INTRODUCTION

Top Abstract Introduction References

PHOSPHOGLYCERATE MUTASE (PGAM) is a glycolytic enzyme that catalyzes the conversion of 2-phosphoglycerate to 3-phosphoglycerate. PGAM exists as a functional homodimer consisting of two M-type isoforms (PGAM-MM) or two B-type isoforms (PGAM-BB) or as a heterodimer (PGAM-MB). Distinct genes encode both isoforms and are expressed differentially among body tissues. The M-type gene (PGAM-M) is expressed in adult cardiac and skeletal muscle, whereas the B-type isoform (PGAM-B) is expressed in most other tissues, including fetal skeletal muscle (14). During myogenesis the PGAM-BB isoform is replaced by PGAM-MM (17). Previous studies have shown that, in adult skeletal muscle, PGAM-MM accounts for 80-90% of the total isoform pool and PGAM-BB for <2%, with the remainder made up of PGAM-MB (1, 10). The requirement for an active PGAM-MM enzyme in skeletal muscle is seen in families that harbor mutations within the PGAM-M coding region. Members of these families experience exercise intolerance, myoglobinuria, and muscle cramps (27).

PGAM enzymatic activity has previously been shown to be elevated in skeletal muscles associated with higher glycolytic activity. For example, the extensor digitorum longus and gastrocnemius muscles demonstrate a sixfold higher activity than the heart or the oxidative soleus muscle (1, 2, 10). In addition to the muscle-type distribution pattern of PGAM enzymatic activity, this enzyme has been found to be responsive to interventions that induce fiber-type changes within skeletal muscle. Hypothyroidism and denervation induce decreased expression of PGAM activity in fast-twitch muscles, whereas hyperthyroidism leads to elevated levels of PGAM activity (2, 11). However, no studies have examined whether PGAM-M expression can be modified with the fiber-type changes that occur with altered loading states on skeletal muscle.

Hindlimb suspension and functional overload of the plantaris muscle are commonly used to study the changes that occur in skeletal muscle in response to altered muscle activity. The changes that occur with hindlimb suspension and functional overload have been well characterized, including changes in muscle mass and the induction/repression of gene expression associated with changes in fiber type and energy metabolism (4). Although the mechanisms that drive altered gene expression in hindlimb suspension and functional overload are beginning to be addressed for contractile proteins such as -myosin heavy chain (MHC) (16, 29), MHC IIB (25), slow myosin light chain 1 (29), and slow troponin I (9), virtually nothing is known about the differential expression of genes involved in metabolism. Thus the primary goals of this study were 1) to test the hypothesis that PGAM-M expression is responsive to muscle loading, 2) to test the hypothesis that the major regulatory step governing the expression of the PGAM-M gene is pretranslational, and 3) by utilizing somatic gene transfer, to determine whether transcriptional activity of the PGAM-M promoter is responsive to muscle type and unweighting.

	MATERIALS AND METHODS

Hindlimb suspension. Adult female Sprague-Dawley rats weighing ~180 g were assigned to one of two groups: 1) normal control (n = 5) and 2) hindlimb suspension (n = 5). Hindlimb suspension was performed as previously described (25). Animals were suspended for 4 wk. The animals were killed, and the soleus and tibialis anterior (TA) muscles were removed, weighed, quickly frozen in liquid nitrogen, and stored at 80°C until analysis.

Protein extraction. Muscles were homogenized using a Polytron (Brinkman Instruments) in a solution (10% wt/vol) containing 100 mM Tris, pH 7.8, 1 mM EDTA, 5 µg/ml leupeptin, 0.7 µg/ml pepstatin, 170 µg/ml phenylmethylsulfonyl fluoride, 2 µg/ml aprotinin, and 1 mM dithiothreitol. Homogenates were centrifuged at 10,000 g for 15 min at 4°C. Supernatants were decanted into tubes, quick frozen in liquid nitrogen, and stored at 80°C until analysis.

PGAM assay. PGAM enzymatic activity was measured as previously described (2). Briefly, the oxidation of NADH was measured spectrophotometrically at 340 nm in a coupled reaction containing 8.5 mM triethanolamine, pH 7.4, 1 mM MgSO₄, 4.7 mM 3-phosphoglycerate, 120 µM 2,3-bisphosphoglycerate, 200 µM NADH, 0.57 mM ADP, 8.5 U/ml lactate dehydrogenase, 8.5 U/ml pyruvate kinase, and 3.4 U/ml enolase. The room-temperature reaction was initiated with the addition of 20 µl of soleus muscle extract. Enzymatic activity was expressed per milligram of protein, as determined with a Bio-Rad kit with BSA as a standard.

Western blot analysis. Western blotting of denatured protein was performed with enhanced chemiluminescence (New England Biolabs) according to the manufacturer's recommendations, with the following changes. After electrophoretic transfer and blocking of the membranes (1 h in 5% nonfat milk in Tris-buffered saline with 0.1% Tween-20, the membranes were incubated overnight with an affinity-purified antibody to PGAM-M (28). The antibody was diluted 1:100,000 in 5% BSA in Tris-buffered saline with 0.1% Tween-20.

PGAM-M mRNA analysis. To extract total RNA, muscles were homogenized in 4 M guanidine thiocyanate, 20 mM NaOAC, and 100 mM -mercaptoethanol. Homogenates were extracted in phenol-chloroform and precipitated with isopropanol. RNA pellets were incubated in 4 M LiCl for 30 min on ice and resuspended in formamide. Electrophoresis and transfer of RNA were performed as described previously (25, 26). Blots were probed with a random-primed ³²P-probe cDNA insert for the entire rat PGAM-M coding region (7). Blots were stripped and subsequently probed with a ³²P end-labeled oligo specific for 28S rRNA. Total RNA from a study that involved functional overload of the rat plantaris muscle was used as well (26).

Generation of a reporter construct for the rat PGAM-M promoter. PCR primers were designed to anneal to the sequenced region of the rat PGAM-M promoter (22). With use of high-fidelity PCR (Boehringer-Mannheim) of rat genomic DNA, the PCR product generated was cloned upstream of the firefly luciferase gene (pGL3 basic, Promega) and termed pGL3PGAM0.4. The cloned promoter region was checked for fidelity by sequencing of both strands. A single nucleotide difference from the initial reported sequence (22) was detected at position 360 (T  C) relative to the transcription start site. This difference does not lie within a putative regulatory site of the promoter, as determined by use of MatInspector (21).

DNA injections and determination of reporter gene activity. Plasmid DNA was purified using Qiagen columns, resuspended in sterile PBS, quantitated using absorbance at 260 nm, and stored at a concentration of 4 mg/ml. pGL3PGAM0.4 plasmid DNA was combined with an equal quantity of pRL-CMV plasmid DNA, a cytomegalovirus-driven Renilla luciferase reporter gene (Promega). Plasmid DNA was injected into the TA muscle (100 µg of each construct) and the soleus muscle (50 µg of each construct) of 12 rats weighing ~120 g, as previously described (25). Six animals were then hindlimb suspended. One week after the injections, all animals were killed, and the injected muscles were quickly removed, trimmed of all connective tissue, weighed, and frozen in liquid nitrogen. Muscles were homogenized and assayed for firefly and Renilla luciferase gene expression, as described previously (25).

Statistics. Values are means ± SE. Statistically significant differences were determined using standard t-tests after an ANOVA. The 0.05 level of confidence was accepted for statistical significance.

	RESULTS

PGAM-M expression among muscles of the rat hindlimb. PGAM enzymatic activity has been well described among different skeletal muscles of the hindlimb. PGAM enzymatic activity is higher in muscles with higher glycolytic capacities (1, 2). However, these characterizations did not address potential levels of regulation, i.e., PGAM-M protein or mRNA levels. Western blotting was performed to detect levels of PGAM-M in muscle extracts with use of a previously described antibody generated against PGAM-M (28). Only a single band of ~29 kDa was detected in a Western blot (Fig. 1). Protein and antibody concentrations were such (1-10 µg of total muscle protein and 1:100,000 dilution of the primary antibody) that a linear signal was observed with increasing amounts of the blotted protein (Fig. 1A). A Western blot was performed on total muscle extracts from four muscles of the hindlimb: TA, soleus, plantaris, and extensor digitorum longus. After scanning of blots similar to those in Fig. 1B, it was determined that immunoreactive PGAM-M protein content was approximately fourfold lower in the soleus muscle than in the three other glycolytic muscles. These data parallel the reported PGAM enzymatic activity among these four muscles (1, 2). A Northern blot was performed using total RNA isolated from these same muscles of the hindlimb. Only one ~0.8-kb band was detected in total RNA from skeletal muscle when the rat PGAM-M cDNA was used as a probe (Fig. 2). The level of PGAM-M mRNA was similarly fourfold lower in the oxidative soleus muscle than in the faster muscles examined (Fig. 2B).

View larger version (45K): [in this window] [in a new window]   View larger version (41K): [in this window] [in a new window]   Fig. 1.   Western blot analysis of muscle-specific isoform of phosphoglycerate mutase (PGAM-M) expression among different muscles of rat hindlimb. A: increasing amounts of total tibialis anterior (TA) muscle protein extract were separated using denaturing gel electrophoresis, then subjected to Western blotting with use of a PGAM-M-specific antibody. This antibody detected a 29-kDa single band. Parallel-loaded Coomassie blue-stained denaturing gel is shown below blot. B: Western analysis performed on 5 µg of total muscle protein extract from TA, soleus (Sol), plantaris (Plan), and extensor digitorum longus (EDL) muscle. Parallel-loaded Coomassie blue-stained denaturing gel of 10 µg of same samples is shown below blot. C: blots similar to those in B were scanned and quantitated (n = 5 for each muscle). * P < 0.05 vs. TA muscle.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Northern blot analysis of PGAM-M expression among different muscles of rat hindlimb. A: 10 µg of total RNA isolated from TA, soleus, plantaris, and EDL muscles were probed with a radioactively labeled PGAM-M cDNA. Northern blots were stripped and reprobed with a 28S oligo probe. B: blots similar to those in A were scanned and quantitated for TA (n = 4), soleus (n = 6), plantaris (n = 5), and EDL (n = 5) muscle. Data are expressed relative to 28S signal. * P < 0.05 vs. TA.

Hindlimb suspension. We next tested the hypothesis that the PGAM-M gene product would be upregulated with the metabolic/fiber-type changes that occur with muscle unweighting. Four weeks of hindlimb suspension induced a significant decrease in the muscle weight-to-body weight ratio of the rat soleus muscle (0.36 ± 0.02 and 0.20 ± 0.02 mg/g in normal and unweighted soleus, respectively, P < 0.05). Suspension did not have an effect on the relative weight of the TA muscle (1.77 ± 0.02 and 1.73 ± 0.03 mg/g in normal and unweighted TA, respectively). PGAM enzymatic activity was elevated 2.5-fold within the unweighted soleus muscle over the normal soleus muscle (Fig. 3). By Western blot analysis, the unweighted soleus muscle was determined to have a 1.8-fold higher level of immunoreactive PGAM-M than the normal soleus muscle (Fig. 3). The levels of PGAM-M mRNA in the muscles examined mirrored the changes observed in the protein levels (Fig. 4). Relative to 28S rRNA, PGAM-M mRNA increased significantly 3.5-fold in the unweighted soleus compared with the normal soleus muscle.

View larger version (32K): [in this window] [in a new window]   View larger version (29K): [in this window] [in a new window]   Fig. 3.   PGAM-M protein expression after muscle unloading. A: typical Western blot with use of 5 µg of total protein extract from normal soleus, unweighted soleus, and normal TA. Parallel-loaded Coomassie blue-stained gel is shown below blot. B: enzymatic activity for PGAM and Western blotting performed on set of total protein extracts in A (n = 5 for normal and hindlimb unweighting). PGAM enzymatic activity was determined spectrophotometrically by oxidation of NADH in a coupled reaction. Data are expressed as µmol NADH oxidized · mg protein1 · min1. Unweighted soleus muscle demonstrated a 2.5-fold higher level of PGAM activity than did normal soleus muscle. Blots similar to those in A were scanned and quantitated. Expression of immunoreactive PGAM-M was increased 1.8-fold in unweighted soleus muscle above that of normal soleus muscle. * P < 0.05 vs. soleus muscle.

View larger version (65K): [in this window] [in a new window]   View larger version (41K): [in this window] [in a new window]   Fig. 4.   Northern blot of PGAM-M expression among normal and unloaded soleus muscles. A: 10 µg of total RNA isolated from normal and unweighted soleus muscles probed for PGAM-M. Northern blots were stripped and reprobed with a 28S oligo probe. B: blot was scanned and quantitated (n = 5 for both groups). Data are shown relative to 28S signal. * P < 0.05 vs. normal soleus muscle.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Northern blot of PGAM-M expression among normal and functionally overloaded plantaris muscles. A: total RNA from a previous study of functional overload of plantaris muscle analyzed for PGAM-M mRNA content. Total RNA (10 µg) of normal (NP) or functionally overloaded plantaris muscle (OP) was loaded. Northern blots were stripped and reprobed with a 28S oligo probe. B: blot was scanned and quantitated (n = 5 for both groups). Data are shown relative to 28S signal. * P < 0.05 vs. normal plantaris muscle.

Somatic gene transfer. Inasmuch as PGAM-M expression appeared to be regulated pretranslationally among muscles of different glycolytic capacities as well as in the unweighted soleus muscle, somatic gene transfer was employed to determine whether the activities of promoter and upstream regulatory regions of the PGAM-M gene are different between muscle types and in muscle unweighting. A reporter gene construct was generated that contained the firefly luciferase coding region driven by the proximal 400 bp upstream of the transcriptional start site of rat PGAM-M gene (pGL3PGAM0.4). This promoter was cloned by PCR and sequenced in its entirety. A single difference (T  C) at position 360 from the reported sequence (22) was found. This nucleotide difference was not in a region that contains homology with other muscle-specific promoters or in any known putative regulatory element (21). This construct was coinjected with pRL-CMV, a constitutively active promoter driving Renilla luciferase coding region, for normalizing reporter gene expression. Plasmids were injected into the TA and soleus muscles of young rats. One-half of those rats (n = 6) were then suspended by their tails to unweight the soleus and TA muscles. One week after injection, firefly luciferase activity relative to Renilla luciferase activity was significantly higher (2-fold) in the normal TA than in the normal soleus muscle (Fig. 6). Furthermore, unweighting the soleus muscle induced a 2.5-fold increase in reporter gene activity above that of the normal soleus muscle. Unweighting the TA had no effect on promoter activity relative to the normal TA. As controls, promoters from previously identified genes that demonstrate a fiber-type-specific pattern of expression were also injected into normal TA and soleus muscles (25, 29). The rat -MHC promoter driving firefly luciferase (a kind gift from Dr. Ken Baldwin) was expressed at 125-fold higher levels in the soleus than in the TA muscle (0.0249 ± 0.0032 vs. 0.0002 ± 0.0001). The -MHC gene product accounts for 90% of the total MHC isoform pool in the soleus muscle and only ~5% of the MHC pool in the TA muscle. The MHC IIB gene, however, is not expressed in the normal soleus muscle, whereas its product accounts for 70% of the MHC isoform pool in the TA muscle. The mouse MHC IIB promoter was expressed at 30-fold higher levels in the TA than in the soleus muscle (0.0228 ± 0.0041 vs. 0.0007 ± 0.0002), which agreed with previous work in which the same promoter was used (25).

View larger version (37K): [in this window] [in a new window]   Fig. 6.   Reporter gene analysis of PGAM-M upstream sequence with use of somatic gene transfer. Approximately 400 bases of upstream sequences of rat PGAM-M gene were cloned upstream of a firefly luciferase reporter gene (pGL3PGAM0.4). This construct was combined with pRL-CMV, a strong constitutively active promoter that drives Renilla luciferase reporter gene, and injected into soleus or TA muscle. Animals were then randomized into normal or hindlimb suspension: normal soleus (n = 12), unweighted soleus (n = 8), normal TA (n = 9), and unweighted TA muscle (n = 12). Firefly luciferase activity relative to Renilla luciferase activity after 1 wk of unweighting was significantly higher in unweighted than in normal soleus muscle. TA reporter gene activity was unchanged with unweighting. Reporter gene activity in normal TA muscle was 2-fold higher than in normal soleus muscle. * P < 0.05 vs. normal soleus muscle.

	DISCUSSION

It is well established that the enzymatic activities of proteins involved in glycolysis or fatty acid oxidation are modulated by altered loading conditions. Previous studies on hindlimb unweighting have demonstrated an increase in glycolytic capacity of an oxidative muscle such as the soleus, with a concomitant decrease in oxidative capacity (12, 23, 24). However, very little is known about the regulation of these activities. In fact, only one gene associated with metabolism has been characterized at any level other than enzymatic activity in response to hindlimb unweighting. Cytochrome c mRNA levels decrease ~50% after 7 days of soleus unweighting (3). In the absence of such data for metabolic enzymes, we wished to characterize PGAM in normal and unweighted skeletal muscle. PGAM was examined, because all the tools with which to characterize this muscle-specific enzyme, i.e., an enzymatic assay (2), specific antibodies (28), the cloned cDNA (7), and the cloned promoter (22), are available.

The hypothesis that expression of PGAM-M is regulated pretranslationally among muscles of different fiber type was supported with data reported here. Expression patterns of immunoreactive PGAM-M and PGAM-M mRNA parallel each other (Figs. 1 and 2) as well as the reported enzymatic activity among the muscles studied (1, 2). PGAM-M expression was also regulated pretranslationally within the unweighted soleus muscle (Figs. 3 and 4). The increases in PGAM-M expression with 4 wk of soleus unweighting reflect well the differential expression of other enzymes involved in glycolysis, i.e., hexokinase (23). Although PGAM can be posttranslationally modified by phosphorylation in a phosphocreatine-dependent manner (20), the data presented here suggest that posttranslational control does not play a significant role in determining the relative PGAM activity in normal and unweighted soleus muscles. PGAM-M mRNA was also upregulated 20% after only 2 days of hindlimb unweighting and 50% after 7 days (data not shown). This suggests that PGAM-M mRNA is not only robust in its response to unweighting (3.5-fold) but also responds quickly to the unweighting stimulus. This response is similar to results obtained with another model of disuse, immobilization. Others have found that carbonic anhydrase III mRNA is downregulated in the soleus muscle after 5 days of immobilization (6), whereas cytochrome c mRNA is decreased 70% after 7 days (5). Thus monitoring the pattern of PGAM gene expression among fiber types and muscle unweighting is advantageous, because 1) there was fourfold differential expression of PGAM-M protein and mRNA between muscle types, 2) it appears that PGAM-M expression is indicative of the general changes, both magnitude and temporal expression patterns, in metabolic enzyme expression that are induced with unweighting, and 3) the tools to analyze the regulation of gene expression are available.

We further found that PGAM-M mRNA was also responsive to muscle overloading. Mirroring the fast-to-slow transition of MHC isoforms observed with this model of functional overload (26), PGAM-M mRNA was decreased significantly in the overloaded plantaris muscle (Fig. 5), suggesting that the overloaded plantaris became less glycolytic. Hence, PGAM-M mRNA was found to be differentially expressed in different muscles of the hindlimb during unweighting and during functional overload.

To understand the molecular mechanisms that govern differential expression of genes within fast- and slow-twitch muscle or in response to altered loading conditions, it is important to establish the level at which the pretranslational control is manifest: mRNA synthesis or degradation. The proximal 400 bp of the PGAM-M promoter, which are sufficient to drive high-level expression of a reporter gene in cell culture (22), were sufficient to drive a two- to threefold higher level of reporter gene expression within the unweighted soleus and the TA muscle over the normal soleus muscle (Fig. 6), reflecting the differential mRNA and protein expression of this gene. We also found that reporter gene expression driven by the PGAM-M promoter was not altered between the normal and unweighted TA muscle, an observation that mirrors endogenous mRNA levels in normal and unweighted TA muscles (data not shown). These data suggest that transcriptional processes may be responsible for the observed pattern of muscle-type distribution of this gene.

The cis element(s) within this 400-bp region responsible for 1) elevated expression in the TA muscle and 2) responsiveness to soleus unweighting may not be the same. For example, the 4.2-kb upstream regulatory region of the troponin I slow promoter is sufficient for slow-twitch fiber-type-specific expression of a reporter gene (8, 15) but not for a decrease in expression after unweighting (9). Although the -MHC promoter and the MHC IIB promoter are differentially expressed among muscle types (see RESULTS; Refs. 25 and 29) and are responsive to unweighting (16, 25) and functional overload (29), the element(s) within these promoters for fiber-type-specific expression and load responsiveness has yet to be demonstrated to be identical or mutually exclusive.

Numerous putative binding sites exist within the proximal 400 bp of the PGAM-M promoter. These include multiple E-boxes, an M-CAT site, and an MEF-2 site conserved with the human PGAM-M promoter (18, 22). The MEF-2 site is critical for this promoter in cell culture (18). The muscle-specific -enolase promoter also contains a critical MEF-2 site (13), which may be of particular importance for coregulation of PGAM and -enolase, inasmuch as these two enzymes interact in vitro (19). However, elements important for activity in cell culture may not be important for expression of reporter genes in vivo. For example, elements within the -MHC promoter critical for activity in cell culture are not important for unweighting-induced downregulation of this promoter in vivo (16).

In conclusion, PGAM-M expression appears to be regulated pretranslationally among muscles of the hindlimb and in response to altered muscle loading conditions. Furthermore, somatic gene transfer suggests that transcriptional processes may play some role in dictating muscle-type-specific and unweighting responsiveness of expression of the PGAM gene.