Contractile Activity Modifies Fru-2,6-P2 Metabolism in Rabbit Fast Twitch Skeletal Muscle*

Modification of muscular contractile patterns by denervation and chronic low frequency stimulation induces structural, physiological, and biochemical alterations in fast twitch skeletal muscles. Fructose 2,6-bisphosphate is a potent activator of 6-phosphofructo-1-kinase, a key regulatory enzyme of glycolysis in animal tissues. The concentration of Fru-2,6-P2 depends on the activity of the bifunctional enzyme, 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase (PFK-2/FBPase-2), which catalyzes the synthesis and degradation of this metabolite. This enzyme has several isoforms, the relative abundance of which depends on the tissue metabolic properties. Skeletal muscle expresses two of these isoforms; it mainly contains the muscle isozyme (M-type) and a small amount of the liver isozyme (L-type), whose expression is under hormonal control. Moreover, contractile activity regulates expression of muscular proteins related with glucose metabolism. Fast twitch rabbit skeletal muscle denervation or chronic low frequency stimulation can provide information about the regulation of this enzyme. Our results show an increase in Fru-2,6-P2 concentration after 2 days of denervation or stimulation. In denervated muscle, this increase is mediated by a rise in liver PFK-2/FBPase-2 isozyme, while in stimulated muscle it is mediated by a rise in muscle PFK-2/FBPase-2 isozyme. In conclusion, our results show that contractile activity could alter the expression of PFK-2/FBPase-2.

Modification of muscular contractile patterns by denervation and chronic low frequency stimulation induces structural, physiological, and biochemical alterations in fast twitch skeletal muscles. Fructose 2,6-bisphosphate is a potent activator of 6-phosphofructo-1-kinase, a key regulatory enzyme of glycolysis in animal tissues. The concentration of Fru-2,6-P 2 depends on the activity of the bifunctional enzyme, 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase (PFK-2/FBPase-2), which catalyzes the synthesis and degradation of this metabolite. This enzyme has several isoforms, the relative abundance of which depends on the tissue metabolic properties. Skeletal muscle expresses two of these isoforms; it mainly contains the muscle isozyme (M-type) and a small amount of the liver isozyme (L-type), whose expression is under hormonal control. Moreover, contractile activity regulates expression of muscular proteins related with glucose metabolism. Fast twitch rabbit skeletal muscle denervation or chronic low frequency stimulation can provide information about the regulation of this enzyme. Our results show an increase in Fru-2,6-P 2 concentration after 2 days of denervation or stimulation. In denervated muscle, this increase is mediated by a rise in liver PFK-2/FBPase-2 isozyme, while in stimulated muscle it is mediated by a rise in muscle PFK-2/FBPase-2 isozyme. In conclusion, our results show that contractile activity could alter the expression of PFK-2/FBPase-2.
Skeletal muscle is a heterogeneous tissue, which responds differentially to a large variety of stimuli (1,2). Modifications of basal contractile activity, induced by denervation or chronic low frequency stimulation, promote an adaptive response. These adaptations are reflected in muscle phenotype; i.e. there are variations in muscle type fiber composition, characterized by different biochemical and physiological properties.
One consequence of denervation or chronic low frequency stimulation in rabbit fast twitch skeletal muscle is the phenotypic transition from fast twitch to slow twitch fiber type (3)(4)(5)(6). These changes involve variations in enzyme activities and, in some case, isozyme expression. Fast twitch fibers are characterized by the predominance of anaerobic glycolysis, while in slow twitch fibers aerobic glycolysis is the main pathway by which energy demands are supplied during contraction. Thus, glucose metabolism is involved in this process.
The rate-limiting enzyme of glycolysis is 6-phosphofructo-1kinase (PFK-1), 1 which is regulated by a large number of modulators. The bisphosphorylated sugar fructose 2,6-bisphosphate (Fru-2,6-P 2 ) has been identified as a potent positive effector of this enzyme in several mammalian tissues (7). Moreover, modifications of Fru-2,6-P 2 by factors such as hormonal or contraction pattern have been related with variations of the glycolytic rate (8 -11).
The intramuscular concentration of Fru-2,6-P 2 is controlled by the bifunctional enzyme 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase (PFK-2/FBPase-2), which is responsible for the synthesis and degradation of this metabolite. This enzyme has several isoforms, the relative abundance of which depends on the metabolic properties of the tissue. These isozymes differ in size, kinetics, and immunologic properties as well as in their response to phosphorylation by protein kinases. Skeletal muscle expresses two of these isoforms. It mainly contains muscle isozyme (M-type), and a small amount of the liver isozyme (L-type) (12)(13)(14)(15)(16). Both isoforms are encoded by the same gene, as a result of alternative splicing, whose expression is under hormonal control (17).
Moreover, contractile activity has a regulatory effect on some proteins related with glucose metabolism in fast twitch skeletal muscle. The absence of contractions induces an increase in GLUT-1 (18) and HK-II (19,20), while an increase in contractile activity induces a rise in  and HK-II (22). These studies highlight the importance of muscle contraction in the regulation of protein expression related with glucose metabolism.
Here we examine the influence of contractile activity on Fru-2,6-P 2 metabolism and its regulation through PFK-2/FB-Pase-2. We measured the changes in this enzyme induced by denervation or by chronic low frequency stimulation at 10 Hz in fast twitch rabbit skeletal muscle.

EXPERIMENTAL PROCEDURES
Chemicals-[␥-32 P]ATP (5000 Ci/mmol), [␣-32 P]dCTP (3000 Ci/ mmol), and Hybond N filters were purchased from Amersham Pharmacia Biotech. The random priming DNA labeling kit was from Roche Molecular Biochemicals. Enzymes and biochemical reagents were from either Roche Molecular Biochemicals or Sigma. All other reagents were of analytical grade.
Treatment of Animals-Adult White New Zealand rabbits weighing between 3.0 and 3.5 kg were fed ad libitum and housed in animal quarters with a 12-h light, 12-h dark cycle. For denervation studies, animals were anesthetized by intramuscular injection of 20 mg/kg ketamine (Ketolar, Parke-Davis) and a 2-ml solution of 0.5% azepromazine (Calmoneosan, Smithkline Beecham), the peroneal nerve was unilaterally severed, and 10 -15 mm of the distal stump was removed. Finally, the proximal part of the nerve was ligated in the Biceps femoris muscle to avoid reinnervation. The contralateral control leg was sham-operated.
For stimulation studies, under the same anesthesia protocol as in denervated animals, 1 week before the onset of stimulation period, electrodes were implanted laterally to the peroneal nerve as described elsewhere (23). This allowed the continuous stimulation at 10-Hz trains (trains of pulses of 18 mA for 0.15 ms) for 24 h/day.
After various periods of denervation or stimulation (1, 2, 4, 7, or 14 days), right (control) and left (denervated or stimulated) tibialis anterior (TA) muscles were carefully exposed. Both muscles were removed simultaneously by two researchers and quickly frozen with aluminum clamps precooled in liquid N 2 . The frozen samples were stored at Ϫ80°C until analysis.
Assay of Metabolites-Samples of frozen muscle were powdered in a stainless steel percussion mortar precooled in liquid N 2 and used for the measurement of metabolites.
Glycogen was extracted from about 10 mg of muscle by alkaline treatment. The hydrolysis and measurement of glucose produced were carried out using the anthrone method (24). ATP, glucose 6-phosphate (Glc-6-P) and fructose 6-phosphate (Fru-6-P) were extracted by acid treatment from about 20 mg in ice-cold 0.5 M perchloric acid. After neutralization and centrifugation, aliquots of the extracts were used. Measurements were carried out using enzymatic methods with fluorimetric techniques (25).
Treatment of PFK-2 with the Catalytic Subunit of Cyclic AMP-dependent Protein Kinase-Partial purified PFK-2 from control, denervated, and stimulated TA muscle were incubated at 30°C for 20 min with the catalytic subunit of cyclic AMP-dependent protein kinase (1 milliunit/ml) in a final volume of 0.1 ml containing 100 mM Hepes (pH 7.1), 1 mM dithiothreitol, 0.1% bovine serum albumin. The reaction was started by the addition of 1 mM ATP-Mg 2ϩ and stopped by the addition of 5 mM EDTA. Samples were removed and assayed for PFK-2 activity (28,29).
For immunodetection, two antibodies were used: polyclonal antiserum MCL-2 (30) against L-type and M-type PFK-2/FBPase-2 isozymes and polyclonal antiserum CL against L-type PFK-2/FBPase-2 isozyme (31). Both antibodies were a generous gift from Dr. Hue. MCL-2 and CL IgG were produced in rabbits. Primary antibody was preincubated for 1 h with a secondary antibody (peroxidase-conjugated anti-rabbit). To prevent an excess of secondary antibody, a third IgG, against a maize protein that is not present in mammalian tissue, was added and incubated for 1 h. Later, this mixture was used for incubating the membrane (2 h) and for detecting PFK-2/FBPase-2 by enhanced chemiluminescense, (ECL, Amersham Pharmacia Biotech). A control test of rabbit blood ruled out any possible disturbance attributable to saturation of the secondary with the tertiary antibody.
RNA Isolation and Northern Blot Analysis-Total RNA from muscle was extracted using the acid guanidinium isothiocyanate/phenol/chloroform method as described by Chomczynski and Sacchi (32). All samples had a 260/280 absorbance ratio above 1.7.
After quantification, total RNA (15-30 g) was denatured at 65°C in the presence of formamide, formaldehyde, and ethidium bromide (33) to allow the visualization of RNA. RNA was separated on a 1.2% agarose/ formaldehyde gel and blotted on Hybond N filters. The RNA in gels and filters was visualized with ethidium bromide and photographed by UV transillumination to ensure the integrity of RNA, to check the loading of equivalent amounts of total RNA, and to confirm proper transfer. RNA was transferred in 10ϫ standard saline citrate (SSC; 0.15 M NaCl and 0.015 M sodium citrate, pH 7.4). After transfer, RNA was fixed by irradiating the membrane with UV light for 5 min.
Filters were washed twice in 2ϫ SSC at room temperature for 5 min and twice in 0.1ϫ SSC plus 0.1% SDS at 50°C for 30 min.
The abundance of specific PFK-2/FBPase-2 mRNA was quantified by scanning densitometry of autoradiograms as described above, and data were expressed as a percentage of control values.
Other Methods-Protein was determined with bovine serum albumin as a standard (34). Statistical significance of differences was assessed by nonparametric Kruskal-Wallis test. Results were considered significant when p was Ͻ0.05.
In denervated muscle, a significant increase occurred 2 days after nerve section, and it persisted throughout the 14 days of the experiment. At the end of this period Fru-2,6-P 2 concentration was approximately 3-4-fold higher than the control value.
Chronic low frequency stimulation also increased Fru-2,6-P 2 . By day 1, there was a significant increase, which persisted at day 4. The peak at day 4 was approximately 4 -5-fold higher than control. As stimulation continued, the concentration declined to control values at day 30 (results not shown). Fru-2,6-P 2 values in control remained constant throughout the experiment.
Because levels of Fru-2,6-P 2 in skeletal muscle are controlled by substrate availability (12), its precursors (Glc-6-P, Fru-6-P, ATP) were studied. Denervation or stimulation had different effects on the intracellular concentration of these metabolites (Table I).
In denervated muscle, Glc-6-P and Fru-6-P decreased by more than 50% at day 4, and their concentration remained significantly lower than that of control muscle. ATP showed a significant decrease at day 4, which persisted for 14 days.
In stimulated muscle, although Glc-6-P was already lower on the first day, Fru-6-P followed the same pattern as in denervated muscle. However, in stimulated muscle these metabolites returned to control values at day 14. ATP was constant throughout this period.
Effects of Denervation or Stimulation on PFK-2 and FBPase-2 Activities-In order to explain the increases observed in Fru-2,6-P 2 after denervation or stimulation of muscle, we measured the enzyme activity for the synthesis and degradation of this bisphosphorylated metabolite, using partial purified (6 -21% PEG) TA muscle homogenate.
Denervation and stimulation increased PFK-2 activity ( Fig.  2A). In denervated muscle, kinase activity increased significantly at day 1, reaching the highest value at day 2. PFK-2 activity then decreased rapidly to control values. The pattern of change in stimulated muscle was similar to the changes in denervated muscle. Kinase activity increased significantly at day 1, reached a maximum at day 4, and declined thereafter to control values. The response of FBPase-2 activity depended on the muscle treatment (Fig. 2B). In denervated muscle, there was a significant decrease at day 1, which was maintained for the rest of the study. In contrast, in stimulated muscle FPBase-2 increased at day 1, reached a maximum at day 4, and recovered basal level at day 7.
The effect on the kinase/phosphatase activity ratio also depended on muscle treatment (Fig. 2C). After denervation, the ratio increased significantly at day 1. By day 2, the ratio (2.36 Ϯ 0.30) was approximately 10-fold higher than control, and it declined thereafter to control values. In contrast, stimulation did not modify the ratio (0.17-0.34 Ϯ 0.05).

Effects of sn-Glycerol 3-Phosphate and Treatment of Catalytic Subunit of Cyclic AMP-dependent Protein Kinase on PFK-2
Activity-Because the PFK-2/FBPase-2 ratio in denervated muscle was similar to that previously described for liver (35), we proposed a kinetic study. As described previously (12,35), inhibition by sn-glycerol 3-phosphate and cyclic AMP-dependent protein kinase phosphorylation can be used to differentiate M-type isozyme from L-type. To assess whether the L-type isozyme was present in denervated muscle, we measured the kinase activity in the presence of sn-glycerol 3-phosphate. Moreover, we carried out a treatment with the catalytic subunit of cyclic AMP-dependent protein kinase.
PFK-2 activity from denervated muscle was inhibited by a low concentration of sn-glycerol 3-phosphate. In contrast, PFK-2 activity from stimulated muscle was unaltered at any of the concentrations tested. As expected, activity from control muscle was not modified (Fig. 3). When extracts were incubated in the presence of the catalytic subunit of cyclic AMP-dependent protein kinase, PFK-2 activity from denervated muscles decreased by more than 50%, whereas the PFK-2 activity from control and stimulated muscle did not show any change (Table II). Simultaneously, the extracts were also incubated in the presence of Mg 2ϩ to activate phosphatases. This treatment had no effect on PFK-2 activity, indicating that the enzyme was not phosphorylated at the start of incubation.
Effects of Denervation or Stimulation on PFK-2/FPBase-2 Transcript-In order to assess the effects of contractile muscular activity on the expression of PFK-2/FPBase-2, we determined the mRNA levels.
RNA was obtained from control, denervated, and stimulated TA muscle on the second day of treatment. The abundance of PFK-2/FPBase-2 mRNA was determined by Northern blot analysis (Fig. 4).
Denervated muscle showed a 25% reduction in PFK-2/FP-Base-2 mRNA content, and a further reduction was noted at day 7 (results not shown). In contrast, stimulation for 2 days led to a sharp increase (50%) in the specific mRNA levels.
Effects of Denervation or Stimulation on the Quantity of PFK-2/FPBase-2 Protein-In order to determine the relative amounts of PFK-2/FBPase-2 protein, we used polyclonal antiserum MCL-2 (30), which recognizes L-and M-type isoforms. The signal intensity from denervated muscle (Fig. 5A, lane D) was 10% of control muscle (lane C). In contrast, the signal from stimulated muscle (Fig. 5A, lane S) increased 2.5-fold.
To determine the L-type isoform, we used the polyclonal antiserum CL raised against the L-type isozyme (31). The signal from denervated muscle (lane D) was twice as strong as that signal from control muscle (lane C) (Fig. 5B). DISCUSSION We examined the influence of contractile activity on Fru-2,6-P 2 concentration and the influence of PFK-2/FBPase-2 activity on fast twitch skeletal muscle. To this end we chose two situations: (i) denervated fast twitch skeletal muscle, which is characterized by the absence of contractions, and (ii) chronic low frequency stimulated fast twitch skeletal muscle, which is characterized by increased contractile activity.
Our results show that both conditions induced an increase in Fru-2,6-P 2 and a modification of the L-and M-type PFK-2/ FBPase-2 isozyme expression. Moreover, we found these changes may depend on contractile activity. Denervation provoked an increase in L-type isozyme, while chronic low frequency stimulation induced an increase in M-type isozyme.
It has been found that Fru-2,6-P 2 concentration is dependent on substrate availability (7); however, the lack of variations in Fru-6-P and ATP in our experiments suggested that Glc-6-P, Fru-6-P, and ATP were not responsible for the increase in Fru-2,6-P 2 . Because Fru-2,6-P 2 concentration is also dependent on synthesis/degradation equilibrium, we focused our attention on the study of PFK-2/FBPase-2 activities.
In denervated muscle, the increased PFK-2 activity and the concomitant reduction in FBPase-2 could thus explain the increase in Fru-2,6-P 2 . The PFK-2/FBPase-2 ratio increased after denervation from 0.3 to 2.4. This change toward values characteristic of liver-type isozyme was interpreted as resulting from increased L-type isozyme synthesis. Since there is evidence that L-type could be differentiated from M-type by the inhibitory effect of sn-glycerol-3-phosphate and by the lower kinase activity after phosphorylation by cyclic AMP-dependent protein kinase (35), we proposed using these characteristics to examine the possible change from M-type to L-type. Our findings are consistent with an increase in L-type isozyme. This possibility was supported by an increased signal in Western blot analysis, with specific L-PFK-2/FBPase-2 antibody. In conclusion, denervation elicited a change from M-type to L-type in fast twitch skeletal muscle.
In stimulated muscle, the increase in Fru-2,6-P 2 was coincident with increases in kinase activity, which confirms previous results (11). Moreover, stimulation induces increases in FBPase-2. The PFK-2/FBPase-2 ratio therefore remained constant throughout the stimulation period. This ratio was similar to values described elsewhere in skeletal muscle from rat (14). The lack of variation in this ratio indicated that the isozyme type was not modified. This was confirmed by the absence of sn-glycerol-3-phosphate inhibition (Fig. 3) and the lack of effect of the catalytic subunit of cyclic AMP-dependent protein kinase on kinase activities (Table II). Since isozyme expression did not appear to have changed, the increase in kinase and phosphatase activity was attributed to a rise in M-type PFK-2/FB-  Pase-2 protein. This assumption was confirmed by the stronger signal in denervated than that in control muscle by Western blot analysis with the specific M-type PFK-2/FBPase-2 antibody, and this increase was coincident with a rise in PFK-2/ FBPase-2 mRNA. In conclusion, stimulation increased M-type in fast twitch rabbit skeletal muscle. The variation in contractile activity induced by denervation (3) or stimulation (6) has been described as causing transformation of fast twitch fibers into slow twitch fibers. In concordance with our results, these variations are manifested as modifications of enzyme activities and/or its isozyme distribution. In denervated fast twitch muscle, phosphoglycerate mutase isozyme pattern changes from MM-type toward MB and BB forms (36). Moreover, there is an increase in GLUT-1 (18,37) and HK-II (19,20). In the same way, muscle hyperactivity induced by stimulation causes increases in GLUT-4 (21) and HK-II (22), while lactate dehydrogenase isozyme changes its isozyme distribution toward H-type (38). The literature shows that contractile activity modifies protein expression (for a review, see Ref. 6). Furthermore, in transgenic mice the existence of nerve-and activity-dependent elements in the proximal promoter region of the human aldolase A gene has recently been identified, suggesting that multiple transcription programs are involved in fiber specificity (39).
The relationship between the transcriptional mechanism involved in denervation and/or chronic low frequency stimulation and the expression of L-and M-type PFK-2/FBPase isozyme in skeletal muscle remains to be elucidated.
In summary, in this paper we present evidence showing that PFK-2/FBPase-2 isozyme expression is modulated by contractile activity. The absence of muscular activity induced by denervation increases the expression of L-type PFK-2/FBPase-2, while the muscular hyperactivity induced by chronic low frequency stimulation increases the expression of M-type PFK-2/ FBPase-2. Moreover, both treatments cause a huge rise in Fru-2,6-P 2 concentration.  4. Expression of PFK-2/FBPase-2 mRNA on control, denervated, and stimulated rabbit tibialis anterior muscle. Total RNA was purified from 2-day denervated (D), 2-day stimulated (S), and contralateral tibialis anterior muscles (C). RNA was measured spectrophotometrically, and the integrity and relative amounts of RNA in each sample used were checked by ethidium bromide staining on the same gel (bottom of A). 20 g of total RNA from the different groups was applied on gels. PFK-2/FBPase mRNA was detected after hybridization with rat PFK-2/FBPase cDNA probe, as described under "Experimental Procedures." A representative autoradiogram is shown in A (top). Autoradiograms were subjected to scanning densitometry. The results (mean Ϯ S.E.)of three separate experiments are shown and expressed as a percentage of control value (B). *, significant difference (p Ͻ 0.05) between levels of control and denervated or stimulated muscles.