AMP-activated protein kinase (AMPK) may mediate the stimulatory effect of contraction and 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR) on glucose transport in skeletal muscle. In muscles with different fiber type composition from fasted rats, AICAR increased 2-deoxyglucose transport and total AMPK activity approximately twofold in epitrochlearis (EPI), less in flexor digitorum brevis, and not at all in soleus muscles. Contraction increased both transport and AMPK activity more than AICAR did. In EPI muscles, the effects of AICAR and contractions on glucose transport were partially additive despite a lower AMPK activity with AICAR compared with contraction alone. In EPI from fed rats, glucose transport responses were smaller than what was seen in fasted rats, and AICAR did not increase transport despite an increase in AMPK activity. AICAR and contraction activated both α1- and α2-isoforms of AMPK. Expression of both isoforms varied with fiber types, and α2 was highly expressed in nuclei. In conclusion, AICAR-stimulated glucose transport varies with muscle fiber type and nutritional state. AMPK is unlikely to be the sole mediator of contraction-stimulated glucose transport.
in skeletal muscle, glucose transport is increased by insulin and muscle contraction via translocation of the intracellular glucose transporter GLUT4 to the plasma membrane and T-tubules (for review see Refs. 24,44). These two stimuli have additive effects on glucose transport, suggesting that they act via at least partially independent pathways (40, 42). This view is further supported by the finding that each of the two stimuli recruits distinct pools of intracellular GLUT4 vesicles to the surface membrane of the muscle fibers (8, 15, 43). The insulin-sensitive pool appears to constitute a specialized population of vesicles that is distinct from the endosomal system (1, 43, 48). In contrast, contraction recruits GLUT4 vesicles from recycling endosomes defined by the presence of transferrin receptors (36, 43). Not surprisingly, then, the signal transduction pathways turn out to be distinct. Studies have shown that wortmannin, a phosphatidylinositol 3-kinase inhibitor, selectively impairs insulin- but not contraction-stimulated glucose transport (35, 37, 55). Until recently, the signaling mechanism by which contraction recruits GLUT4 transporters has remained obscure. A rise in the cytoplasmic Ca2+ concentration, as takes place during excitation-contraction coupling, has for a number of years been thought to be involved in contraction-induced glucose transport (for review see Ref. 27). The identification of a specific Ca2+ sensor has been elusive, but recently the protein kinase C inhibitor calphostin C has been shown to selectively impair contraction- but not insulin-stimulated glucose transport (29).
The possibility that another pathway may also be involved in contraction-induced activation of glucose transport was suggested by the recent observation that perfusion of rat hindquarters with the pharmacological agent 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR) resulted in increased glucose uptake (38). AICAR is known to be taken up into cells and converted to the monophosphorylated form ZMP, which mimics effects of AMP on AMP-activated protein kinase (AMPK), including both direct allosteric activation and promotion of phosphorylation by an upstream kinase (9, 25, 26, 49). AMPK is a heterotrimeric enzyme, consisting of a catalytic subunit (α) and two regulatory subunits (β and γ). Several isoforms of all three subunits have been identified (7, 21, 33). AMPK has been suggested to act as a “low-fuel warning system,” being switched on by depletion of ATP and then subsequently initiating energy-saving measures and ATP-generating processes (20). Because AMPK is known to be activated in muscle by contraction (28, 51,52), it was then speculated that the mechanism by which AICAR is able to increase muscle glucose uptake is similar to the mechanism by which contraction and hypoxia (both of which increase AMP in muscle) stimulate glucose transport (38). In favor of this view, it has been found that acute exposure to AICAR increases 2-deoxyglucose (2-DG) transport in skeletal muscle in a wortmannin-insensitive manner (3, 23) and induces GLUT4 translocation to the muscle fiber surface membrane (34). Furthermore, the effect of AICAR on glucose transport was additive to the effect of insulin but not additive to the effect of contractions (3, 23). Most of the cited studies have been performed on muscles from fasted or starved rats and on muscles containing a high proportion of fast-twitch type IIb fibers. However, in a recent study, high glycogen levels inhibited contraction-induced AMPK activation in slow-twitch, but not in fast-twitch, muscles (12). This suggests that mechanisms involved in contraction-stimulated glucose transport may be fiber type dependent and/or influenced by the nutritional state. A fiber type-related plasticity of AMPK regulation is also suggested by the in vivo effect of AICAR on muscle glucose transport (3) and GLUT4 expression (5).
In the present work, we have studied the acute effects of AICAR and contraction on glucose transport and AMPK activity in three different muscles with distinct fiber type composition from both fed and fasted rats. Furthermore, we have looked for fiber type-dependent expression and subcellular localization of the two isoforms of the catalytic subunit of AMPK, since it has been reported that it is primarily the α2-, and not the α1-isoform, that is activated by muscle contractions (51, 53).
MATERIALS AND METHODS
2-DG, glucose, bovine serum albumin fraction V (BSA), pyruvic acid, AICAR, ATP, ADP, AMP, IMP, and ZMP were from Sigma. 2-Deoxy-d-[3H]glucose and [14C]sucrose were from Du Pont-NEN. Polyvinylidene difluoride (PVDF) membranes (Sequi-Blot) were from Bio-Rad and the enhanced chemiluminescence kit from Amersham Pharmacia. Hexokinase, glucose-6-phosphate dehydrogenase, NADP, and NAD+ were from Boehringer, and other reagents were of analytical grade.
Muscle incubation and stimulation.
The experiments were approved by the Animal Experiments Inspectorate of the Danish Ministry of Justice. Fed or overnight-fasted male Wistar rats (60–70 g body wt), obtained from Charles River Laboratories (Sulzfeld, Germany), were anesthetized with an intraperitoneal injection of pentobarbital sodium (5 mg/100 g body wt). The rats were perfused through the left ventricle for 1 min (flow 20 ml/min) with Krebs-Henseleit bicarbonate-buffered medium containing 8 mM glucose, 1 mM pyruvic acid, and 0.2% BSA. The epitrochlearis (EPI), flexor digitorum brevis (FDB), and soleus muscles, each enriched with a distinct fiber type (Table 1), were gently dissected free with intact tendons at both ends and incubated for 30 min in perfusion medium in test tubes at 29°C. The medium was continuously gassed with 95% O2-5% CO2. After the 30-min equilibration period, muscles were placed at 37°C in glucose-free Krebs-Henseleit buffer containing 2 mM pyruvic acid and 0.2% BSA for ≤60 min with or without various concentrations of AICAR (see legends to figures). When the effect of contractions was studied, muscles were directly electrically stimulated to contract during the last 11 min of incubation at 37°C. A small clip was attached to each of the two tendons, and the muscle was vertically suspended in incubation medium, with or without AICAR, with the upper end attached to a force transducer connected to a computer for measurement of total force output (31). Electrodes were placed at both ends of the muscle, and 2 × 5 min of repeated tetanic contractions, separated by a 1-min break, were produced by stimulation with 25 V in 200-ms trains of 100 Hz, each impluse being 0.2 ms, and with a train frequency of 1/s (42). During the initial contractions, the length of the muscle was adjusted to yield maximum force; i.e., the muscle was suspended at its resting length.
Measurement of 2-DG transport.
2-DG transport was measured as described (29). Briefly, immediately after incubation with or without contractions, muscles were placed in test tubes at 29°C for measurement of glucose transport. Transport was measured as 2-deoxy-d-[3H]glucose ([3H]2-DG) uptake, with [14C]sucrose as extracellular marker. Isotopes and unlabeled sugars were added to the incubation medium (Krebs-Henseleit buffer containing 2 mM pyruvic acid and 0.2% BSA with or without AICAR) to yield final concentrations of 0.43 μCi [3H]2-DG and 0.32 μCi [14C]sucrose/ml and 1 mM of both unlabeled 2-DG and sucrose. After 10-min exposure to isotopes, muscles were briefly blotted on filter paper and immediately frozen in liquid nitrogen. Muscles were stored at −80°C until analyzed.
Measurement of metabolites and total and isoform-specific AMPK activity and expression by Western blot.
Muscles were quickly removed from the incubation medium and freeze-clamped with tongs cooled in liquid nitrogen and stored at −80°C until analyzed. For studies of contractions, muscles were freeze clamped while still connected to the electrodes and still contracting. Glycogen was determined by a hexokinase method after hydrolysis with HCl (32). Nucleotide concentrations were determined in neutralized perchloric acid extracts by reverse-phase HPLC (50). Separation was achieved by a 30-min gradient elution using a Hibar Lichrosphere 100 CH-18/2 (Merck) column (250 × 4 mm). The linear gradient program for the mobile phase was as follows: 0 min, 100% A; 0.1–4 min, 75% A; 4–17 min, 0% A; 17–22 min, 0–100% A. The composition of buffer Awas 150 mM ammonium phosphate (pH 5.80). Buffer B also contained 150 mM ammonium phosphate with 20% methanol and 2% acetonitrile added as organic modifiers (pH 5.45). Separation was achieved at room temperature with a flow of 0.8 ml/min. Detection was at 254 nm, and peaks were identified by comparison of retention times with commercially obtained compounds.
For AMPK activity determination, muscles were homogenized with a Polytron PT 3100 (Kinematica, Littau-Luzern, Switzerland) at maximum speed for 10–15 s in ∼20 volumes of ice-cold AMPK buffer (in mM: 210 sucrose,1 EDTA, 5 sodium pyrophosphate, 50 NaF, 1 dithiothreitol, 2 phenylmethylsulfonyl fluoride, 50 HEPES, pH 7.4). The homogenate was centrifuged for 45 s at 15,000 g, and the supernatant was stored in aliquots for later determination of protein concentration and total and isoform-specific AMPK activity. Sample protein was determined by the Coomassie Protein Assay Reagent (Pierce, Rockford, IL) with BSA as a standard.
Total AMPK activity was determined as previously described (12,31). AMPK activity is expressed as picomoles of phosphate incorporated into SAMS-peptide per milligram of protein in muscle sample supernatant during 1 min.
Isoform-specific AMPK activity was determined in immunoprecipitates from 200 μg of supernatant protein in a total volume of 500 μl of AMPK buffer. After 1-h incubation at 4°C with ∼1.5 μg of affinity-purified isoform-specific sheep IgG against either α1-AMPK or α2-AMPK, 25 μl of protein G Sepharose beads (Amersham Pharmacia) were added. Two hours later, beads were pelleted by centrifugation and washed once in AMPK buffer, and AMPK activity was determined in duplicate, as described above for total AMPK activity. The primary antibodies were raised against 15 amino acid peptides corresponding to the rat amino acid sequences (54). To verify the specificity, similar peptides were synthesized (α1: TSPPDSFLDDHHLTR; α2: MDDSAMHIPPGLKPH) and preincubated in a 5 molar excess with corresponding primary antibody before use in Western blot.
For Western blot, 10 μg of protein from either the supernatant or pellet, prepared as for AMPK activity determination, were separated by SDS-PAGE on Criterion 4–20% gradient gels (Bio-Rad) and electrophoretically transferred to PVDF membranes for 1.5 h at ∼1A. Transfer buffer contained 25 mM Tris, 192 mM glycine, 0.008% SDS, and 15% methanol. Membranes were blocked in 5% defatted milk powder plus 5% BSA in TS buffer (10 mM Tris, pH 7.4, 150 mM NaCl) and incubated for 90 and 60 min with primary and horseradish peroxidase-conjugated labeled secondary antibodies, respectively, diluted in blocking solution. Antigen-antibody complexes were visualized by enhanced chemiluminescence and preflashed film (Hyperfilm ECL, Amersham Pharmacia) and quantitated by a GS-710 densitometer (Bio-Rad). The aforementioned affinity-purified isoform-specific sheep antibodies against α1-AMPK and α2-AMPK were used at a concentration of 0.4 μg/ml.
Determination of muscle fiber type composition and immunohistochemical localization of AMPK α-subunits.
Pieces of muscles were mounted in Tissue-Tek (Miles, Elkhart, IN) and frozen in isopentane, cooled to its freezing point in liquid nitrogen. Serial transverse sections (20 μm) of the embedded muscle pieces were cut with a microtome at −20°C, mounted on glass slides, and stained for myofibrillar ATPase to identify fiber types (4). For localization of AMPK α1- and α2-subunits, transverse sections were fixed for 10 min in 2% depolymerized paraformaldehyde in 0.1 M phosphate buffer and blocked for 30 min with 50 mM glycine, 0.25% BSA, 0.04% saponin, and 0.05% sodium azide in PBS. Sections were then incubated for 1.5 h with primary antibodies diluted in blocking buffer. As primary antibodies, the previously described isoform-specific sheep anti-α1-and anti-α2-AMPK antibodies were used at a concentration of 0.2 μg/ml IgG together with a mouse monoclonal antibody against the slow myosin heavy-chain isoform (Sigma) for fiber type identification. After three washes of 10 min each in blocking buffer, sections were incubated for 45 min with secondary antibody conjugates (Alexa Fluor 488-labeled anti-sheep IgG and Alexa Fluor 594-labeled anti-mouse IgG from Molecular Probes, Eugene, OR) in blocking buffer, stained for 5 min with Hoechst 33342 (0.5 μg/ml) in blocking buffer, washed three times, and mounted in Vectashield (Vector Laboratories, Burlingame, CA). Confocal immunofluorescence microscopy was performed with a Leica TCS SP2 system (Leica, Mannheim, Germany).
Groups were compared by Student's unpaired t-test and one- or two-way ANOVA as applicable. Statistically significant differences were localized by the Student-Newman-Keuls method. P < 0.05 was considered significantly different in two-tailed tests.
Effect of rat size on AICAR-induced 2-DG uptake in EPI muscles.
Most studies of the effect of AICAR on glucose transport in skeletal muscle have been performed on rat EPI muscles. Due to its rather small size, this muscle is often obtained from larger rats of ∼100–130 g (22, 23) compared with the smaller rats (60–70 g) from which we usually obtain soleus muscles for studies of glucose transport regulation (29, 31). Figure1 shows the effect of 30-min preincubation with 2 mM AICAR on 2-DG uptake in EPI muscles from four groups of overnight-fasted rats weighing ∼65, 125, 185, and 245 g, respectively. AICAR had the largest effect in muscles from the smallest rats (P < 0.05), and the effect decreased progressively in the larger rats. In rats ∼65 g, AICAR increased glucose transport 2.7-fold compared with basal, whereas in muscles from rats ∼245 g transport was increased only 2.1-fold. Muscles were then obtained from rats weighing between 60 and 70 g in all subsequent experiments.
Time course effect and concentration dependency of AICAR on 2-DG uptake in EPI, FDB, and soleus muscles.
To find the preincubation time nescessary to give maximum effect of 2 mM AICAR on 2-DG uptake, EPI, FDB, and soleus muscles from 60- to 70-g overnight-fasted rats were incubated with or without AICAR for either 0, 15, 30, 45, or 60 min at 37°C before transport was assayed at 29°C. Incubation at 37°C without AICAR (basal) resulted in a progressive increase in glucose transport with time in EPI muscles (P < 0.05), whereas basal transport in soleus and FDB muscles was unaffected by incubation for up to 60 min (Fig.2). Incubation with AICAR gave the largest increase, both in absolute terms and relative to corresponding basal values, in EPI muscles. In EPI muscles, the effect peaked after 45 min, whereas in FDB muscles the effect of AICAR was already maximal after 15 min and stayed constant throughout the rest of the 60-min period. In soleus muscles, no effect on 2-DG uptake was observed even after 60-min incubation with AICAR (Fig. 2). By use of a preincubation time of 45 min at 37°C with different concentrations (0, 0.5, 1, 2, or 4 mM) of AICAR, it was found that 2 mM gave maximum 2-DG uptake in EPI muscles, whereas 1 mM gave maximum uptake in FDB muscles (Fig.3). In the soleus muscle, AICAR had no effect on 2-DG uptake, even when using 4 mM (Fig. 3). A preincubation time of 45 min at 37°C with or without 2 mM AICAR was then used in all subsequent experiments.
Comparison of effects of AICAR and contractions on 2-DG uptake and AMPK activity in different muscles from fed and fasted rats.
In EPI and FDB muscles from rats that had been fasted overnight, muscle glycogen concentrations were decreased ∼40% (P < 0.05) compared with muscles from fed animals [25.6 ± 2.2 vs. 14.9 ± 1.1 and 23.1 ± 2.6 vs. 14.4 ± 0.8 mmol glycosyl units/kg wet wt (means ± SE), n = 8–10]. In contrast, glycogen concentrations in soleus muscles obtained from fed and fasted animals were not significantly different (21.8 ± 1.0 vs. 19.6 ± 0.9 mmol glycosyl units/kg wet wt,n = 10, P > 0.05).
In muscles from fasted rats, AICAR increased 2-DG transport and total AMPK activity approximately twofold in EPI, less in FDB, and not at all in soleus muscles. In both fed and fasted rats, the effect of contractions on both glucose transport and AMPK activity was always higher than the effect of AICAR (Figs. 4and 5). Furthermore, the effects of the two stimuli combined on glucose transport was in EPI, but not in the other muscles, partially additive (Fig. 4). In contrast, the effect of combined stimulation on AMPK activity tended in all muscles to be lower than the effect of contractions alone (Fig. 5). Although responses in fed rats in general paralleled those seen in fasted rats, the effect of AICAR, contraction, or both stimuli combined on glucose transport was around twice as high in EPI muscles from fasted compared with fed rats. In fact, AICAR did not significantly increase glucose transport in EPI muscles from fed rats (P > 0.05), despite a similar increase in total AMPK activity, as seen in EPI muscles from fasted rats (Figs. 4 and 5). In FDB and soleus muscles, glucose transport responses were essentially identical in fed compared with fasted rats (Fig. 4). When all mean values from both fed and fasted animals for each muscle were combined, glucose transport varied directly with AMPK activity (Fig. 6). However, the correlation was not significant for EPI muscles (Fig. 6).
AMPK α1- and α2-isoform activity, expression, and immunohistochemical localization.
To determine which of the two known catalytic isoforms of the AMPK α-subunit that becomes activated after stimulation, α1- and α2-isoforms were immunoprecipitated from basal and AICAR- and contraction-stimulated EPI muscles from fasted rats. AICAR treatment increased the activity of both isoforms about threefold, and contraction increased the activity of both isoforms about ninefold compared with basal (Fig.7). Western blot showed that EPI muscles had the highest expression of both α1- and α2-isoforms. Furthermore, the soleus had the lowest expression of α1-AMPK, whereas FDB muscles had the lowest expression of α2-AMPK (Fig.8). In an attempt to morphologically verify the fiber type-specific expression, cryostat sections were stained with isoform-specific antibodies against either the α1- or α2-subunit. However, the staining did not reveal any difference between fiber types. In contrast, it was clearly seen that the α2-, but not the α1-subunit, was highly expressed in nuclei (Fig.9).
Nucleotide concentrations in muscles from fed and fasted rats.
The inability of AICAR to increase glucose transport and AMPK activity in soleus muscles might be due to lack of uptake or conversion of AICAR to ZMP. However, the concentration of ZMP varied between 1 and 2 μmol/g in all three different muscle types, being highest in the FDB and lowest in the soleus (Table 2). A few differences were found in nucleotide concentrations between experimental groups. Of note is the finding that AICAR treatment reduced AMP concentrations ∼20% in all three muscles, irrespective of the nutritional state (P < 0.05, Table 2).
The focus of the present study was to explore possible fiber type-specific differences in the ability of AICAR and contractions to increase AMPK activity and glucose transport in skeletal muscle. AICAR was first demonstrated to increase glucose uptake, measured as arteriovenous difference, in the perfused rat hindlimb (38), which contains a mixture of predominantly fast-twitch muscle fibers (2). It was proposed that AICAR mimics the effect of contraction on AMPK and, in turn, glucose transport (38). Shortly afterward, it was demonstrated that AICAR treatment of incubated EPI muscles from fasted rats increased glucose transport [measured using 3-O-methylglucose (23)] and GLUT4 translocation (34) and that the effect was additive with the effect of insulin but not additive with the effect of contraction (3, 23). In the present study, we found that AICAR increased glucose transport (measured using 2-DG) in EPI muscles from fasted, but not from fed, rats and, surprisingly, that the combined effect of AICAR and contractions was significantly higher than the effect of each stimulus alone (Fig. 4). This partial additivity suggests that AICAR and contraction activate glucose transport only in part by the same mechanism. The lack of additivity in previous studies (3, 23) compared with the present findings may relate to the different protocols that were used for inducing muscle contractions.
In contrast to our findings in EPI muscles, AICAR-stimulated glucose transport in FDB muscles was rather small (50% vs. 2-fold in EPI) and in soleus muscles was completely absent (Fig. 4). Furthermore, in FDB and soleus muscles, the combined effect of AICAR and contraction on glucose transport was not any different from the effect of contraction alone. This was not due to submaximal effect of AICAR, as a maximum effective dose and exposure time of AICAR had been determined in advance (Figs. 2 and 3). The inconsistencies among different muscles with distinct fiber type compositions (Table 1) were also not due to differential accumulation of ZMP, which reached concentrations between 1 and 2 μmol/g in all three muscles, similar to what has been reported elsewhere (34). This shows that AICAR is taken up and converted to its monophosphorylated form ZMP, irrespective of muscle type. The concentration of ZMP is ∼50-fold higher than the endogeneous AMP concentration (Table 2). It may be speculated that ZMP displaces AMP from intracellular binding sites, making it more accessible to degradation. This could provide an explanation for the observation that AICAR treatment reduced AMP concentrations ∼20% in all three muscles studied, irrespective of nutritional state (Table 2). However, this is not a universal finding (3, 38), which may be ascribed to methodological differences.
AICAR-stimulated glucose transport was significantly higher (Fig. 4), and glycogen levels lower, in EPI and FDB muscles from fasted than from fed rats (P < 0.05). This corresponds with the fact that both contraction- and insulin-stimulated glucose transport and cell surface GLUT4 content have been shown to correlate negatively with the initial glycogen content (12-14). Accordingly, it is possible that the lack of enhancement by overnight fasting of the AICAR effect in the soleus muscle (Fig. 4) is related to the lack of glycogen depletion during overnight fasting in this muscle. A crucial role of glycogen was suggested more than 15 years ago by studies showing that insulin-independent glucose uptake and transport during and after contraction are increased only if muscle glycogen is below the resting level (16, 41, 45, 46).
It has previously been reported that glucose transport in rat soleus muscle, as also found in the present study, is resistant to acute treatment with AICAR (11), even though this is not a universal finding (3, 17). The discrepancies among various studies observed after acute AICAR treatment of soleus muscles may relate to the nutritional state of the animals or depend on whether experiments were carried out in vitro or in vivo. In one study, 24-h starvation of very small rats (17) may have resulted in glycogen depletion, even in the soleus muscle. In an in vivo study, glucose uptake was also increased in the soleus muscle after intravenous infusion of AICAR (3). However, it cannot be excluded that, in that study, an observed ∼50% reduction in plasma free fatty acid levels after AICAR infusion may have contributed to the increase in muscle glucose uptake (19). Recently, it has been reported that treatment of rats for 5 days with subcutaneous injections of AICAR increased insulin-stimulated glucose transport in muscles containing a high proportion of type IIb fibers, but not in soleus muscle, with a majority of type I fibers (5). It thus appears that both acute and chronic treatment with AICAR increase glucose transport mainly in muscles with a high proportion of type IIb fibers. This may indicate that important signaling components in AICAR-stimulated glucose transport are variably expressed in different muscle fiber types.
The mechanism by which AICAR and contraction increase glucose transport is thought to be through an activation of AMPK (3, 6, 17, 23,52). If AMPK is an essential intermediary in the signaling pathway for these two stimuli, then variations in stimulus-induced AMPK activation are expected to be paralleled by corresponding variations in glucose transport. In fact, the present study provides such evidence. In conditions that increased glucose transport, total AMPK activity was always increased (Figs. 4 and 5). Furthermore, in all muscles and dietary conditions studied, the effects of contraction on both glucose transport and total AMPK activity were always higher than the effects of AICAR (Figs. 4 and 5). Corresponding with this, the effect of contraction on the activity of the α1- and α2-catalytic subunits of AMPK (see below) was also higher than the effect of AICAR. When all mean values from both fed and fasted rats were combined, glucose transport varied directly with total AMPK activity within each type of muscle (Fig. 6). However, the correlation was not significant in EPI muscle. The latter finding reflects our finding that glucose transport in EPI muscles was almost twofold higher in muscles from fasted compared with fed rats despite identical AMPK activities. However, it must be kept in mind that the degree of AMPK activity measured in cell-free extracts only reflects the degree of phosphorylation of the enzyme, whereas allosteric modifications likely to play a role in vivo are lost during preparation. Several findings in support of this view have been discussed elsewhere (53). Another reason for the lack of significant correlation between glucose transport and total AMPK activity in EPI muscle is that, in this muscle, glucose transport was highest upon combined stimulation with AICAR and contraction, whereas AMPK activity was highest on stimulation with contraction alone (Fig. 5). In fact, in all three muscles treated with both AICAR and contraction, the total AMPK activity was decreased compared with muscles stimulated with contraction alone (Fig. 5). This decrease in AMPK activity was not due to less force being produced when muscles were stimulated to contract in the presence of AICAR (data not shown). However, it is possible that activation of the AMPK is less efficient when intense contractions producing much AMP are carried out in the presence of a high intracellular concentration of ZMP (Table 2). It has been shown with purified rat liver AMPK that, at high concentrations, both AMP and ZMP become less efficient at allosteric activation of the enzyme, possibly due to competition with ATP at the catalytic site (9, 26). It is conceivable that a similar mechanism may be operating for the covalent activation of AMPK by the upstream kinase and thus may explain the present observation that combined treatment with contraction and AICAR was less efficient in activating AMPK than contractions alone (Fig. 5).
Expression of the two isoforms, α1 and α2, of the catalytic subunit of AMPK did not vary in parallel with fiber type as determined by Western blot. Furthermore, the two isoforms showed distinct subcellular compartmentalization, as evidenced by at least partial nuclear localization of α2 (Fig. 9) but not α1, which has also been observed in INS-1 cells, a transformed cell line derived from rat pancreatic β-cells (47). Interestingly, the α2-subunit has been implicated in the regulation by glucose of islet β-cell gene expression (10). It may thus be speculated that, in muscle, the α2-subunit is involved in, for example, AICAR- and possibly exercise-induced GLUT4 expression (5). Despite the fact that the combined expression of both isoforms was at least twice as high in EPI as in FDB or soleus muscles (Fig. 8), basal and contraction-stimulated total AMPK activity tended to be higher in the soleus compared with the other two muscles (Fig. 5). The explanation for this is not readily apparent. However, one possibility is that, during incubation conditions, the comparatively rather thick soleus muscle was under more hypoxic stress than the other muscles. In line with this interpretation, we have found that, in the soleus muscle, AMPK activity is lower during basal conditions as well as after contraction, when incubation is at 29°C compared with 37°C (data not shown).
Immunoprecipitation with the use of isoform-specific antibodies showed that both α1- and α2-catalytic subunits of AMPK are activated by AICAR as well as by muscle contraction (Fig. 7). An effect of AICAR on both isoforms is consistent with previous studies (22). In contrast, muscle contraction induced by intense electrical stimulation of the sciatic nerve in anesthetized rats increased the activity of only the α2- but not the α1-isoform of AMPK (51). Furthermore, during high-intensity exercise in humans (∼70–75% of maximal oxygen uptake), only activation of α2, but not α1, could be detected (18, 53). However, a 30-s bicycle sprint exercise resulted in a two- to threefold increase in the activity of both α1- and α2-isoforms in human skeletal muscle (6). It is thus possible that activation of the α1-isoform requires more intense muscle contractions as used during the present in vitro conditions or during in vivo sprint exercise (6). It has been suggested that this makes α1 less likely to play a role in the metabolic response to ordinary exercise in skeletal muscle (18, 51, 53). Surprisingly, however, also during exercise of relatively low intensity in humans, where glucose uptake is known to be increased, no activation of α2 could be detected (18, 53). This suggests that mechanisms other than activation of AMPK may increase glucose transport during contractions, which is consistent with several observations in the present study and with our previous observation that, in glycogen-supercompensated soleus muscles, glucose transport was increased sixfold during contractions with no increase at all in AMPK activity (12). In further support of a dissociation between AMPK and contraction-stimulated glucose transport, expression in mouse muscle of a dominant inhibitory mutant of AMPK completely blocked the ability of hypoxia and AICAR to increase glucose uptake, whereas the effect of contraction was only partially reduced (39).
In conclusion, the present study has shown that both AICAR- and contraction-stimulated glucose transport in muscle vary with muscle fiber type as well as nutritional state. Contraction is a more potent stimulus for activation of both glucose transport and AMPK activity than is AICAR. The amount of AMPK α1- and α2-catalytic subunit isoforms is higher in fast-twitch than in slow-twitch fibers. Activity of both isoforms is enhanced by AICAR as well as by contraction, which is consistant with the idea that AMPK is involved in increasing glucose transport during contraction. However, several of the present observations may also be interpreted as indicating a dissociation between AMPK- and contraction-stimulated glucose transport. For instance, the combination of contraction plus AICAR reduces total AMPK activity, whereas glucose transport is either maintained or even increased compared with the effect of contraction alone. Furthermore, the effect of contraction on AMPK is highest in the soleus muscle, which at the same time has the smallest contraction-induced glucose transport. Available data are thus consistent with a scenario in which contraction may enhance glucose transport by a dual mechanism: a feed-forward control elicited early during muscle contractions and possibly involving Ca2+ as well as a feedback control associated with mechanical performance and, in turn, metabolism and elicited by reduced intracellular energy status and AMPK (30, 31).
We thank Gerda Hau and Conni Temdrup for skillful technical help.
This work was supported by grants from the Danish National Research Foundation (504–14), the Velux Foundation, the Danish Diabetes Foundation, and the Novo Nordic Foundation. H. Ai was the recipient of a fellowship sponsered by DANIDA, Danish Ministry of Foreign Affairs. G. Hardie was supported by a Programme Grant from the Wellcome Trust, UK.
Address for reprint requests and other correspondence: T. Ploug, Dept. of Medical Physiology, The Panum Institute, Blegdamsvej 3, DK-2200 Copenhagen N, Denmark (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2002 the American Physiological Society