Regulation of Na+–K+ pump activity in contracting rat muscle

Authors


Abstract

  • 1In rat soleus muscle, high frequency electrical stimulation produced a rapid increase in intra-cellular Na+ (Na+i) content. This was considerably larger in muscles contracting without developing tension than in muscles contracting isometrically. During subsequent rest a net extrusion of Na+ took place at rates which, depending on the frequency and duration of stimulation, approached the maximum transport capacity of the Na+–K+ pumps present in the muscle.
  • 2In isometrically contracting muscles, the net extrusion of Na+ continued for up to 10 min after stimulation, reducing Na+i to values 30% below the resting level (P < 0.001). This undershoot in Na+i, seen in both soleus and extensor digitorum longus muscles, could be maintained for up to 30 min and was blocked by ouabain or cooling to 0 °C.
  • 3The undershoot in Na+i could be elicited by direct stimulation as well as by tubocurarine-suppressible stimulation via the motor endplate. It could not be attributed to a decrease in Na+ influx, to effects of noradrenaline or calcitonin gene-related peptide released from nerve endings, to an increase in extracellular K+ or the formation of nitric oxide.
  • 4The results indicate that excitation rapidly activates the Na+–K+ pump, partly via a change in its transport characteristics and partly via an increase in intracellular Na+ concentration. This activation allows an approximately 20-fold increase in the rate of Na+ efflux to take place within 10 s.
  • 5The excitation-induced activation of the Na+–K+ pump may represent a feed-forward mechanism that protects the Na+–K+ gradients and the membrane potential in working muscle. Contrary to previous assumptions, the Na+–K+ pump seems to play a dynamic role in maintenance of excitability during contractile activity.

In skeletal muscle, excitability and thus contractility depend on the membrane potential and the chemical gradients for Na+ and K+. In resting muscle, the passive Na+–K+ fluxes are low and the active transport necessary to maintain the chemical gradients of these ions only amounts to a few per cent of the theoretical maximum transport capacity of the Na+–K+ pump (Clausen, 1986). During contractions, however, the influx of Na+ and efflux of K+ can be dramatically elevated, and an efficient activation of the Na+–K+ pump at the onset of contraction is necessary to prevent a rapid run-down of the Na+–K+ gradients and an ensuing reduction in excitability. Several studies have shown that electrical stimulation can increase the rate of active Na+–K+ transport substantially. In perfused dog hindlimb, Hazeyama & Sparks (1979) reported a 65% increase in Na+–K+ pump activity during 4 Hz electrical stimulation. Measurements of intracellular Na+ (Na+i) activity showed that repetitive stimulation with 40 Hz trains in isolated mouse soleus muscle led to a 10-fold increase in active Na+–K+ transport, corresponding to a transport rate of 45% of the theoretical maximum capacity of the Na+–K+ pumps in the muscle (Juel, 1986). Using the same approach, a net Na+ extrusion rate close to the maximum capacity for active Na+–K+ transport was observed after intermittent 120 Hz stimulation in frog semitendinosus muscle (Balog & Fitts, 1996). In rat soleus, 60 Hz stimulation induced a 12-fold increase in the activity of the Na+–K+ pump within 10s, corresponding to a transport rate of 66% of the maximum Na+–K+ pump capacity (Everts & Clausen, 1994). The high transport rate observed after high frequency stimulation indicates that during intensive exercise skeletal muscle may utilize the major part of its Na+–K+ pump capacity.

The increase in active Na+–K+ transport in contracting muscle can partly be explained by an excitation-induced increase in intracellular Na+ concentration ([Na+]i), which is a potent stimulus for the Na+–K+ pump (for review see Semb & Sejersted, 1996). During recovery from electrical stimulation the net extrusion of Na+ may, however, continue even after [Na+]i has reached the pre-stimulated value (Juel, 1986; Everts & Clausen, 1994). Moreover, in rat soleus, 2 Hz electrical stimulation has been shown to increase Na+–K+pump activity 2-fold without any change in [Na+]i (Everts & Clausen, 1992, 1994). This indicates that mechanisms other than increased [Na+]i contribute to the activation of the Na+–K+ pump during excitation.

Based on these observations it was of interest to test to what extent the activity of the Na+–K+ pumps in skeletal muscle is increased by various types of high frequency excitation, and to examine the underlying mechanisms. Part of the observations presented here have briefly been reported in a preliminary form (Clausen & Nielsen, 1994; Nielsen & Clausen, 1996a, b).

METHODS

Animals and muscle preparation

All handling and use of animals complied with Danish animal welfare regulations. Most experiments were carried out using 4-week-old female or male Wistar rats weighing 60–70 g. For examination of the effect of denervation, rats weighing 55–59 g were anaesthetized with ether. The sciatic nerve of one of the legs was cut, and the cut end of the proximal part of the nerve was looped back and ligated. The animals had free access to food and water, and were kept at a constant temperature (21°C) and day length (12 h).

Animals were killed by decapitation, and the intact soleus or extensor digitorum longus (EDL) muscles with tendons were dissected out as described previously (Nielsen & Clausen, 1996c). The wet weight of the muscles from 4-week-old rats ranged from 20 to 25 mg. In rats denervated for 7 days, the soleus muscle of the denervated leg weighed from 11 to 22 mg and the contralateral muscle from 33 to 37 mg. Incubation took place at 30°C in a standard Krebs–Ringer bicarbonate buffer containing (mm): 120.1 NaCl, 25 NaHCO3, 4.7 KCl, 1.2 KH2PO4, 1.2 MgSO4, 1.3 CaCl2 and 5 d-glucose (pH7.4) under continuous gassing with 95%O2-5%CO2.

Electrical stimulation

In vitro experiments

For electrical stimulation in vitro, isolated muscles were either mounted at optimal resting length (leading to isometric contractions when stimulated) or placed on sets of platinum electrodes positioned 3 mm apart and surrounding the central part of the muscle. Using the latter set-up, muscles were allowed to shorten freely without exerting external force. Following mounting, the muscles were equilibrated in standard Krebs–Ringer bicarbonate buffer for 30–60 min without stimulation. If not otherwise noted, the muscles were stimulated using pulses of 1 ms duration and a supramaximal voltage of 12 V. In some experiments the pulse duration was reduced to 0.02 ms in order to produce maximal stimulation through the motor nerve of the muscle preparation. Control experiments demonstrated that when 0.02 ms pulses were used, pre-incubation of muscles with 10−5m tubocurarine completely abolished both the excitation-induced accumulation of Nat and force development.

In situ experiments

To determine net fluxes of Na+ during electrical stimulation of soleus muscles in situ, rats were anaesthetized with an intraperitoneal injection of pentobarbitone (60 mg (kg body weight)−1 Mebumal; SAD, Copenhagen, Denmark). The gastrocnemius–plantaris–soleus muscle group was exposed leaving the blood and nerve supply intact, and a pair of platinum electrodes was placed in close contact with the central part of the soleus muscle. The legs were clamped, with the knee and the ankle joint at fixed angles to a disc of cork, using needles. The fixation ensured that no movements of the leg or of the two joints took place during electrical stimulation. The soleus muscle was stimulated directly using 12 V pulses of 1 ms duration. For isometric contractions, the muscle was stimulated with tendons intact at both ends, whereas for contractions without the exertion of external force the distal tendon of the soleus muscle was separated from the other muscles and cut at the ankle joint before stimulation. At the end of the experiment the animals were killed by decapitation whilst still anaesthetized.

Isometric force development

Isolated muscles were mounted vertically with their tendons intact in thermostatically controlled chambers and were stimulated directly by supramaximal pulses (12 V, 1 ms pulse duration) through two platinum electrodes (field stimulation). Isometric force was measured using a force displacement transducer (Grass FTO3) and recorded with a chart recorder calibrated with standard weights. After adjustment to optimal length (resting tension, 0.015 ± 0.002 N, n= 8) and initial tests of contractility, muscles were equilibrated for at least 30 min without stimulation before the test of force production. Contractile force was expressed per gram of wet weight of muscle. In sixteen muscles the maximal tetanic force exerted at 60 Hz was 16.4 ± 0.4 N (g wet wt)−1.

Intracellular Na+ and muscle water

For determination of Na+i content muscles were immediately moved to ice-cold Na+-free Tris–sucrose buffer and washed for 15 min 4 times, a procedure which in a previous study was shown to remove all extracellular Na+ (Everts & Clausen, 1992). In addition the same study showed that part of the Na+i was lost and that this loss could be corrected for by multiplying the Na+ content of the muscles at the end of washout by 1.46. Several control experiments made in the present study (for methods see Everts & Clausen, 1992) showed that the fractional loss of Na+i during washout at 0°C was independent of prior electrical stimulation and the mode of contraction. A correction factor of 1.46 was therefore used for the calculation of Na+i content in all muscles undergoing washout. After the 4 × 15 min washout at 0°C, tendons were cut off, and the muscles were blotted and weighed for determination of wet weight. They were then soaked overnight in 0.12 m trichloroacetic acid (TCA) which gave complete extraction of Na+i from the tissue (Clausen, Andersen & Flatman, 1993). The concentration of Na+ in the TCA extract was measured by flame photometry (FLM3; Radiometer, Copenhagen, Denmark) and the muscle Na+ content was expressed as μmol (g wet wt)−1. If not otherwise noted, this procedure was used for determination of Na+i content in all experiments.

In a few experiments the Na+i content was determined without washout on the basis of measurements of total Na+ content and the extracellular water space. Muscles were equilibrated for 70 min in buffer containing [14C]sucrose, blotted, weighed and subsequently dried overnight at 60°C for determination of dry weight. These experiments also allowed the calculation of the intracellular water content of muscles that had not been washed in Tris buffer by using the total water content and the distribution volume of [14C]sucrose. This showed that in resting soleus muscles the intracellular water content was 0.58 ± 0.01 ml (g wet wt)−1n= 6). A similar value (0.57 ± 0.01 ml (g wet wt)−1, n= 6) was found in muscles stimulated for 30 s at 60 Hz, 10 min before the determination of water content. These values were used for calculation of [Na+]i expressed as mmol (1 cell water)−1.

22Na influx and Na+–K+ pump activity

Passive 22Na influx during recovery from electrical stimulation was determined by exposing muscles to the standard incubation buffer containing 22Na(1 μCi ml−1) for 5 min starting 2 min after cessation of electrical stimulation. For determination of the excitation-induced influx of 22Na, pairs of contralateral muscles were pre-incubated for 5 min in standard incubation buffer containing 22Na (0.5 μCi ml−1), whereafter one of the muscles was stimulated for 10 or 30 s at 60 Hz. The excitation-induced influx was calculated from the difference in intracellular 22Na between the resting and the stimulated muscle. At the end of the influx experiments the muscles were immediately transferred to ice-cold Na+-free Tris–sucrose buffer and underwent a 4 × 15 min washout. The muscles were then blotted and weighed, and the activity of 22Na retained in the muscles was determined by γ-counting. After correction for loss of intracellular 22Na during the washout at 0°C (see above), the Na+ influx was calculated from the specific activity of 22Na in the incubation buffer.

The basal Na+–K+ pump activity in resting control muscles was calculated from the ouabain-sensitive efflux of 22Na. The efflux of 22Na was determined as described in detail elsewhere (Clausen & Kohn, 1977). In brief, muscles were loaded with 22Na for 60 min in Krebs–Ringer buffer containing 10 μCi ml−1. This was followed by an 8 × 10 min washout in non-radioactive buffer with or without ouabain (10−3M). Based on measurements of the 22Na activity retained in the muscle at the end of washout and that released to the incubation tubes in each 10 min period, the fractional loss of 22Na per gram of wet weight was calculated as previously described (Everts & Clausen, 1994). The fractional loss of 22Na that was sensitive to ouabain was then converted to Na+ efflux by multiplying with the Nat content of resting muscles. In muscles stimulated electrically the activity of the Na+–K+ pump above the baseline level of resting muscles was determined from the net decrease in Na+i content.

Chemicals and isotopes

All chemicals were of analytical grade. [14C] Sucrose and 22Na were from Amersham. Insulin was a gift from Novo Nordisk (Copenhagen, Denmark). Saxitoxin was bought from Calbiochem, calcitonin gene-related peptide (CGRP) from Peninsula Laboratories, Inc. (St. Helens, Merseyside, UK), and dl-propranolol from Ferrosan (Søborg, Denmark). Dantrolene, salbutamol, l-NAME, ouabain and tubocurarine were from Sigma.

Statistics

All data are expressed as means ±s.e.m. The statistical significance of any difference was ascertained using Student's two-tailed t test for non-paired observations.

Results

Effects of electrical stimulation on intracellular Na+

Figure 1 shows the time course of the changes in Na+i content in isometrically contracting soleus muscles stimulated at 60 Hz. During 30 s of stimulation Na+i increased by 3.2 ± 0.9 μmol (g wet wt)−1. After the cessation of stimulation, Na+ was rapidly re-extruded and within 1 min Na+i content was reduced to the level in unstimulated control muscles. The net extrusion of Na+ continued, however, for several minutes resulting in a highly significant ‘undershoot’ in Na+i content to 32% below that of unstimulated control muscles. This was followed by a slow reversal to the control level, and after 45 min of recovery Na+i was no longer significantly different from that of the controls. A significant undershoot in Na+i content could also be elicited using lower frequencies or shorter duration of stimulation. Thus, 10 min after the cessation of 5 or 15 s stimulation at 60 Hz, Na+i was reduced by 23 and 29%, respectively, compared with resting controls (4 versus 4 muscles, P < 0.05 and 4 versus 4 muscles, P < 0.005). Likewise, 10 min after 1 min of stimulation at 5 Hz, Na+i was reduced by 27% (4 versus 4 muscles, P < 0.001). In experiments where Na+i content was measured upon the cessation of stimulation, 20 min at 2 Hz was found to lower Na+i content by 24% (from 15.3 ± 0.5 to 11.6 ± 0.8 μmol (g wet wt)−1; 5 versus 6 muscles, P < 0.005).

Figure 1.

Time course of changes in intracellular Na+ content (Na+i) during and after electrical stimulation in soleus muscles contracting isometrically

At each time point indicated, muscles were immediately transferred to ice-cold Na+-free Tris–sucrose buffer, washed 4 × 15 min, blotted and taken for determination of Na+ content (see Methods for details). ○, resting control muscles; •, muscles stimulated for 30 s at 60 Hz as indicated by the bar. Each point represents a mean value ±s.e.m. of 6–27 muscles. Significant difference between stimulated muscles and corresponding controls: *P < 0.05; **P < 0.001.

In EDL muscles a similar excitation-induced undershoot in Na+i content was observed. As shown in Fig. 2, 5 min after 2 s stimulation at 30 or 90 Hz, Na+i was reduced by 26 and 27%, respectively, compared with resting controls, P < 0.01). These observations indicate that the undershoot in Na+i content during recovery from excitation is a general phenomenon in both slow- and fast-twitch muscles.

Figure 2.

Na+i content 5 min after 2 s of electrical stimulation in EDL muscles contracting isometrically

Na+i was determined as described in Fig. 1. □, resting control muscles; inline image, muscles stimulated for 2 s at 30 or 90 Hz. Each point represents a mean value ±s.e.m. of 6–12 muscles. Controls and stimulated muscles are significantly different, *P < 0.01.

To examine the excitation-induced undershoot in Na+i content in detail a series of control experiments were carried out where the Na+i of soleus muscles stimulated for 30 s at 60 Hz and allowed to rest for 10 min were compared with control muscles receiving no stimulation. Firstly, since the Na+i of the muscles was expressed per gram of wet weight, a possible excitation-induced increase in muscle water content could reduce the values obtained for Na+i. However, experiments where muscles were dried before extraction with TCA (see Methods for details) showed that when Na+i was expressed per gram of dry weight a similar undershoot was observed (from 75 ± 2 μmol (g dry wt)−1 in controls to 51 ± 2 μmol (g dry wt)−1 in the muscles stimulated at 60 Hz; n= 19 in both cases, P < 0.001). Moreover, the water content of the muscles, as determined after the 4 × 15 min washout at 0°C (see Methods for details), was unaffected by prior electrical stimulation (3.50 ± 0.19 g H2O(g dry wt)−1 in controls versus 3.42 ± 0.19 g H2O (g dry wt)−1 in stimulated muscles; n= 6 in both cases). Another potential source of error is that during the 4 × 15 min washout in the ice-cold Na+-free buffer, the stimulated muscles might lose more Na+ than the unstimulated controls. Therefore, in a separate experiment, Na+i content was calculated from measurements of total Na+ content, deducting the Na+ present in the extracellular space available to [14C] sucrose. Using this procedure without washout in the cold, Na+i content was 9.5 ± 0.5 μmol (g wet wt)−1 in controls and 5.4 ± 0.3 μmol (g wet wt)−1 in the muscles stimulated at 60 Hz (n= 12 in both cases, P < 0.001). The conclusion from these control experiments is that the excitation-induced undershoot in Na+i (Figs 1 and 2) cannot be explained by experimental errors introduced by water shifts in the muscles or by increased Na+ loss during the washout procedure, but reflects a true reduction in Na+i content.

Excitation-induced changes in Na+–K+ pump activity

Figure 3 shows that both the initial net extrusion of Na+i following 60 Hz stimulation and the development of an undershoot in Na+i were abolished by cooling the muscle preparation to 0°C. Likewise, the extrusion of Na+ was blocked by pre-incubation of muscles with ouabain (10−3 M) before stimulation (data not shown). Measurements of the uptake of 22Na from the second to the seventh minute after the cessation of stimulation gave values for Na+ influx identical to those measured in unstimulated contralateral control muscles (1300 ± 300 and 1300 ± 200 nmol (g wet wt)−1 min−1, respectively; n= 6 in both cases). Taken together, these results show that the net extrusion of Na+i following excitation (Fig. 1) reflects an increased Na+–K+ pump activity rather than decreased Na+ permeability of the muscle fibres.

Figure 3.

Effect of cooling to 0°C on the re-extrusion of Na+i following electrical stimulation in soleus muscles contracting isometrically

Na+i was determined as described in Fig. 1. ○, resting control muscles at 30°C; •, muscles at 30°C stimulated for 30 s at 60 Hz as indicated by the bar. Dashed lines indicate groups of stimulated muscles that were transferred to 0°C upon the cessation of stimulation (▪) or 2 min after the cessation of stimulation (▴). Each point represents a mean value ±s.e.m. of 6–27 muscles.

Figure 4 shows the rate of net extrusion of Na+i in isometrically contracting soleus muscles during the first 50 s following 10–30s of stimulation at 60–120 Hz. The values were compared with the theoretical maximum Na+–K+ pump capacity calculated on the basis of reported values for the concentration of 3H-labelled ouabain binding sites in soleus muscles from 4-week-old rats (Clausen, Everts & Kjeldsen, 1987). It appears that after stimulation the net rate of Na+ extrusion reached values between 30 and 60% of the maximum transport capacity of the Na+–K+ pumps in the preparation. As calculated from measurements of the ouabain-sensitive efflux of 22Na, the rate of active Na+ extrusion in resting muscles suspended at optimal length was 500 ± 90 nmol (g wet wt)−1 min −1 (Fig. 4). Thus, the net extrusion rates of Na+i found after 60–120 Hz stimulation corresponds to a 5- to 10-fold increase in Na+–K+ pump activity (Fig. 4).

Figure 4.

Na+ efflux in resting and electrically stimulated soleus muscles contracting isometrically

Na+i was determined as described in Fig. 1. □, rate of ouabain-suppressible Na+ efflux in resting muscles. inline image, rate of net Na+ extrusion after 10 or 30 s stimulation at 60, 90 or 120 Hz. The efflux of Na+ in resting muscles was calculated from measurements of the ouabain-sensitive efflux of 22Na. The Na+ extrusion rate in stimulated muscles was calculated from the net changes in Na+i content during the first 50 s of recovery from stimulation. The theoretical maximum capacity for active Na+ transport at 30°C was calculated from Clausen et al. (1987). Each point represents a mean value ±s.e.m. of 10–18 pairs of muscles.

A comparison of the excitation-induced increase in Na+i in muscles contracting isometrically with previously published data from muscles contracting without exerting external force indicated that the accumulation of Na+ was less in the former. To examine this, the accumulation of Na+i during 10s of stimulation at 60 or 120 Hz was determined in soleus muscles contracting either isometrically or without exerting external force (see Methods for details). Table 1 shows that whereas the mode of suspension was without effect on the Na+i content in resting muscles, the net accumulation of Na+i during stimulation was around 5-fold larger in muscles contracting without exerting force than in muscles contracting isometrically. To further substantiate this finding the excitation-induced influx of 22Na was determined in both isometrically contracting muscles and in muscles contracting without exerting external force. The muscles were stimulated at 60 Hz for 30 s (isometric contractions) or 10 s (contractions without external force). Based on these measurements the influx of Na+ per action potential was 2.4 ± 0.6 and 9.5 ± 1.6 nmol (g wet wt)−1 in the two modes of contraction, respectively, corresponding to a 4-fold difference.

Table 1. Intracellular Na+ content before and after 10 s of electrical stimulation at 60 or 120 Hz in rat soleus muscles contracting without exerting external force or isometrically
 Na+i 
Experimental conditionNo external force(μmol (g wet wt)−1)Isometric (μmol (g wet wt)−1)P
  1. P values indicate significance of difference between muscles contracting isometrically and muscles contracting without exerting external force. n. s.: not significant at the 5% level. Values are means ±s.e.m. with the number of muscles in parentheses.

Isolated muscles
   Rest14.7 ± 0.4 (15)15.5 ± 0.6 (16)n. s.
   10 s at 60 Hz22.0 ± 0.6 (19)16.5 ± 0.3 (18)< 0.001
   10 s at 120 Hz27.3 ± 0.9 (20)18.2 ± 0.3 (14)< 0.001
In situ
   Rest11.6 ± 0.5 (6)10.7 ± 0.4 (6)n. s.
   10 s at 120 Hz18.1 ± 0.9 (6)14.6 ± 0.5 (6)< 0.01

To test whether the difference in the excitation-induced Na+ influx between the two modes of contraction was related to the difference in the force exerted, the Na+ accumulation during 10 s of 60 Hz stimulation was examined in isometrically contracting muscles where the resting tension was increased by 0.1 N from the 0.015 N corresponding to the optimal length (see Methods). This reduced the active force exerted during 60 Hz stimulation by 41 ± 3%, but had no significant effect on the increase in Na+i during excitation (1.8 ± 0.9 μmol(g wet wt)−1 (n= 5 pairs of muscles) compared with 1.0 ± 0.7 μmol(g wet wt)−1 in isometrically contracting muscles with maximal force development; Table 1). Likewise, 30 min incubation at increased resting tension (0.1 N) was without effect on Na+i content in resting muscles (data not shown).

A difference in the net Na+ accumulation between the two modes of contraction was also observed in muscles stimulated at 120 Hz in situ although it appeared to be considerably smaller (1.7-fold difference, Table 1). Because of the time needed to isolate the muscles after in situ stimulation, the transfer to ice-cold Na+-free buffer was delayed by 10–15 s. During this interval part of the Na+ taken up during stimulation may have been re-extruded by the Na+–K+ pump. Therefore, the values obtained for Na+i content probably underestimate the actual excitation-induced increase in Na+i. In spite of this, it seems reasonable to conclude that in isolated muscles as well as in situ, the excitation-induced accumulation of Na+i depends on the mode of contraction, with isometric contractions causing less increase in Na+i than contractions without force production. It was of interest, therefore, to examine the re-extrusion of Na+ after high frequency stimulation in muscles contracting without exerting external force.

Figure 5 shows the rate of ouabain-suppressible 22Na efflux in resting control muscles and the rate of net Na+ extrusion after 10 s of stimulation at 60–120 Hz. As estimated from the net change in Na+i during the first 50 s of recovery after stimulation, excitation led to a 10 to 18-fold increase in Na+ extrusion, approaching the theoretical maximum transport capacity of the Na+–K+ pumps available. Indeed, when the rate of Na+ extrusion was estimated over the first 30 s of recovery after 120 Hz stimulation, it was not significantly different from the theoretical maximum capacity, corresponding to a 22-fold increase in Na+–K+ pump activity.

Figure 5.

Na+ efflux in resting and electrically stimulated soleus muscles contracting without exerting external force

The Na+ extrusion in stimulated muscles was estimated from the net changes in Na+i content during the first 50 s ( inline image) or the first 30 s (•) of recovery from 10 s of stimulation at 60, 90 or 120 Hz. □, rate of ouabain-suppressible Na+ efflux in resting muscles. Each point represents a mean value ±s.e.m. of 12–33 muscles. Other details as in Fig. 4.

Mechanisms for the excitation-induced activation of the Na+–K+ pump

Increased Na+i is a potent stimulus for the Na+–K+ pump and the excitation-induced increase in Na+i probably contributes to the increase in Na+–K+ pump activity after electrical stimulation. However, as is evident from Fig. 1, the net extrusion of Na+ following stimulation may continue for several minutes after Na+i content is reduced to the level in resting control muscles. Measurements performed 10 min after 30 s stimulation at 60 Hz (corresponding to the time of maximal undershoot in Na+i content in Fig. 1) indicated that electrical stimulation was without effect on intracellular water content (see Methods for details). Therefore, the changes in Na+i content are reflected in similar changes in [Na+]i. This indicates that other mechanisms, independent of elevated Na+i, contribute to the excitation-induced increase in Na+–K+ pump activity following electrical stimulation. To evaluate this, the rate of net Na+ extrusion during the first 50 s of recovery after electrical stimulation (data from Figs 4 and 5) was related to the mean [Na+]i in the same period in Fig. 6. It appears that in general the activity of the Na+–K+ pump increased with increasing [Na+]i with a tendency for the effect to plateau at high [Na+]i. Additionally, the relation between [Na+h and Na+–K+ pump activity was shifted considerably upwards in stimulated muscles compared with resting muscles. In particular, it should be noted that following 10 s of stimulation at 60 Hz in muscles contracting isometrically, [Na+i was reduced from 28.4 ± 0.5 to 24.8 ± 0.7 mmol (1 cell water)−1 within 50 s. Thus, although the mean [Na+]i was identical to the 26.6 ± 0.9 mmol (1 cell water)−1 observed in resting muscles, a 5-fold increase in Na+–K+ pump activity was observed (Fig. 6). Taken together, these data strongly indicate that a substantial part of the excitation-induced activation of the Na+–K+ pump is independent of an increase in [Na+]i.

Figure 6.

Relation between Na+ efflux and mean [Na+]i in resting and electrically stimulated soleus muscles

Data from the experiments depicted in Figs 4 and 5. [Na+], was calculated as the mean of the [Na+]i, at the cessation of stimulation and after 50 s of recovery. ○, resting control muscles; •, muscles contracting isometrically; ▪, muscles contracting without exerting external force. Each point represents a mean value ±s.e.m. of 10–33 muscles. The curve was fitted by eye.

To examine the mechanism behind this Na+-independent activation of the Na+–K+ pump various treatments were tested for their effect on the undershoot in Na+i content measured 10 min after 30 s stimulation at 60 Hz (Table 2). Except for saxitoxin and dantrolene, which reduced contractile force by 95 ± 1 and 42 ± 5%, respectively, (n= 4 in both cases) none of the substances listed in Table 2 affected maximum force production during 60 Hz stimulation with pulses of 1 ms duration (data not shown). In denervated muscles the maximal contractile force at 60 Hz was reduced to 7.2 ± 1.1 N (g wet wt)−1 compared with 17.0 ± 1.2 N (g wet wt)−1 in control muscles (n= 4 in both cases).

Table 2. Effect of various treatments on intracellular Na+ content in resting soleus muscles and 10 min after 30 s of 60 Hz electrical stimulation using pulses of 1.0 or 0.02 ms duration
  Na+i 
 Pulse duration(ms)Resting(μmol (g wet wt)−1)Stimulated(μmol (g wet wt)−1) P
  1. The muscles were mounted at optimal resting length and contracted isometrically. The indicated substances were added 30 min before electrical stimulation corresponding to 40.5 min before the determination of Na+i content. P values indicate significance of difference between resting and stimulated muscles treated the same way except for electrical stimulation. Significantly different from resting control muscles, *P < 0.01. Values are means ±s.e.m. with the number of observations in parentheses. n. s., not significant at the 5% level.

Control1.015.5 ± 0.3 (36)11.2 ± 0.3 (37)< 0.001
 0.0215.3 ± 0.6 (4)12.9 ± 0.4 (4)< 0.05
Tubocurarine (10−5 M)1.015.5 ± 0.9 (4)10.7 ± 1.2 (4)< 0.02
 0.0214.6 ± 0.6 (4)15.9 ± 0.4 (4)n. s.
Saxitoxin (10−7 M)1.012.4 ± 0.1* (3)12.7 ± 0.0 (8)n. s.
Dantrolene (2 × 10−5 M)1.014.9 ± 1.0 (6)10.5 ± 0.7 (6)< 0.02
L-NAME (10−5 M)1.015.7 ± 0.7 (5)10.7 ± 0.4 (6)< 0.01
Propranolol (10−6 M)1.016.6 ± 0.5 (10)11.9 ± 0.6 (10)< 0.001
Salbutamol (10−6 M)1.08.8 ± 0.6*(12)8.2 ± 0.7 (12)n. s.
CGRP (10−7 M)1.09.6 ± 0.6* (8)8.9 ± 1.2 (8)n. s.
Denervated for 7 days1.012.7 ± 0.5* (9)11.4 ± 0.7 (9)n. s.

As shown in Table 2, the undershoot in Na+i could be elicited both by direct stimulation of the muscles using 1 ms pulses and by indirect stimulation through the motor nerve using pulses of 0.02 ms duration. Pre-incubation for 30 min with tubocurarine (10−5 M) was without effect in preparations stimulated by pulses of 1 ms duration, whereas in preparations stimulated using pulses of 0.02 ms duration it completely abolished the undershoot in Na+i during recovery. Pre-incubation of muscles with saxitoxin led to a small but significant reduction in Na+i content of resting muscles (Table 2) presumably by blocking the passive Na+ influx through Na+ channels. At the same time it completely prevented the undershoot in Na+i following electrical stimulation (Table 2). These data strongly indicate that the increase in Na+–K+ pump activity was not caused by the exposure of the muscles to the electrical field per se, but depended on the opening of Na+ channels. To examine the possible role of increased free intracellular Ca2+ for the excitation-induced undershoot in Na+i content, muscles were pre-incubated for 30 min with 2 × 10−5m dantrolene. This reduced tetanic force by 42% but had no effect on the undershoot in Na+i content (Table 2).

Studies on rabbit aorta has implicated a role for endothelium-derived nitric oxide in the activation of the Na+–K+ pump in smooth muscle (Gupta, Sussman, McArthur, Tornheim, Cohen & Ruderman, 1992; Gupta, McArthur, Grady & Ruderman, 1994). We therefore incubated muscles with L-NAME, an inhibitor of nitric oxide synthase, to test if nitric oxide was involved in the excitation-induced activation of the Na+–K+ pump in skeletal muscle. Pre-incubation with L-NAME was, however, without effect on Na+i– both in resting and stimulated muscles, indicating that the undershoot was not related to the formation of endogenous nitric oxide (Table 2). Noradrenaline and CGRP have been shown to lower Na+i of resting skeletal muscle by stimulating the Na+–K+ pump, presumably via an increase in the Na+ affinity of the Na+–K+ pump (for review see Clausen, 1986). It is conceivable, therefore, that the undershoot in Na+i after excitation was caused by an excitation-induced release of these hormones from endogenous stores of the muscle preparation. However, pre-incubation of muscles with propranolol at a concentration (10−6 m) shown to suppress the stimulating effect of noradrenaline on the Na+–K+ pump (Clausen & Flatman, 1977), did not prevent the undershoot in Na+i (Table 2). On the other hand, when the muscles were maximally stimulated with salbutamol (a β-agonist), resulting in lowered Na+i, electrical stimulation failed to induce an additional reduction in Na+i content (Table 2). This phenomenon was also observed when Na+i was reduced prior to electrical stimulation by pre-incubation with CGRP. Moreover, the excitation-induced undershoot in Na+i was absent in muscles denervated for 7 days prior to the experiment (Table 2). This procedure has been shown to deplete the nerve endings of skeletal muscle for CGRP (Kashihara, Sakaguchi & Kuno, 1989) and the lack of an undershoot in these muscles may, therefore, indicate a role for CGRP in this phenomenon.

Discussion

Na+–K+ pump activity following excitation

The present study demonstrates that in rat skeletal muscle excitation is a potent and long-lasting stimulus for the Na+–K+ pump. Based on the net change in Na+i content in the first 50 s following excitation, stimulation at 60–120 Hz resulted in a 5- to 18-fold increase in Na+–K+ pump activity. As illustrated in Fig. 1, a considerable reduction in Na+i took place during the 50 s interval used to determine the rate of net Na+ extrusion. This will reduce the stimulating effect of [Na+]i on the Na+–K+ pump, and the transport rates obtained are therefore likely to underestimate the Na+–K+ pump activity in the first few seconds after cessation of stimulation. In keeping with this, the transport activity following 120 Hz stimulation of muscles contracting without exerting external force was 40% larger when estimated from the net change in Na+i content in the first 30 s after stimulation than when estimated from the net change in the first 50 s after stimulation (Fig. 5). Taken together, the data indicate that during contractions involving a high frequency of action potentials, a large proportion of the Na+–K+ pump capacity present in the muscles is utilized.

Increased [Na+]i is in most tissues believed to be the most important stimulus for the Na+–K+ pump. In skeletal muscle, high frequency excitation is associated with an accumulation of Na+i (Table 1) and part of the rapid excitation-induced stimulation of the Na+–K+ pump is, therefore, probably related to increased [Na+]i. Indeed, the relation between [Na+]i and initial net Na+ extrusion following stimulation (Fig. 6) suggests that the differences between Na+–K+ pump activity in muscles stimulated for 10–30s at 60–120 Hz were related to a difference in the excitation-induced increase in [Na+]i. The effect of the Na+i accumulation on the Na+–K+ pump may be augmented by compartmentation of Na+i, which has been suggested to cause a preferential increase in subsarcolemmal [Na+] during excitation (Semb & Sejersted, 1996). Two observations indicate, however, that increased [Na+]i is not the only cause of increased Na+–K+ pump activity in contracting muscle. Firstly, in isometrically contracting muscles 60 Hz stimulation increased the activity of the Na+–K+ pump 5-fold within 10 s, although mean [Na]i was similar to the level in resting control muscles (Fig. 6). Secondly, following stimulation the active re-extrusion of Na+i continued even after Na+i content was reduced below the resting level, resulting in a considerable long-lasting undershoot in Na+i. Thus, following excitation the activity of the Na+–K+ pump can be increased even though Na+i concentration is at or below the resting level.

Although the mechanism behind the Na+-independent stimulation of the Na+–K+ pump cannot readily be identified a number of possibilities can be excluded. Since it was blocked by saxitoxin (Table 2) it was probably elicited by the generation of action potentials in the muscle preparation. Due to K+ lost from the muscle fibres during the action potentials, excitation is associated with an increase in extracellular [K+]. Single-cell studies on cardio-myocytes have, however, shown that the K½ of the Na+–K+ pump for extracellular K+ is around 1–1.5 mm (Cohen, Datyner, Gintant, Mulrine & Pennefather, 1987; Nakao & Gadsby, 1989). Therefore, an increase in extracellular [K+] to above the 5–93 μm present in the incubation buffer is unlikely to contribute to the increase in Na+–K+ pump activity during recovery from excitation.

Dantrolene has been shown to reduce the release of Ca2+ from the sarcoplasmic reticulum in response to excitation (Hainaut & Desmedt, 1974). In the present study, free intracellular Ca2+ was not measured, but pre-incubation of muscles with 2 × 10−5 M dantrolene reduced tetanic force by 42% indicating a substantial reduction in free intracellular Ca2+ during electrical stimulation. Dantrolene was, however, without effect on the excitation-induced undershoot in Na+i, indicating that there was no simple relation between the activation of the Na+–K+ pump and the increase in free intracellular Ca2+ during excitation.

Several hormones have been demonstrated to stimulate active Na+–K+ transport in resting muscle without increasing [Na+]i, presumably via an increase in the Na+ affinity of the Na+–K+ pump (Clausen, 1986), and a similar change in the transport characteristics of the Na+–K+ pump could account for the excitation-induced undershoot in Na+i. Experiments with L-NAME and propranolol indicate that the activation of the Na+–K+ pump cannot be ascribed to a release of NO or catecholamines from endogenous stores in the muscle preparation (Table 2). CGBP is known to be released from nerve endings of skeletal muscle during intense stimulation (Uchida et al. 1990; Sakaguchi, Inaishi, Kashira & Kuno, 1991) and this peptide might, therefore, contribute to the increase in active Na+–K+ transport in electrically stimulated muscles. This possibility tallies with the absence of an excitation-induced undershoot in Na+i in muscles from denervated rats (Table 2), where the release of CGRP during electrical stimulation is reduced (Kashihara et al. 1989). On the other hand, in muscles stimulated through the nerve the inhibition of signal transduction in the motor endplate by tubocurarine completely abolished the undershoot in Na+i after excitation. This indicates that the activation of the Na+–K+ pump is not related to an excitation-induced release of CGRP from motor nerves but caused by mechanisms distal to the motor endplate. Although the excitation-induced activation of the Na+–K+ pump cannot readily be related to the release of endogenous catecholamines or CGRP from the muscles preparation, it is interesting that electrical stimulation induced no further reduction in Na+i in muscles pre-incubated with supra-maximal doses of either salbutamol or CGRP (Table 2). This suggests that some of the steps in the regulatory pathway involved in the excitation-induced activation of the Na+–K+ pump are in common with the regulatory pathway used by catecholamines and CGRP. Several studies have indicated that the phosphorylation of a regulatory site on the Na+–K+ pump could influence its transport characteristics (Vasilets & Schwarz, 1993) and such phosphorylation may, therefore, be a final common step through which hormones and excitation activate the Na+–K+ pump without an increase in [Na+i.

Excitation-induced net changes in intracellular Na+

In soleus muscles contracting without exerting external force, 10s of stimulation at 60 or 120 Hz increased Na+i content by 7.3 and 12.7 μmol(g wet wt)−1, respectively (Table 1), corresponding to a net Na+ influx of around 11 nmol (g wet wt)−1 per action potential. In experiments with 22Na a Na+ influx of 9.5 nmol (g wet wt)−1 per action potential was observed. These values are in good agreement with earlier determinations of 22Na influx and 42K efflux during 2–5 Hz stimulation in rat muscles contracting without exerting external force, where values of around 7–11 nmol (g wet wt)−1 per action potential were found (Creese, Hashish & Scholes, 1958; Everts & Clausen, 1988). In muscles contracting isometrically the increase in Na+i during 60 or 120 Hz stimulation was between 18% (in vitro) and 60% (in situ) of that observed in muscles contracting without exerting external force (Table 1). This difference in Na+ influx during excitation was confirmed in experiments where the influx of 22Na was determined in isolated muscles. Since activation of the Na+–K+ pump in general was lower in muscles contracting isometrically than in muscles contracting without exerting external force, the difference in Na+ accumulation during contractions cannot be related to differences in Na+–K+ pump activity but rather reflects a marked difference in the passive Na+ influx associated with the action potentials. The underlying mechanism is at present unknown. One possibility might be that Na+ influx was related to the passive or active force of the muscles. However, in isometrically suspended muscles at rest the Na+i content was not affected by an increase in resting tension. Likewise, the accumulation of Na+i during 60 Hz stimulation was not significantly affected by an increase in the resting tension although this reduced the active force exerted by 41%. Thus, the effect of the mode of contraction on the excitation-induced influx of Na+ cannot readily be related to differences in the force exerted by the muscles.

Physiological significance of Na+–K+ pump stimulation during contractile work

The physiological implications of a rapid activation of the Na+–K+ pump in contracting muscle are 2-fold. Firstly, it improves the maintenance of the chemical gradients for Na+ and K+ during contractions. This effect is illustrated by the observation that partial inhibition of the Na+–K+ pumps by ouabain increases the accumulation of Na+i in rat soleus during 60 Hz electrical stimulation (Nielsen & Clausen, 1996c). Moreover, during 2 Hz stimulation of isolated rat soleus Na+i can be maintained at the resting level (Everts & Clausen, 1992, 1994) or even reduced (this study) despite the increased passive fluxes of Na+ and K+ during action potentials. Secondly, as shown by Hicks & McComas (1989) the activation of the Na+–K+ pump increases its electrogenic contribution to the membrane potential. In rat muscles, intermittent 20 Hz electrical stimulation was followed by an around 10 mV hyperpolarization that after cessation of stimulation showed a time course similar to the undershoot in Na+i (Fig. 1), i.e. the hyperpolarization was maximal 9 min after the cessation of stimulation and still present after 15 min. The hyperpolarization was blocked by ouabain and was suggested to result from increased Na+–K+ pump activity. The same mechanism was proposed to cause the potentiation of the M-wave during intermittent voluntary contractions in human subjects (Hicks, Fenton, Garner & McComas, 1989). Several studies indicate that both the increase in the electrogenic contribution of the Na+–K+ pump and in the active Na+–K+ transport serve to limit the development of contractile failure arising from reduced excitability in working muscles (for review see Nielsen & Overgaard (1996). In addition, because the Na+-independent activation of the Na+–K+ pump lasts for several minutes after the cessation of contraction, it may prime the muscle for subsequent contractions making the regulation of active Na+–K+ transport more efficient.

In conclusion, excitation of rat skeletal muscle leads to a rapid activation of the Na+–K+ pump that can only be partly explained by an increase in [Na+]i. This activation probably protects muscles from contractile failure arising from depolarization and run-down of the chemical gradients for Na+ and K+ during contractions. The Na+–independent activation of the Na+–K+ pump may act as a feed-forward mechanism that in contracting muscle allows the activity of the Na+–K+ pump to increase before the chemical gradients for Na+ and K+ are severely compromised. This mechanism may be of particular importance during contractile work of lower intensity where Na+i can be maintained at the resting level or even reduced. During intensive exercise, involving a high frequency of action potentials, the passive Na+–K+ fluxes exceed the active transport of the two ions and in this case increased [Na+]i constitutes an additional early stimulus for increasing Na+–K+ pump activity, which then may reach a level close to the theoretical maximum transport capacity.

Acknowledgements

This study was supported by the Danish Biomembrane Research Center, The Danish Medical Research Council (j.nr. 12–1336), the Ib Henriksen Fond and the Novo Nordisk Foundation. The technical assistance of Marianne Stürup-Johansen, Ebba de Neergaard, Tove Lindahl Andersen and Ann-Charlotte Andersen is gratefully acknowledged.

Ancillary