INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

REGULAR EXERCISE HAS BENEFICIAL EFFECTS for some pathological conditions, such as cardiovascular disease and diabetes (12, 14). It is believed that many of these salutary effects originate in skeletal muscle (36, 40). Skeletal muscle, by virtue of the mass, is an important mechanism for extrarenal regulation of potassium balance. Associated with the aforementioned diseases is an attending or potential consequence of disturbed potassium balance. Several reports have provided convincing evidence for a benefit from exercise training on the regulation of potassium balance (7, 13, 20). Therefore, transport mechanisms that mediate potassium uptake by muscle tissue, and their adaptation, are of interest.

A large body of evidence demonstrates activation of Na⁺-K⁺-ATPase in response to muscle contraction or exercise, and this increase in activity is associated with more balanced potassium plasma concentrations during and after exercise (7, 18, 20, 22, 23). However, other transport mechanisms also appear to be active (13, 17). In this regard, we have recently provided evidence that stimulated Na⁺-K⁺-2Cl cotransporter (NKCC) activity in skeletal muscle may contribute to the overall potassium uptake. Our data indicate -adrenergic agonists and contractile activity, known stimulators of Na⁺-K⁺-ATPase activity, are also significant stimulators of NKCC activity (6, 43). However, the exercise adaptation of skeletal muscle NKCC expression and activity has not been evaluated. The potential regulation of specific molecular mechanisms controlling NKCC activity in response to exercise are also unknown.

Numerous studies have pointed out the significance of mitogen-activated protein kinase (MAPK) signaling pathways and signaling through Akt (protein kinase B) in the metabolic and mitogenic effects of muscle contraction or exercise training (1, 33, 36, 39, 40, 45). A classical extracellular signal-regulated kinase (ERK) MAPK cascade is a major signaling network by which cells can regulate cell growth, differentiation, and adaptation (41). The emerging evidence suggests that contractile activity and exercise are powerful stimulators of the ERK MAPK pathway (1, 11, 15, 29, 33), although the physiological relevance of this effect in skeletal muscle is still unclear (11, 42). Akt can also be activated by muscle contraction (30, 36, 37). To elucidate the physiological importance of ERK and Akt pathways, recent studies conducted in our laboratory (6, 43) showed that stimulus-specific ERK activation is necessary for NKCC-mediated potassium uptake by skeletal muscle, and Akt activation can be a point of convergence to control the signals for NKCC activation. Also, fiber type-specific differences add to the complexity of signaling and metabolic events in response to distinct stimuli (6). Therefore, it was of interest to evaluate the impact of exercise training on the signaling pathways that regulate NKCC activity in different muscle phenotypes.

Here we show that exercise increased total potassium transport and NKCC activity in skeletal muscle in association with activation of the ERK-MAPK signaling pathway. Furthermore, the data also indicate a rapid and persistent exercise adaptation of signaling through the -adrenergic receptor such that NKCC activity is inhibited.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials. ⁸⁶RbCl was from New England Nuclear (Boston, MA). Isoproterenol (Iso) and bumetanide were from Sigma Chemical (St. Louis, MO). Phospho-specific antibodies to ERK and anti-ERK2 were obtained from Santa Cruz Biotechnology, (Santa Cruz, CA). Polyclonal rabbit antibodies to phospho-Akt-Ser⁴⁷³, phospho-Akt-Thr³⁰⁸, Akt, phospho-p38 MAPK, and p38 MAPK were from Cell Signaling Technology, (Beverly, MA). All other chemicals were purchased from Sigma Chemical.

Animal care and exercise protocol. Female Sprague-Dawley rats were used for all experiments. Rats were housed in light- and temperature-controlled quarters where they received food and water ad libitum. The Animal Care and Use Committee of the University of Tennessee, Health Science Center, approved all procedures.

Muscle incubation and ⁸⁶Rb/K influx constant calculation. One tendon of each soleus and plantaris muscle was attached with 4-0 surgical silk suture to glass wands for rapid transfer among solutions. The unloaded condition has been shown by Nielsen and Clausen (22) to allow larger ion transients in muscle. Muscles were preincubated for 15 min at 30°C in preincubation medium [oxygenated Krebs-Ringer containing bumetanide (10⁵ M) or vehicle (DMSO) for the contralateral muscle]. For Iso stimulation, after preincubation, muscles were transferred directly to incubation medium (oxygenated Krebs-Ringer containing 1 µCi/ml ⁸⁶Rb and either bumetanide or vehicle) at 30°C, which contained 30 µM Iso. Incubation was for 10 min. After the acute bout of exercise, muscles were taken directly in the incubation medium. To assess the involvement of MAPK signaling in the activation of NKCC, a MAPK kinase (MEKK) 1,2 inhibitor, PD-098059 (20 µM), was added to the incubation medium (11, 42, 43). To determine whether Iso worked by activating a -adrenergic receptor pathway, the -adrenergic antagonist propranolol (1 µM) was used in the medium. Muscles were immediately washed with ice-cold 0.9% saline solution. After the muscles were washed, they were blotted, weighed, and homogenized in 2 ml of 0.3 M TCA. ⁸⁶Rb uptake by the muscle was measured by Cerenkov counting. ⁸⁶Rb transport was expressed as a rate constant, as described previously (43).

NKCC activity. The bumetanide-sensitive ⁸⁶Rb rate constant was used as an index of NKCC activity (43). The bumetanide-sensitive portion of ⁸⁶Rb uptake was calculated by subtracting the bumetanide treatment rate constant for the muscle of one hindlimb from the vehicle treatment value of the contralateral muscle.

Intracellular signaling assays. Muscles were preincubated as described. Incubation medium did not include ⁸⁶Rb. After incubation, muscles were removed and clamp frozen in liquid nitrogen and stored at 80°C for immunoblot analysis. Muscles were homogenized in 10% ice-cold lysis buffer containing 10 mM Tris · HCl, pH 7.4, 5 mM EDTA, 50 mM NaCl, 30 mM Na₄P₂O₇, 50 mM NaF, 100 µM Na₃VO₄, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 2 µg/ml aprotinin, 2 µg/ml leupeptin, 2 µg/ml antipain, and 1 µg/ml pepstatin A with a Teflon pestle and centrifuged at 4°C for 5 min at 5,000 g. Protein concentration of the supernatant was measured by the bicinchoninic acid assay (Pierce, Rockford, IL).

NKCC protein expression. One hundred micrograms of protein were mixed with SDS denaturing buffer, warmed to 95°C for 5 min, and electrophoresed on a 7.5% SDS-PAGE gel. The gels were electroblotted onto polyvinylidene difluoride membranes. The membranes were incubated at room temperature for 1 h in the blocking buffer (1.5 mM NaH₂PO₄, 8 mM Na₂HPO₄, 0.15 M NaCl, and 0.3% Triton X-100, pH 7.4) supplemented with 5% nonfat dry milk. Immunologic reactions were performed at room temperature for 1 h in PBS-T containing 5% nonfat dry milk and a specific monoclonal antibody raised against the 310 C-term residues of human colonic NKCC1 (T4, from the Developmental Studies Hybridoma Bank, University of Iowa) used at a 1:50 dilution. The membrane was subsequently washed with PBS-T buffer and incubated for 1 h at room temperature with a horseradish peroxidase-conjugated goat anti-mouse IgG secondary antibody. The proteins of interest were visualized by chemiluminescent exposure of X-ray film (ECL Plus, Amersham). Bands were quantitated by video densitometry.

Analysis of ERK, p38 MAPK, and Akt phosphorylation. Aliquots of supernatants containing an equal amount of protein were mixed with SDS denaturing buffer, warmed to 95°C for 5 min, and electrophoresed on a 10% SDS-PAGE gel. The gels were electroblotted onto polyvinylidene difluoride membranes. The membranes were incubated overnight at 4°C in blocking buffer supplemented with 3% BSA. All immunologic reactions were performed at room temperature for 1.5 h in blocking buffer containing 1% BSA and the specific antibody. All primary antibodies were used in the dilution of 1:1,000. Phospho-specific antibodies to Akt phosphorylated on Ser⁴⁷³ and Thr³⁰⁸, ERK1/ERK2 dually phosphorylated on Thr²⁰² and Tyr²⁰⁴, and p38 MAPK dually phosphorylated on Thr¹⁸⁰ and Tyr¹⁸² were used to detect the catalytically activated forms of the kinases. Incubation for 45 min in 1% BSA blocking buffer with horseradish peroxidase-conjugated goat anti-mouse IgG or goat anti-rabbit IgG secondary antibody followed. Total protein expression was determined by stripping and reprobing the Western blots with antibodies against the total specific protein. The proteins of interest were visualized and quantitated as described for NKCC. Protein phosphorylation was calculated as the ratio of phospho- to total protein expression normalized to the basal level value (taken as 1.0). Because mechanical damage or hypoxia during muscle bath incubation (4, 31) may inadvertently initiate signaling through the MAPK pathway (16), we examined ERK1,2 and p38 MAPK phosphorylation in animals whose muscles were clamp-frozen immediately on removal. There were no differences between these muscles and the muscles that have been incubated for 25 min in the Krebs-Ringer solution.

Citrate synthase activity. To asses the training effect on the skeletal muscles, citrate synthase (CS) activity of muscle homogenates was measured as described by Srere and Brooks (34).

Statistics. Comparisons within and among treatments were made by ANOVA and analysis of covariance. Differences between treatments were considered significant at P < 0.05. Data are reported as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effect of acute exercise and Iso on total and NKCC-mediated ⁸⁶Rb uptake. A 30-min bout of treadmill running or Iso stimulated total ⁸⁶Rb uptake in skeletal muscle by up to 60% (Fig. 1A). This effect was mediated, in part, by activation of bumetanide-sensitive ⁸⁶Rb uptake (Fig. 1B). Bumetanide is a specific inhibitor of NKCC activity. Recently, we have demonstrated that inhibitors of ERK1,2 MAPK activation abolished contractile activity-induced NKCC activation in skeletal muscle (43). To test whether ERK1,2 MAPK stimulation by acute exercise could contribute to the increase of NKCC-mediated ⁸⁶Rb uptake, we inhibited ERK1,2 activation with the MEK inhibitor PD-098059 (11, 42). Indeed, the stimulatory effect of acute exercise on bumetanide-sensitive ⁸⁶Rb uptake was blunted by PD-098059 in both the soleus and plantaris muscles (Fig. 1B). This was accompanied by a significant decrease of total ⁸⁶Rb uptake (Fig. 1A). In nonexercised skeletal muscle, PD-098059 alone has no effect on NKCC-mediated ⁸⁶Rb uptake (data not shown and Ref. 43). In contrast to the effects of exercise and Iso alone on NKCC-mediated ⁸⁶Rb uptake, we observed that Iso inhibited NKCC activity in acutely exercised muscle (Fig. 1B). Similarly, Iso stimulation of exercised muscle decreased total ⁸⁶Rb uptake by 15-20% compared with Iso alone (Fig. 1A). The attenuating effect of Iso on exercise-stimulated NKCC activity was reversed by the -adrenergic antagonist propranolol. Blockade of the -adrenergic receptor by propranolol in either nonexercised or exercised skeletal muscle in the basal conditions did not alter the rate of ⁸⁶Rb uptake (data not shown and Ref. 43). Because the 30-min exercise bout exhibited a profound stimulatory effect on NKCC activity, we tested whether an exercise training program would also upregulate NKCC activity.

View larger version (36K): [in this window] [in a new window]   Fig. 1.   Acute exercise stimulated total and bumetanide-sensitive 86Rb uptake. Both activation of the -adrenergic receptor and a mitogen-activated protein kinase kinase (MEK) 1,2 inhibitor abolished the exercise-induced effect. Isolated soleus and plantaris muscles were incubated in the presence or absence of 30 µM isoproterenol, 1 µM propranolol, or 20 µM PD-098059 during 10 min of radiolabel uptake. Total (A) and Na+-K+-2Cl cotransporter (NKCC)-mediated (B) 86Rb uptake were determined. NKCC-mediated 86Rb uptake was calculated as the difference between vehicle- and bumetanide-treated muscles. Values are given as means ± SE of n = 6-8 experiments. P < 0.05 compared with *nonstimulated (basal) muscle, isoproterenol (Iso)-stimulated muscle of sedentary rats, and exercised values.

Effect of exercise training on body weight and CS activity. Rats began the 2- and 4-wk running programs such that they were the same age at death. Body weights of trained rats were not significantly different compared with sedentary controls after 2 or 4 wk of the exercise protocol (165 ± 5 vs. 172 ± 7 g and 172 ± 2 vs. 177 ± 3 g, respectively). Also, the weights of the both soleus and plantaris muscles were not changed by exercise training (Table 1). Endurance exercise training is known to increase oxidative metabolism. The exercise protocol used here produced a distinct increase in CS activity. After 2 wk of training, CS activity was significantly increased in the soleus muscle (P < 0.05) and was not altered in the plantaris muscle. Four weeks of training did not change CS activity in the soleus muscle but increased CS activity in the plantaris muscle (P < 0.01; Table 1).

Effect of exercise training and Iso on total and NKCC-mediated ⁸⁶Rb uptake. Total basal ⁸⁶Rb uptake increased 32 and 23% in the soleus muscle after 2 and 4 wk of exercise, respectively (P < 0.05) (Table 1). The plantaris muscle of the trained rats also had significantly greater total ⁸⁶Rb uptake by 21 and 15% after 2 and 4 wk, respectively (P < 0.05; Table 1). Thus we further investigated whether NKCC-mediated ⁸⁶Rb uptake contributed to the overall increase in total uptake by trained muscle.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Exercise training stimulated bumetanide-sensitive 86Rb uptake in a phenotype-specific and time-dependent manner. Iso abolished the training-induced effect. After preincubation with bumetanide or vehicle for 15 min, isolated soleus (A) and plantaris (B) muscles were stimulated with 30 µM Iso during 10 min of radiolabel uptake. NKCC-mediated 86Rb uptake was calculated as the difference between vehicle- and bumetanide-treated muscles. Values are given as means ± SE of n = 6-8 experiments. P < 0.05 compared with *nonstimulated (basal) muscle, Iso-stimulated muscle of sedentary rats, and exercised values.

Expression of NKCC protein after exercise training. As shown in Fig. 2, endurance exercise led to a significant upregulation of NKCC activity in a phenotype-specific manner. To establish whether there was an increase in NKCC protein expression, we performed immunoblot analysis of muscle lysates. Figure 3 depicts upregulation of NKCC by 14% (P < 0.01) in the soleus muscle after 2 wk and in the plantaris muscle after 4 wk by 27% (P < 0.05). This could account for upregulation of NKCC activity in the trained muscles. However, inverted regulation of NKCC activity by Iso and a lack of correlation between expression and activity in sedentary and acutely exercised muscle indicated additional, acute regulatory mechanisms.

View larger version (39K): [in this window] [in a new window]   Fig. 3.   Exercise training significantly elevated NKCC-protein expression in slow-twitch soleus muscle after 2 wk and in fast-twitch plantaris muscle after 4 wk of training. Representative blots are shown. Values are means ± SE of n = 20 experiments. *P < 0.01 compared with sedentary muscle.

Effect of exercise and Iso on ERK1,2 phosphorylation. Our previous reports demonstrate that ERK activity is critical for NKCC activation by Iso and contractile activity (6, 43). Because Iso inhibited NKCC activity in exercised muscle, we next characterized the effect of exercise on Iso-induced ERK phosphorylation. Consistent with recent reports (1, 5, 25, 26, 40), exercise significantly increased the basal level of ERK phosphorylation compared with sedentary control muscle. As shown in Fig. 4, an acute bout of exercise increased phosphorylation of ERK in both soleus and plantaris muscle by 97 and 30% (P < 0.05), respectively. However, exercise training showed a time-dependent pattern of ERK activation. In the soleus muscle, 2 wk of training increased ERK phosphorylation by 40% (P < 0.05), and this effect disappeared with 4 wk of exercise (Fig. 5A). In contrast, exercise training resulted in a 90% increase in ERK phosphorylation in the plantaris muscle after only 4 wk of the training regimen (P < 0.01; Fig. 5B). In agreement with our laboratory's previous reports (6, 43), Iso stimulation of skeletal muscle of sedentary rats significantly increased ERK phosphorylation by up to 65% from basal level (Figs. 4 and 5). Iso-stimulated ERK1,2 phosphorylation in exercise-trained soleus muscle was not significantly different from Iso-stimulated sedentary muscles (Fig. 5A). In contrast, Iso-stimulated ERK1,2 phosphorylation was dramatically lower in the trained plantaris muscle compared with Iso stimulation of the sedentary plantaris muscle (Figs. 4B and 5B). ERK-modulating effects of exercise and Iso stimulation cannot be attributed to the changes in ERK2 protein expression. As shown in Fig. 4, there were no differences in ERK2 abundance (per µg protein) among sedentary, exercised, and stimulated muscles (P > 0.05). Likewise, exercise training did not alter ERK2 protein expression in skeletal muscle (not shown). The characteristic ineffectiveness of Iso to stimulate NKCC activity or even inhibit existing NKCC activity and ERK phosphorylation in exercised muscle is consistent with the hypothesis that additional signaling pathways can regulate stimulus-specific NKCC activation (6).

View larger version (33K): [in this window] [in a new window]   Fig. 4.   Effect of acute exercise and Iso on extracellular signal-regulated kinase (ERK) 1,2 phosphorylation. Isolated soleus and plantaris muscles were incubated in the presence or absence of 30 µM Iso. Twenty-five micrograms of protein from muscle lysates were immunoblotted with an antibody to phospho-ERK1,2. Immunoblots were then stripped and reprobed with an anti-ERK2 antibody. Representative blots are shown. The intensity of phosphorylated ERK1,2 was quantified by densitometry, and the phospho-protein-to-total ERK2 protein ratio was calculated. Values are expressed relative to the basal level of phosphorylation in sedentary muscle (taken as 1.0). Data were obtained from 6-8 different muscles (means ± SE). P < 0.05 compared with *nonstimulated (basal) muscle, Iso-stimulated muscle of sedentary rats, and exercised values.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Exercise training enhanced the basal level of ERK1,2 phosphorylation. Isolated soleus (A) and plantaris (B) muscles were incubated in the presence or absence of 30 µM Iso. Twenty-five micrograms of protein from muscle lysates were immunoblotted with an antibody to phospho-ERK1, 2. Immunoblots were then stripped and reprobed with an anti-ERK2 antibody. The intensity of phosphorylated ERK1,2 was quantified by densitometry, and the phospho-protein-to-total ERK2 protein ratio was calculated. Values are expressed relative to the basal level of phosphorylation in sedentary muscle (taken as 1.0). Insets: representative immunoblots of dually phosphorylated ERK1,2 from lysate of soleus (A) and plantaris (B) muscle corresponding to an appropriate experimental condition. Data were obtained from 6-10 different muscles (means ± SE). P < 0.05 compared with *nonstimulated (basal) muscle, Iso-stimulated muscle of sedentary rats, and exercised values.

Effect of exercise and Iso on Akt and p38 MAPK phosphorylation. Akt and p38 MAPK activation can inhibit ERK1,2 phosphorylation and, in turn, NKCC stimulation (6). Therefore, we assessed the activation of these molecules by exercise and Iso. For immunodetection of Akt, two different antibodies that recognize Akt phosphorylation on Ser⁴⁷³ and Thr³⁰⁸ were used. An acute bout of exercise produced a significant increase in Akt Ser⁴⁷³ and Thr³⁰⁸ phosphorylation compared with sedentary muscles. In the soleus muscle, the magnitude of exercise-induced phosphorylation on Thr³⁰⁸ was larger than on Ser⁴⁷³ (Fig. 6). Because an acute bout of exercise exhibited a significant activation of Akt, we next determined phosphorylation of Akt after exercise training. Consistent with acute exercise results, 2- and 4-wk exercise training enhanced the phosphorylation of Akt on Ser⁴⁷³ by twofold in both the soleus and the plantaris muscle (P < 0.01; Fig. 7). Exercise training produced a similar activation of Akt phosphorylation on Thr³⁰⁸ (not shown). Iso stimulation of the muscle did not alter the basal level of Akt phosphorylation on Ser⁴⁷³ or Thr³⁰⁸ exhibited in muscle from sedentary or trained rats (Figs. 6 and 7). Acute exercise or Iso stimulation did not alter Akt protein expression in either muscle (Fig. 6). The pattern of Akt expression in the trained muscle was similar to that in the sedentary muscle (not shown). In agreement with previous reports (6, 44), p38 MAPK phosphorylation was not activated by an acute bout of contractile activity or Iso stimulation in the slow-twitch soleus muscle but was significantly increased in the fast-twitch plantaris muscle (1.8 ± 0.2- and 1.5 ± 0.1-fold, respectively, n = 6, P < 0.05). Stimulation of acutely exercised soleus muscle with Iso resulted in a 2 ± 0.3-fold increase of p38 MAPK phosphorylation (n = 6, P < 0.05). The effects of acute exercise and Iso on p38 MAPK activity were additive in the plantaris muscle.

View larger version (31K): [in this window] [in a new window]   Fig. 6.   Acute exercise increased the phosphorylation of Akt on Ser473 and Thr308. Isolated soleus and plantaris muscles were incubated in the presence or absence of 30 µM Iso. One hundred micrograms of protein from muscle lysates were immunoblotted with an anti-phospho-Ser473 (A) or a Thr308 Akt (B) antibody. Immunoblots were then stripped and reprobed with an anti-Akt antibody. Representative blots are shown. The intensity of each band was quantified by densitometry, and the phospho-protein-to-total protein ratio was calculated. Values are expressed relative to the basal level of phosphorylation in sedentary muscle (taken as 1.0). Data were obtained from 6 different muscles (means ± SE). P < 0.05 compared with *nonstimulated (basal) and Iso-stimulated muscle of sedentary rat.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Exercise training increased the phosphorylation of Akt on Ser473. Isolated soleus (A) and plantaris (B) muscles were incubated in the presence or absence of 30 µM isoproterenol. One hundred micrograms of protein from muscle lysates were immunoblotted with an anti-phospho-Ser473 Akt antibody. Immunoblots were then stripped and reprobed with an anti-Akt antibody. The intensity of each band was quantified by densitometry, and the phospho-protein-to-total protein ratio was calculated. Values are expressed relative to the basal level of phosphorylation in sedentary muscle (taken as 1.0). Insets: representative immunoblots of Akt phosphorylated on Ser473 from lysates of soleus (A) and plantaris (B) muscle corresponding to an appropriate experimental condition. Data were obtained from 6-10 different muscles (means ± SE). P < 0.05 compared with *nonstimulated (basal) and Iso-stimulated muscle of sedentary rats.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

These data provide evidence that not only does exercise lead to the adaptation of NKCC-mediated potassium transport by skeletal muscle but also to a surprising reversal of stimulatory and inhibitory signal pathways. In this context, the potential physiological consequences of the exercise are twofold: improved potassium buffering and remodeling of muscle intracellular signal pathways. These will be discussed in the following paragraphs.

Improved potassium buffering as a result of regular exercise (10, 32) may provide a significant margin of safety against the risk of cardiotoxic hyperkalemia at a given level of work. Although Na⁺-K⁺-ATPase activity is necessary, previous evidence raises the possibility that other mechanisms also exist to regulate potassium transport by skeletal muscle (13, 17). Our current and recent findings indicate that NKCC activity could be an additional mechanism. Catecholamines, contractile activity (6, 43), and acute exercise are able to activate NKCC (Fig. 1). Similarly, exercise training also increased the basal level of NKCC activity (Fig. 2). It appears that training produced part of its effect on NKCC activity by increasing expression of NKCC (Fig. 3). Catecholamines are critical for reducing the exercise-mediated hyperkalemia that is a consequence of the mismatch between release of potassium by contracting muscle and the uptake of potassium by muscle and other tissues (32). -Adrenergic agonists activate Na⁺-K⁺-ATPase, reduce the magnitude of hyperkalemia associated with exercise, and delay the onset of contraction-induced fatigue (3, 9, 23, 27). Conversely, exercise during -adreno blockade impairs the clearance of increased plasma potassium (8, 38). Although the evidence for the necessity of -adrenergic receptor stimulation to reduce the exercise-induced rise in plasma potassium is clear, less is known about the consequences of catecholamine action in contracting muscle. Intriguingly, it has been observed that, during -adrenergic stimulation, working muscle accumulates more potassium in the extracellular space than in the absence of -agonists. Administration of the -adrenergic agonist terbutaline before an exercise bout significantly reduces the rate of potassium reuptake by working muscle, indicating that the action of terbutaline on exercised muscle differs from the resting muscle (9, 27). This raises the possibility that exercising muscle reverses the action of catecholamines, possibly due to altered G-protein-coupled signaling (24). Our results indicate that exercise-induced changes in the regulation of NKCC activity may participate in this phenomenon. We demonstrated the lack of NKCC activation by Iso in exercised muscle, opposite to the -adrenergic agonist action in sedentary muscle (Figs. 1 and 2). Indeed, Iso stimulation of exercised muscle could not evoke further stimulation of total ⁸⁶Rb uptake above that elicited by exercise in the soleus muscle; Iso stimulation of exercised plantaris muscle inhibited total ⁸⁶Rb uptake relative to exercise alone (Fig. 1). In agreement with previous reports (8, 38), our findings support the hypothesis that, during or after exercise, catecholamines act mainly on nonexercised, resting muscle to buffer potassium. Adaptation of the skeletal muscle intracellular milieu was considered for the regulation of NKCC activity in response to exercise. As discussed below, insight into the molecular mechanism for NKCC regulation by exercise comes from analysis of ERK MAPK, p38 MAPK, and Akt signaling, recently shown to be regulators of NKCC activity (6).

The physiological consequences of the ERK MAPK pathway remodeling in skeletal muscle should not be underestimated. This pathway is activated in response to insulin, contractile activity, and exercise (1, 15, 29, 33). Although it is clear that these stimuli likely produce changes in gene expression through the ERK pathway, the metabolic consequences are less well-known. Our recent reports indicate that ERK signaling plays an essential role in hormone-mediated NKCC activation (6, 43). In the present study, our finding of pharmacological blockade of NKCC activity by an inhibitor of ERK activation and correlation between NKCC and ERK activation demonstrate that ERK signaling is important for exercise-induced cotransporter activity (Figs. 1 and 2). However, the data presented here demonstrate the complexity of interactions involving the ERK pathway that are uncovered by exercise. On one hand, Iso inhibited exercise-mediated ERK phosphorylation in the fast-twitch plantaris muscle, and this was consistent with the ERK dependency of NKCC activity. On the other hand, in the slow-twitch soleus muscle, the apparent inhibition of NKCC by the -adrenergic agonist was not accompanied by ERK MAPK inactivation (Figs. 4 and 5). We previously reported that other intracellular processes play a role in conjunction with the ERK pathway to regulate NKCC activity (6).

The phosphatidylinositol 3-K/Akt pathway appears to be one mechanism that can inhibit NKCC activation (6). Although we have not directly shown that Akt inhibits the ERK MAPK pathway, others have reported that Akt can inhibit Raf-1 activation in skeletal muscle cells (28); Raf-1 activates MEKK1,2, which in turn activates ERK1,2. In the present study, it is unlikely that Akt plays an inhibitory role. Iso action in exercised muscle, which abolished NKCC activity (Figs. 1 and 2), did not further increase Akt activity compared with trained muscle that was not stimulated with Iso (Figs. 6 and 7). Another mechanism to downregulate ERK MAPK activity, and subsequently NKCC activity, is stimulation of p38 MAPK (6). p38 MAPK may activate MAPK phosphatases, which, in turn, leads to the dephosphorylation and inhibition of ERK MAPK activity (2, 41). In the plantaris muscle, treatment with Iso after exercise resulted in a significantly higher level of p38 MAPK activation and a notable decrease in ERK MAPK activation. Therefore, it is possible that Iso may inhibit NKCC activity in exercised fast-twitch muscle through a mechanism involving p38 MAPK activation of MAPK phosphatases. Although this mechanism is possible in the plantaris muscle, ERK MAPK activation in the soleus muscle was not affected by Iso stimulation after an acute bout of exercise despite the stimulation of p38 MAPK. These data indicate possible phenotypic differences in signal transduction.

Others have reported that different intracellular signaling pathways are activated in distinct muscle phenotypes in response to contractile activity (19, 21, 44). In agreement with these reports, we found that ERK MAPK activation is somewhat greater in the predominantly slow-twitch soleus muscle than in the predominantly fast-twitch plantaris muscle. Also consistent with the previous study (44), soleus muscle p38 MAPK was not activated after exercise. It may be that the lower level of ERK MAPK activation in the plantaris muscle was due to p38 MAPK activation, considering the antagonizing action of p38 MAPK on ERK MAPK activation (2, 41). However, this does not appear to be the case for exercised soleus muscle stimulated with Iso. Nevertheless, phenotypic differences in ERK activation in response to exercise training, whether through expression or signal pathway crosstalk, likely reflect the distinct patterns of functional demand between slow- and fast-twitch muscle (35). Taken together, our data and the reports of others indicate that exercise training produces a consistent pattern of adaptation of intracellular signal pathways. In turn, these adaptations appear to have an impact on muscle NKCC activity.

An emerging body of evidence indicates that muscle contraction and exercise are able to activate Akt in skeletal muscle (30, 36, 37). Although initial reports demonstrated no effect of contractile activity or exercise on Akt activity (33, 40), more detailed examination reveals a consistent increase in Akt phosphorylation and activation by muscle contraction, irrespective of muscle phenotype (30). Consistent with these findings, our data show that both an acute bout of exercise and exercise training increased the level of Akt phosphorylation (Figs. 6 and 7). Therefore, these results corroborate recent evidence that upregulation of Akt activity may be a significant contributor to the beneficial effects of exercise on the metabolic disturbances associated with insulin resistance and diabetes.

Despite these data, the size of the muscles from the 170-g rats that were used for the in vitro tissue bath treatments (necessitated by the training duration) raises the possibility of hypoxia due to limited oxygen diffusion to the core of the muscle (4, 31). It appears that hypoxia was not a dominant factor in the results obtained after 25 min of incubation at 30°C. Both Iso and contractile activity stimulate total and bumetanide-sensitive ⁸⁶Rb uptake in that soleus and plantaris muscles of smaller rats (50-100 g) (6, 43), and the pattern and magnitude of the stimulation are not different from the data reported here (Fig. 1). Furthermore, the pattern of ERK MAPK and p38 MAPK phosphorylation of muscle frozen immediately on removal from control and exercised animals was not different from basal levels observed in the muscle incubated at 30°C for 25 min. In addition, the magnitude and mode of ERK MAPK, p38 MAPK, and Akt activation for slow- and fast-twitch phenotypes reported here in response to Iso stimulation and contractile activity are not different from those reported previously for much smaller muscles (6, 30, 43, 44). Therefore, although certainly a concern, the concurrence of the data for large and small muscles indicates that hypoxia due to muscle size is not responsible for the reported data.

In summary, these experiments demonstrate that an exercise adaptation of the skeletal muscle signaling pathways may influence NKCC activity. As a potential transport mechanism for potassium, NKCC activity in muscle provides a component of whole-body potassium metabolism. Thus exercise-induced muscle adaptations could affect potassium dynamics in response to physiologically meaningful stimulators such as -adrenergic agonists.