significantly (P < 0.05) suppressed in response to the reloading stimulus at 7 and 14 days in 30 mg·kg⁻¹·day⁻¹ CsA-treated compared with vehicle-treated rats (data not shown).

Fig. 1.

RT-PCR analysis of myocyte-enriched calcineurin-interacting protein-1 (MCIP-1) mRNA levels in rats treated with cyclosporin A (CsA; 20, 30, and 50 mg·kg⁻¹·day⁻¹) and vehicle (open bar). Values are means ± SE. AU, arbitrary units. *Significantly different (P < 0.05) from vehicle.

RNA analysis.

The TriReagent protocol (Molecular Research Center) was used to extract total RNA from frozen soleus muscle. Extracted RNA was treated with DNase [1 U of RQ1 RNase-free DNase (Promega) per microgram of total RNA] and incubated at 37°C for 30 min, and a second RNA extraction was carried out using TriReagent LS (Molecular Research Center).

RT-PCR was used to assess pre-mRNA and mRNA of target genes. RT-PCRs were performed with the OneStep RT-PCR kit (Qiagen), where the RT and PCR are performed in a single reaction tube, with some modifications to the manufacturer's protocol, and as described previously (30). This protocol has been optimized to avoid amplification of nonspecific transcripts, which are known to be coamplified with pre-mRNA and mRNA transcripts and can, thus, preclude accurate measurement (19, 30). These one-step RT-PCR analyses were performed using 100–200 ng of total RNA and 15 pmol of specific primers in 25 μl of total volume and were carried out on a Robocycler (Stratagene). Compared samples were run under similar conditions (template amounts and PCR cycle numbers). RT reactions were performed at 50°C for 30 min followed by 15 min at 95°C and 16–32 PCR cycles. The annealing temperature was based on the PCR primer's optimal annealing temperature. PCR primers used for RNA analysis are shown in Table 1. The amount of RNA and the number of PCR cycles were adjusted so that the accumulated product was in the linear range of the exponential curve of the PCR amplifications. PCR products were separated by electrophoresis on agarose gels and stained with ethidium bromide. The ultraviolet light-induced fluorescence of stained DNA was captured by a digital camera, and band intensities were quantified by densitometry with ImageQuant software (GE Healthcare) on digitized images. RNA levels are reported as concentration, in arbitrary units per milligram of muscle weight.

Table 1.

PCR primer sequences, their specific target, and PCR product size

	RT-PCR Primers (5′→3′)	PCR Product, bp
Type I pre-mRNA
Forward	CCTGGTCCTATGTGCCGATCTCTAACGA	215
Reverse	CGGTCCCCAATGGCAGCAATAAC
Type I mRNA
Forward	GGAGCTCACCTACCAGACAGA	308
Reverse	CTCAGGGCTTCACAGGCATCC
Type IIa pre-mRNA
Forward	TGCTTCCCAATGCTGCCATATCTACAT	295
Reverse	TTCCTACTGCTTCCCTTGGTCTTGTCA
Type IIa mRNA
Forward	CCTCTTACTTCCCAGCTGCACCTTCT	239
Reverse	ACTTTCCCTGCGTCTTTGCTCTGAAT
aII NAT in intergenic region between IIa and IIx
Forward	ATCTTCACGGGTATTTTTGGTTT	294
Reverse	GCTGGGGCTCATTTTCTTC
Type IIx pre-mRNA
Forward	TGCCACAGAAAGAGGGACGC	290
Reverse	CTGGCTGTGGTGTGGCTGAAA
Type IIx mRNA
Forward	ACGGTCGAAGTTGCATCCCTAAAG	263
Reverse	CACCTTCGGTCTTGGCTGTCAC
xII NAT in intergenic region between IIx and IIb
Forward	CAGCTGTGCCTGGATCAAGTTAGTC	227
Reverse	TGCATCACGGAAGGAGATACAGACG
Type IIb pre-mRNA
Forward	GGCCATGCCAGCTAGCTTTTACG	270
Reverse	GCGTTTTGATTGGTGGAAGAGTCC
Type IIb mRNA
Forward	AGCCTGCCTCCTTCTTCATCTGG	229
Reverse	CACGGTTGCTTTCACATAGGACTC
bII NAT in intergenic region between IIb and Neo
Forward	CTAATGAGGAGGCCACTTTGAGAA	470
Reverse	GTATTATGTGGGCAGTCCGAGATG
Embryonic pre-mRNA
Forward	CAGCCAACACTATGAGTAGCGACACC	473
Reverse	GGCCCAGAACAAGGCAGTGATAAA
Embryonic mRNA
Forward	CAGCCAACACTATGAGTAGCGACACC	252
Reverse	TCATGGCATACACGTCCTCTGGTTTA
Embryonic NAT
Forward	TCCCGCTGAAATATACAAACAACT	212
Reverse	TACGTGGAAATTAAGCAGGATGGT
MCIP-1 mRNA
Forward	TCCCCCTGCCTCTCCACCTGT	194
Reverse	CCTCCTCTTCCTCCTCCTCCTCTTG

NAT, natural antisense transcript; MCIP-1, myocyte-enriched calcineurin-interacting protein-1.

Pre-mRNAs are the nascent, unprocessed, transcriptional products. Pre-mRNA transcript abundance serves as a better marker of a gene's level of transcriptional activity than the mRNA, because its half-life is much shorter. Assessment of the transcriptional activity of other genes by measurement of pre-mRNA with RT-PCR has been validated as an alternative to the nuclear run-on approach (14).

MHC mRNA isoform distribution.

The MHC mRNA isoform distribution was assessed by RT with oligo(dT)/random primers followed by PCR with primers targeting the embryonic, neonatal, and type I, IIa, IIx, and IIb MHC mRNAs, as described previously (11, 47). In these PCRs, each MHC mRNA signal was corrected to an externally added control DNA fragment that was coamplified with the MHC cDNAs using the same PCR primer pair. This approach provides a means to correct for any differences in the efficiency and/or pipetting of each PCR. A correction factor was used for each control fragment band on the ethidium bromide-stained gel to account for the staining intensity of the variably sized (224- to 324-bp) fragments, as reported previously (11).

SDS-PAGE MHC isoform separation.

A preweighed frozen muscle piece was homogenized in 20 volumes of PBS with use of a tight-fitting glass homogenizing tube and a pestle. Total protein concentration was determined using the Bio-Rad Protein Assay and γ-globulin as a standard. Total muscle homogenate proteins were diluted to 1 mg/ml in a solution consisting of 50% glycerol, 50 mM Na₄P₂O₇, 2.5 mM EGTA, and 1 mM β-mercaptoethanol (pH 8.8) and stored at −20°C. To quantify MHC protein isoform distribution, we subjected 2.5 μg of the stored total protein to SDS-PAGE (43). In addition to the four primary adult MHCs, this technique enabled us to separate neonatal and embryonic MHCs. These procedures were performed as described previously (1). These gels allow us to obtain the relative proportions of MHC isoforms and to quantify type I MHC by densitometry of the type I MHC band by Image Quant volume integration of the pixel density within a rectangle containing the band with local background correction (Image Quant Software, GE Healthcare). This procedure was performed to quantitate the pool of type I MHC protein content per muscle after HS + T₃ and at 7 and 14 days of reloading. Values per muscle were calculated on the basis of protein concentration and muscle weight.

Statistical analyses.

Values are means ± SE. Differences between muscle groups were analyzed by one-way ANOVA with Newman-Keuls post hoc test (GraphPad Software). For percent MHC protein and percent mRNA data sets, significant differences were determined by one-way ANOVA, with Newman-Keuls post hoc test within each individual MHC isoform only, across the six experimental groups. Differences between two groups were analyzed using an unpaired t-test. Statistical significance was set at P < 0.05.

RESULTS

Body weight and soleus muscle weight.

Soleus muscle wet weight per gram of body weight was significantly (P < 0.05) reduced by 26% in the HS + T₃ group compared with the Con group (Table 2). During reloading, soleus muscle weight per gram of body weight was significantly (P < 0.05) increased only in the 14d RL group compared with the HS + T₃ group. There was no effect of CsA on body weight or muscle weight per gram of body weight in either reloading group. However, statistical analysis by unpaired t-test showed a significant difference (P < 0.05) between reloading and reloading + CsA at 7 and 14 days. Total protein content per muscle was also significantly decreased from the Con group to the HS + T₃ group. Protein content was increased in the 7d RL group, but not in the 7d RL + CsA group relative to the HS + T₃ group (Table 2). Protein content was increased in the 14d RL and 14d RL + CsA groups relative to the HS + T₃ group (Table 2). There was no effect of CsA on total protein content. Total RNA concentration was increased in each reloading group compared with the Con or HS + T₃ group, but no effect of CsA treatment was observed (Table 2). CsA was effective in achieving significant inhibition of Cn on the basis of suppression of MCIP-1 mRNA expression (see methods; Fig. 1).

Table 2.

Effect of treatments on body weight, soleus wet weight, and protein, DNA, and RNA concentrations

	Experimental Group
	Con	HS + T3	7d RL	7d RL + CsA	14d RL	14d RL + CsA
Body wt, g	208±4.6	152±5.1*	195±3.4*†	191±2.7*†	215±3.6†‡§	212±3.6†‡§
Soleus wet wt/body wt, mg/g	0.43±0.01	0.32±0.02*	0.37±0.02*	0.32±0.02*	0.41±0.01†§	0.36±0.01*
Total protein concn, mg/g	206±4	220±7	183±6†	196±8	216±10‡	212±11
Total protein content per muscle, mg/muscle	18±0.8	11±0.8*	14±1*†	12±0.9*	19±1†‡§	16±0.9†§
DNA concn, mg/g	1.7±0.047	2.3±0.12*	2.0±0.059*†	2.1±0.078*†	1.8±0.048†	2.1±0.057*†
DNA content per muscle, μg/muscle	148±6	111±6*	159±9†	126±6*‡	157±9†§	158±9†§
RNA concn, μg/mg	1.2±0.075	1.2±0.070	1.9±0.063*†	1.8±0.021*†	1.4±0.066*†‡§	1.4±0.056*†§

Values are means ± SE. Con, control; HS + T₃, hindlimb suspension + triiodothyronine (T₃); 7d RL, HS + T₃ followed by 7 days of muscle reloading; 7d RL + CsA, 7d RL + cyclosporin A (CsA); 14d RL, HS + T₃ followed by 14 days of reloading; 14d RL + CsA, 14d RL + CsA; concn, concentration. Significant differences by 1-way ANOVA with post hoc test:

^*P < 0.05 vs. Con;

^†P < 0.05 vs. HS + T₃;

^‡P < 0.05 vs. 7d RL;

^§P < 0.05 vs. 7d RL + CsA.

Relative proportions of MHC mRNA.

A common internal DNA control fragment coamplified by PCR with each MHC gene transcript was used to examine the six experimental groups for MHC mRNA, which was expressed as a percentage of total MHC (Fig. 2). The soleus muscle expresses primarily type I and IIa MHC (80% and 15%, respectively) in the Con group (Fig. 2). In the HS + T₃ group, type I and IIa MHC were markedly decreased, with proportional increases in the fast type IIx and IIb MHCs (Fig. 2). Thus, in the HS + T₃ group, the slow soleus muscle was remodeled to a fiber type resembling that of a predominantly fast muscle at the pretranslational level. In the 7d RL and 7 d RL + CsA groups, type I MHC reemerges as the dominant MHC isoform, and no difference was observed between the 7d RL and 7d RL + CsA groups. Type IIa MHC also returned to control levels in the 7d RL group (14%), although type IIa MHC represents a higher percentage of total MHC (21%) in the 7d RL + CsA group than in the 7d RL group (Fig. 2). Type IIx MHC decreased in the 7d RL group compared with the HS + T₃ group, returning to control levels (<1%), but remained elevated in the 7d RL + CsA group (6%; Fig. 2).

Fig. 2.

RT-PCR analysis of myosin heavy chain (MHC) mRNA distribution. Each MHC mRNA is shown as percentage of total MHC mRNA pool for soleus muscle. Effect of each experimental treatment [control (Con), hindlimb suspension + triiodothyronine (HS + T₃), HS + T₃ followed by 7 days of muscle reloading (7d RL), 7d RL + CsA, HS + T₃ followed by 14 days of muscle reloading (14d RL), and 14 d RL + CsA] is shown for each MHC gene [type I, IIa, IIx, and IIb, embryonic (Emb), and neonatal (Neo)]. Each MHC cDNA is coamplified with and normalized to a common internal control fragment. Values are means ± SE. Significant differences of intra-MHC by 1-way ANOVA with post hoc test: *P < 0.05 vs. Con; ΦP < 0.05 vs. 7d RL; #P < 0.05 vs. 14d RL. For clarity, only significant differences between Con and HS + T₃ and between vehicle and respective CsA treatment are indicated.

Type I MHC mRNA relative expression returned to control levels in the 14d RL group (80%) but did not fully recover to control levels in the 14d RL + CsA group (66%). Similarly, type IIa MHC mRNA was reduced to near-control levels but remained elevated with CsA (28%). The percentage of type IIx MHC mRNA largely returned to control levels in the 14d RL and 14d RL + CsA groups but remained slightly elevated as a percentage in the 14d RL + CsA group. Little, if any, type IIb mRNA was detected in any of the reloading groups.

Embryonic MHC mRNA was strongly increased in the 7d RL group, such that it consisted of 34% of total MHC mRNA (Fig. 2). It decreased in the 14d RL group but remained higher than in the Con group (13.5% vs. 5%). The reloading-induced increase in percent embryonic MHC mRNA was attenuated by CsA at 7 and 14 days.

RNA transcripts of MHC genes.

In the percent MHC mRNA presentation, each isoform is expressed relative to the total MHC mRNA pool and, thus, provides information to understand how the MHC composition responds to treatment. However, it does not provide insight into the transcription states of each MHC at the absolute level. Thus we examined pre-mRNA and mRNA transcripts of the MHC genes in each experimental group (Fig. 3). The unspliced pre-mRNA products (Fig. 3, A–E) show congruency with the spliced mRNA products (Fig. 3, F–J) when expression levels are compared among the six experimental groups.

Fig. 3.

Pre-mRNA (A–E) and mRNA (F–J) analysis of MHC genes by RT-PCR. Effect of each experimental treatment is shown for type I, IIa, IIx, and IIb and embryonic MHC genes in soleus muscle. Transcript levels are shown as fold change compared with Con, except for type IIb MHC mRNA (I), because negligible mRNA levels were detected in the Con state. Values are means ± SE. Significant differences by 1-way ANOVA with post hoc test: *P < 0.05 vs. Con; §P < 0.05 vs. HS + T₃; ΦP < 0.05 vs. 7d RL; $P < 0.05 vs. 7d RL + CsA; #P < 0.05 vs. 14d RL.

Many of the broad alterations in MHC mRNA distribution described in Fig. 2 were similar at the absolute pre-mRNA and mRNA levels. HS + T₃ resulted in a near-complete repression of type I MHC gene transcription (Fig. 3, A and F). The reloading stimulus strongly induced transcription of type I MHC after 7 days in vehicle- and CsA-treated groups. Transcription of type I MHC was, in fact, higher in vehicle- and CsA-treated than in control rats (Fig. 3, A and F). However, CsA significantly blunted the accumulation of type I MHC pre-mRNA and mRNA compared with vehicle. By 14 days of reloading, type I MHC in vehicle- and CsA-treated rats was similar to control levels, although CsA treatment resulted in significantly decreased pre-mRNA and mRNA compared with the vehicle-treated group.

Type IIa MHC pre-mRNA and mRNA were reduced in the HS + T₃ group compared with the Con group and returned to control levels in the 7d RL group (Fig. 3, B and G). Type IIa MHC pre-mRNA was significantly increased in the 7d RL + CsA group compared with the 7d RL and Con groups. After 14 days of reloading, the CsA-treated group continued to maintain significantly higher levels of type IIa MHC pre-mRNA and mRNA than the vehicle-treated group. Type IIx MHC pre-mRNA and mRNA were strongly increased in the HS + T₃ group compared with the Con group and returned to near-control levels in the 7d RL and 14d RL groups (Fig. 3, C and H). However, type IIx MHC pre-mRNA was significantly greater in the 7d RL + CsA and 14d RL + CsA groups than in the 7d RL and 14d RL groups, although no difference was observed at the mRNA level between the 14d RL and 14d RL + CsA groups. Also, type IIa and IIx MHC pre-mRNA were significantly reduced in the 14d RL + CsA group compared with the 7d RL + CsA group, and there was no significant difference between the 14d RL + CsA and Con group. Type IIb MHC pre-mRNA and mRNA levels were almost undetectable in the control soleus (Fig. 3, D and I). HS + T₃ was highly effective in the induction of the type IIb MHC gene. Reloading resulted in a return to the repressed state for type IIb MHC, and no effects of CsA were observed.

Embryonic MHC pre-mRNA and mRNA levels were significantly reduced in the HS + T₃ group compared with the Con group (P < 0.05, t-test; Fig. 3, E and J). Embryonic MHC pre-mRNA and mRNA levels were strongly increased in the 7d RL group (Fig. 3, E and J). Upregulation of embryonic MHC pre-mRNA and mRNA was significantly inhibited in the 7d RL + CsA group compared with the 7d RL group. Embryonic MHC pre-mRNA and mRNA levels were further reduced in the 14d RL group compared with the 7d RL group, and there was also a significant decrease in the 14d RL + CsA group compared with the 14d RL group (P < 0.05, t-test). It is curious that the pattern of embryonic MHC gene transcription across the six experimental groups closely parallels that of type I MHC (cf. Fig. 3, A and E) but corresponds inversely with the pattern of transcription of type IIx MHC (cf. Fig. 3, C and E), indicating coordination that is likely to be actively regulated.

Regulated antisense RNA transcripts.

We previously reported the discovery and regulation of natural antisense transcripts (NATs), which are transcribed in the antisense orientation to the type II MHC genes in skeletal muscle and correlate with inhibited transcription of the sense gene (30, 35). The NAT transcribed in the antisense orientation to the type IIa MHC gene (aII NAT) was highly upregulated in response to HS + T₃ compared with control and returned to low levels of expression (similar to control) in all muscle reloading groups (Fig. 4A). Similarly, an NAT to the type IIx MHC gene (xII NAT) was highly upregulated in response to HS + T₃ compared with control and returned to low levels of expression (similar to control) in all reloading groups (Fig. 4B). We also report the regulation of a newly discovered NAT: an antisense transcript to the embryonic MHC gene is highly upregulated with HS + T₃ compared with control or reloading (Fig. 4C). Embryonic and type IIa and IIx MHC genes are tandemly linked on the same chromosome (40). Thus, with HS + T₃ compared with control, antisense RNA is transcribed at high levels across this MHC gene cluster. Further research is needed to determine whether antisense RNA could therefore provide a link to coordinate MHC expression across this ∼100-kb locus. An antisense transcript to the type IIb MHC gene could not be detected in any of the experimental groups.

Fig. 4.

RT-PCR analysis of antisense RNA products of type IIa and IIx MHC genes. A: antisense RNA to the type IIa MHC gene (aII NAT). B: antisense RNA to type IIx MHC gene (xII NAT). C: antisense RNA to embryonic MHC (Emb NAT). Values are means ± SE. Antisense product to type IIb MHC gene (bII NAT) was not detected in any of the experimental groups. *Significantly different (P < 0.05) from all other groups.

MHC isoform protein distribution.

The turnover rate is higher and the half-life is longer for MHC proteins than for pre-mRNA and mRNA products. Nevertheless, protein isoform content was quantified by SDS-PAGE to determine the relative proportions of type I, IIa, IIx, and IIb, embryonic, and neonatal MHC protein in the six experimental groups. As a percentage of total MHC protein, type I was significantly decreased in the HS + T₃ group compared with the Con group (Fig. 5). There was a further reduction with 7 days of reloading, with no difference between the 7d RL group and the 7d RL + CsA group (Fig. 5). Percent type I MHC protein returned to control levels in the 14d RL group but remained lower in the 14d RL + CsA group. Type IIa MHC relative protein was unaltered in the HS + T₃ and 7d RL groups, with no significant difference between the 7d RL group and the 7d RL + CsA group (Fig. 5). The relative proportion of type IIa MHC protein was significantly greater in the 14d RL + CsA group than in the 14d RL group. Type IIx MHC relative protein was significantly increased in the HS + T₃ group compared with the Con group (Fig. 5). The proportion of type IIx MHC protein was also significantly greater in the 7d RL + CsA and 14d RL + CsA groups than in the 7d RL and 14d RL groups, respectively (Fig. 5). Type IIb MHC relative protein was significantly increased in the HS + T₃ group compared with the Con group and progressively lost over 7 and 14 days of muscle reloading (Fig. 5). There was no effect of CsA on the percent expression of type IIb MHC protein. The proportion of embryonic MHC was significantly increased in the 7d RL and 7d RL + CsA groups compared with the Con or the HS + T₃ group (P < 0.05; not shown on Fig. 5). There was a significant effect of CsA relative to vehicle treatment at 7 days of reloading, but not at 14 days of reloading, for embryonic MHC.

Fig. 5.

SDS-PAGE analysis of MHC protein distribution. MHC isoforms of soleus muscle were separated by SDS-PAGE for each of the experimental groups. A: representative gel showing samples from each group, with each of the 7 MHC isoforms. Mix, MHC isoforms in protein extract from a mix of adult and 5-day-old neonatal skeletal muscles, which contained all 7 MHC isoforms. B: proportion of each MHC isoform as a percentage of total MHC protein pool for soleus muscle. Values are means ± SE. Significant differences were analyzed by 1-way ANOVA with post hoc test for each MHC isoform separately: *P < 0.05 vs. Con; ΦP < 0.05 vs. 7d RL; #P < 0.05 vs. 14d RL. For clarity, only significant differences between Con and HS + T₃ and between vehicle and respective CsA treatment are indicated.

CsA treatment during normal loading.

To contrast the effects of CsA treatment on MHC transcription in muscle reloading from an atrophied state with those of a normal loading condition, we treated a separate group of rats with vehicle or CsA for 7 days under normal ambulatory conditions. Type I MHC transcription was significantly reduced in CsA- compared with vehicle-treated rats (Fig. 6). There was a small, but nonstatistically significant, increase with CsA treatment in type IIa MHC pre-mRNA but a more robust, and statistically significant, accumulation of type IIa MHC mRNA levels with CsA compared with vehicle treatment (Fig. 6). There was a small, but significant, increase in type IIx MHC pre-mRNA and a robust effect of CsA on type IIx MHC mRNA expression (Fig. 6). Type IIb MHC transcriptional products could not be detected in the soleus muscle in either treatment group.

Fig. 6.

RT-PCR analysis of effect of CsA on MHC RNA expression during normal loading. CsA (30 mg·kg⁻¹·day⁻¹) was administered, and rat soleus muscle was analyzed for type I, IIa, and IIx MHC pre-mRNA (A) and mRNA (B). Transcript levels are depicted as fold change, with vehicle-treated group normalized to 1. Type IIb MHC transcripts could not be detected in vehicle- or CsA-treated groups and are not shown. Values are means ± SE. *Significantly different (P < 0.05) from respective vehicle-treated group.

DISCUSSION

Using a combination of mechanical unloading and thyroid hormone treatment, we essentially remodeled the slow fiber-type soleus muscle, which expresses primarily the type I MHC in the control state, transcriptionally to a fast fiber-type muscle. This is demonstrated by the expression of MHC genes, which showed very little expression of type I and IIa MHC pre-mRNA and mRNA, whereas 90% of the MHC mRNA pool consisted of the fast type IIx and IIb MHCs in the HS + T₃ condition. On resumption of normal ambulation under a euthyroid state, the now-fast fiber-type profile rapidly returned to an MHC profile largely resembling that of the control state by 14 days after reloading was initiated. By comparing CsA- with vehicle-treated rats during the reloading phase, we further demonstrated that reexpression of type I MHC was attenuated somewhat with CsA treatment, although transcriptional upregulation remained robust at 7 and 14 days in the CsA-treated group (Fig. 3). Also, reexpression of type IIa MHC during reloading was significantly enhanced in the 14d RL + CsA group compared with the 14d RL group, as demonstrated by a higher proportion of type IIa MHC mRNA and protein. Rerepression of type IIx MHC transcription after reloading was delayed with CsA after 7 days, but by 14 days type IIx MHC mRNA had returned to control levels in CsA- and vehicle-treated rats. Transcriptional activity, however, may have remained elevated in the 14d RL + CsA group, inasmuch as the level of type IIx MHC pre-mRNA was significantly higher than in the 14d RL group. HS + T₃ resulted in strong upregulation of type IIb MHC transcripts; however, reloading repressed type IIb MHC to control levels in CsA- and vehicle-treated rats at 7 and 14 days.

These findings, as summarized above, provide evidence that Cn inhibition seems to impede, to some degree, the return to the slow MHC phenotype with respect to the upregulation of type I MHC and downregulation of type IIx MHC during reloading. This is in contrast to type IIa MHC, which appears to be more responsive to reloading in the presence of CsA. Increased expression of type IIa MHC by CsA-mediated Cn inhibition was also reported previously in normal rat soleus muscle, when expressed as a percentage of total MHC protein (6, 21, 36, 51) or mRNA (21, 22) levels. Our findings in normally loaded soleus muscle (Fig. 6) showed a small, but statistically insignificant, increase with CsA treatment in type IIa MHC pre-mRNA but a more robust, and statistically significant, accumulation of mRNA with CsA than vehicle treatment. This indicates that inhibition of Cn results in increased transcription of type IIa MHC during reloading and normal loading.

The approach utilized in this study to repress type I MHC transcription with hyperthyroidism and unloading, followed by reloading in the context of normal thyroid levels, resulted in above-normal type I MHC transcriptional activation (i.e. pre-mRNA production). Thus reversal of type I MHC repression with mechanical loading plus resumption of normal thyroid hormone action presumably activates the signaling pathways and recruits the transcriptional complexes that are necessary for full activation of the type I MHC gene. Under these conditions, inhibition of Cn with CsA attenuated, but did not prevent, the “supranormal” activation of type I MHC (Fig. 3A).

We also report a significant shift from type I to type IIx MHC in the normally loaded soleus muscle with CsA treatment, indicating a significant role for Cn signaling in the maintenance of fiber-type phenotype. Thus, during remodeling and maintenance of fiber type, Cn signaling appears to be a regulator of the slow muscle phenotype. However, the large increase in transcription observed in the type I MHC gene during reloading with CsA treatment provides new insight into the contribution of the mechanical loading stimulus that may not be appreciated if one were to only use the normal loading state to assess the role of Cn in type I MHC transcriptional regulation. Accumulation of transcriptional products of the type I MHC gene, although attenuated, is still prevalent in the normally loaded CsA-treated soleus muscle. Thus the effect of CsA treatment during both loading paradigms clearly indicates that although Cn does play a role in type I MHC transcriptional modulation, other pathways and factors are likely the predominant regulators. For example, an alternative Ca²⁺-dependent signaling pathway, such as the Ca²⁺/calmodulin-dependent kinase pathway, may be involved (9). In addition, Ca²⁺-independent pathways, such as Ras/MAPK-ERK1/2, protein kinase D1, AMP-activated protein kinase-peroxisome proliferator-activated receptor (PPAR)-δ, and PPAR coactivator-1α, have been identified to play some role in expression of genes associated with the slow muscle fiber (20, 25, 28, 38). Therefore, of the multitude of pathways involved in muscle loading/activity of soleus muscle (e.g., neuromuscular activity-initiated, mechanical force-sensing, and metabolite-sensing), it appears that the signaling pathways initiated by Cn to regulate type I MHC are modulatory, rather than dominant, in slow fiber-type muscle.

Several of the published reports that have examined the role of Cn in MHC expression have reported the data as a percentage of total MHC protein, showing that Cn results in alteration of the percent type I MHC (6, 26, 27, 36, 44, 51). In many of these cases, it is impossible to determine the effect of Cn at an absolute level of MHC expression, inasmuch as alterations in the expression of type IIa, IIx, or IIb MHC may change the relative proportions of type I MHC, while possibly having little or no specific effect on the type I MHC gene itself. The data presented here, showing an effect of CsA on the percent type I MHC protein after 14 days of reloading, appear to be consistent with these studies (6, 26, 27, 36, 44, 51). However, the other analyses performed in our experiments provide further insights into the role of Cn, showing that the relative changes in type I MHC protein are reflective of the alteration of transcription of type I MHC by CsA treatment. Moreover, we confirmed these relative changes by examining type I MHC protein content from SDS-polyacrylamide gels (Fig. 7), which shows correspondence to the changes in pre-mRNA and mRNA. This analysis indicates that type I MHC protein is lost rapidly with 7 days of HS + T₃ and is regained with 14 days of reloading, with a significant attenuation effect in the CsA-treated group at 14 days of reloading. Also, Dunn et al. (12) suggested that the effect of Cn may be mediated primarily by posttranscriptional events during the early stages of MHC remodeling. However, our analysis of the pre-mRNA and mRNA transcriptional products, as well as relative protein content of MHC isoforms, indicates that effects of Cn are mediated primarily at the transcriptional level.

Fig. 7.

Type I MHC protein content. Type I MHC band of SDS-PAGE gels (as shown in Fig. 5A) was quantified by densitometry (as described in methods). Protein levels were normalized to Con levels. Values are means ± SE. Significant differences as determined by 1-way ANOVA with post hoc test: *P < 0.05 vs. Con; §P < 0.05 vs. HS + T₃; ΦP < 0.05 vs. 7d RL; $P < 0.05 vs. 7d RL + CsA; #P < 0.05 vs. 14d RL.

We previously reported on antisense transcripts to type IIa, IIx, and IIb MHCs in skeletal muscle and type I MHC in cardiac muscle that appear to play roles in coordinating shifts in the expression of the MHC genes in response to several different types of stimuli (17, 18, 30, 35). Embryonic, type IIa, IIx, and IIb, and neonatal MHC genes are tandemly linked on the same chromosome (40). As reported previously, the MHC NATs correlate with inhibited transcription of the sense gene product. NATs have also been shown mechanistically to inhibit their sense counterparts (34, 50). Regulated NATs to type IIa and IIx MHCs are also reported here. Transcription of the aII NAT was increased in response to HS + T₃, in strong correspondence to the decreased type IIa MHC pre-mRNA and mRNA. HS + T₃ also resulted in increased xII NATs, which are antisense to the type IIx MHC gene product. In this case, however, the xII NAT was not associated with decreased type IIx MHC transcription. This is likely because the xII NAT is coupled to the type IIb MHC promoter (30, 35), which was also highly activated by HS + T₃ (presumably more so than the type IIx MHC); thus, in this case, no net decrease was observed in type IIx MHC pre-mRNA and mRNA in the HS + T₃ group compared with the Con group. However, it is conceivable that the xII NAT acts to buffer type IIx MHC mRNA accumulation, resulting in the type IIb MHC being the dominant MHC isoform expressed (i.e., 61% of total MHC mRNA expressed is type IIb MHC and 29% is type IIx MHC). This also demonstrates that the effect of T₃ on the type IIb MHC gene was apparently functioning in synergy with the HS stimulus.

Limitations exist regarding the efficiency of Cn inhibition with the use of CsA. From our dose-response experiments with CsA at 20, 30, and 50 mg·kg⁻¹·day⁻¹, we determined that 30 mg·kg⁻¹·day⁻¹ CsA appeared to be optimal in terms of a robust decrease in MCIP-1 mRNA and viability of rat health. There was no difference between 30 and 50 mg·kg⁻¹·day⁻¹ in the mRNA expression of the Cn regulator MCIP-1. The higher dose also had toxicity effects: four of eight rats died during 7 days of reloading with administration of 50 mg·kg⁻¹·day⁻¹ CsA. Moreover, to our knowledge, the 30 mg·kg⁻¹·day⁻¹ dose used here is higher than CsA doses used in any previously published reports that examined the effect of CsA on MHC expression in rats. However, one cannot rule out the effect of residual Cn activity that may continue to activate Cn-NFAT signaling pathways. This is also an issue with transgenic models, in which Cn protein and Cn activity are not completely eliminated in published reports (29, 32, 33). Thus complete elimination of Cn in skeletal muscle has not been achieved with pharmacological or genetic approaches.

Despite this potential limitation, our data argue strongly for Cn-independent mechanisms to also control transcription of the MHC genes, and the slow type I MHC in particular. Reloading of an atrophied muscle provides a stimulus in which a strong loading-specific signaling pathway is prevalent, compared with a normal loading state, in which the contribution of neuromuscular activation and other factors is likely more in balance with loading-sensitive signaling. After 7 days of reloading, the type I MHC transcriptional system responds by more than doubling its production of the short half-life pre-mRNA compared with the control state (Fig. 3A). In type I MHC of CsA-treated rats, transcriptional responses exceeded those of the control soleus. Therefore, the contribution of mechanosensitive signaling pathways to type I MHC transcription would seem to be a major driving pathway. Transcription of type I MHC then reverts to levels more closely resembling the control state by 14 days of reloading, concurrent with a buildup of muscle mass and, thus, a decrease in the relative loading stimulus exerted on the soleus at this time. This parallel with decreasing transcription levels of type I MHC and the decrease in the relative loading stimulus, initially high (at 7d RL) and then significantly lower (at 14d RL), demonstrates the loading state sensitivity of type I MHC gene transcription. Importantly, this parallel is observed in the CsA-treated rats as well, demonstrating that the CsA treatment effect is altered in conjunction with alteration of the relative loading stimulus. Conversely, when mechanical loading was removed, type I MHC transcription was rapidly downregulated, further demonstrating its loading state sensitivity. The highly active type I MHC in the CsA-treated rat soleus during reloading, although attenuated, indicates that a major stimulus driving transcription of type I MHC is mediated by loading-induced factors that are not necessarily linked to the Cn pathway. These type I MHC transcriptional responses sensitive to loading state and CsA treatment are most likely linked to altered levels of type I MHC protein (Fig. 7).

Secondary effects of high doses of CsA have been cited as a reason to question the interpretation of CsA-mediated Cn inhibition effects, with the implication being that altered gene expression could be caused by the secondary effects, and not necessarily by muscle-specific Cn inhibition (e.g., altered mitochondrial respiration and neuronal activity) (10, 39, 41, 42). The reloading model used here, which, after 7 days of unloading, imposes considerable mechanical stress on the soleus muscle, thus, may not be subject to the same secondary effects that may be more prevalent during less taxing paradigms. However, our findings that type I MHC transcription is strongly increased during reloading, despite any obvious secondary effects of CsA, further support the notion that Cn-independent mechanisms primarily regulate the slow type I MHC gene expression.

In an effort to avoid the health consequences and secondary effects of CsA administration and to achieve potentially more complete Cn inhibition, researchers have utilized transgenic approaches to study the role of Cn in MHC gene expression. In these studies, Cn was shown to have a limited, if any, effect on type I MHC (13, 32, 33) or large effects (29, 32, 33), depending on the targeted gene and the particular muscle studied. Of course, there are limitations with transgenic approaches as well as with CsA-induced inhibition of Cn. Care should therefore also be taken in the interpretation of fiber-type shifts in transgenic mice because of possible variation in contributing signaling pathways during early developmental stages vs. the adult animal that results in the observed differential proportions of developmental and adult MHC isoform expression, which may be impacted by the transgenes of the altered mice. For example, type I MHC does not become fully expressed until ∼30 days postpartum in the rat soleus muscle (2). Moreover, perturbations impacting MHC expression, such as functional overload, have markedly different effects on newborn vs. adult rat skeletal muscle (37, 38).

In addition, species differences between rat and mouse may be critical in the study of slow muscle. The slow soleus muscle is a mixed-fiber type in the mouse, with only ∼50% type I MHC (44), in contrast to the rat (Fig. 2) and human (46), which have ∼80–90% type I MHC. Thus alterations attributed to Cn in the mouse soleus may appear to be of greater consequence because of lower baseline type I MHC expression levels than are typically observed in the rat. Therefore, the relatively smaller, but significant, effect of Cn on type I MHC we report in the rat may be assessed as a large effect in the mouse, given equal contributions of Ca²⁺-activated Cn signaling to type I MHC gene expression in both species. Differences in the response to stimuli known to impact type I MHC gene expression (denervation) have been reported between the rat and mouse, thus documenting such species differences (3).

In conclusion, the data reported here indicate that Cn is necessary for complete phenotype shifts in MHC genes during recovery of a slow (soleus) muscle from an atrophied state, although it is not sufficient in itself to induce such large phenotype shifts in MHC. Despite the knockdown of Cn activity with CsA at doses short of a toxic/lethal state, we were unable to diminish the dominant role of loading state in regulating gene expression of the type I MHC in rodent skeletal muscle. Therefore, Cn is likely one of several interconnecting signaling pathways and regulators that influence expression of the slow MHC gene in skeletal muscle.

Perspectives and Significance

The slow skeletal muscle provides vertebrates with a means to sustain posture and assist locomotion in a manner highly resistant to fatigue. The metabolic and contractile components of this fiber type are critical to the overall health of humans, with an impact on obesity and related metabolic disorders, as well as inactivity-related conditions. Thus it is necessary to identify and characterize the signaling pathways and transcriptional regulators that maintain and remodel slow muscle fibers to develop effective interventions to muscle-related disorders, whether pharmacological and/or exercise based. In addition to the role of Cn, as reported here, future research is needed to fully characterize the contribution of the other interconnected pathways that regulate expression of type I MHC in adult skeletal muscle.

GRANTS

This research was supported by National Institute of Arthritis and Musculoskeletal and Skin Diseases Grants AR-30346 (to K. M. Baldwin) and T32 AR-047752.