INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SKELETAL MUSCLE FATIGUE IS a complex phenomenon due to the large number of fatigue-inducing factors acting at multiple cellular sites (10). Although impairment of actomyosin cross-bridge function in skeletal muscle fatigue has been well characterized, a failure to maximally activate the contractile proteins may also contribute to fatigue. Eberstein and Sandow (9) suggested that sarcoplasmic reticulum (SR) Ca⁺ release might be reduced in fatigued muscle. Confirmation of this hypothesis came in a series of papers by Allen and co-workers (1, 24, 37) that used absorptive and fluorescent indicators to measure intracellular Ca⁺ during fatiguing stimulation of skeletal muscle fibers. However, to date, little is known about the mechanism(s) by which SR Ca⁺ release is impaired in fatigued muscle cells.

Skeletal muscle excitation-contraction coupling begins with the spread of the action potential over the sarcolemma and into the transverse tubules (t tubules). Depolarization of the t-tubular membrane is thought to cause the reorientation of the highly charged S4 transmembrane domains of the L-type Ca²⁺ channel, and this intramembrane charge movement triggers, by a poorly understood mechanism, the opening of the Ca²⁺ release channels of the SR (27, 31). Therefore, an impairment of voltage sensor function of the channel could lead to reduced SR Ca²⁺ release and may contribute to muscle fatigue.

Among the well-documented alterations in cellular homeostasis that occur in fatigued skeletal muscle are a decrease in intracellular pH (36, 38) and depolarization of the membrane potential (3, 14, 28). A depolarization of the resting membrane potential of up to 15 mV is commonly seen as a result of repetitive stimulation of skeletal muscle. Although difficult to measure, the depolarization in the t tubules may be more extreme than that seen at the surface membrane. Because of their size and the tortuous path they follow as they penetrate the muscle fiber, diffusion into and out of the t tubules is relatively slow. Thus alterations in the ionic composition of the t-tubular lumen as the result of repeated stimulation may be exaggerated compared with that of the interstitial space. Although direct measurement of the ionic composition of the tubular lumen has not been reported, it has been suggested that Na⁺ is depleted (4, 11), whereas K⁺ (2, 20) and Ca²⁺ (5, 16) accumulate. Depletion of Na⁺ would decrease the height of the action potential overshoot and slow the rate of depolarization. K⁺ accumulation would slow the repolarization rate and, along with Ca²⁺ accumulation, would depolarize the resting membrane potential. This depolarization may be sufficiently large to alter the function of the voltage sensor.

The isoelectric point of the amino acid histidine is pH 7.6; therefore, it is assumed to be partially charged at physiological pH values (25). Changes in pH, therefore, may alter the protonation state and charge of this amino acid. The L-type calcium channel has multiple histidines in the putative cytoplasmic domains. This raises the possibility that the function of the t tubular voltage sensor may be impaired by changes in intracellular pH in the physiological range, in particular the decrease in pH associated with skeletal muscle fatigue. In addition, the voltage sensor is inactivated by prolonged depolarization (30), and this inactivation may possibly be exaggerated by low pH. Finally, the voltage sensor has a high-affinity Ca²⁺-binding site on the extracellular aspect of the molecule, which must be saturated to maintain the sensor in the fully active state (6). Interestingly, Ca²⁺ concentration above that expected to saturate the Ca²⁺-binding site shifts the voltage dependence of contractile inactivation to more positive potentials (34). Therefore, elevated extracellular Ca²⁺ may mitigate some of the depolarization-induced inactivation of the voltage sensor.

We have examined the effects of decreased intracellular pH and membrane depolarization on the voltage sensor of excitation-contraction coupling in single frog skeletal muscle fibers mounted in a double Vaseline gap voltage clamp. In addition, we examined the interaction of elevated extracellular Ca²⁺ and membrane depolarization on the charge movement. Our results suggest that decreased pH, similar to that seen in extreme skeletal muscle fatigue, does not alter the voltage dependence of the t-tubular charge movement. However, elevated extracellular Ca²⁺ partially mitigated the depolarization-induced inactivation of the voltage sensor.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Frogs (Rana pipiens) were obtained from Kons Scientific (Germantown, WI) and housed in a large aquarium with continuously filtered water at room temperature (22-24°C). Frogs were fed live crickets every other day. The muscle fibers were dissected and mounted as described by Kovacs et al. (22). The semitendinosus muscles were removed from double-pithed frogs and dissected in Ringer solution (in mM: 115 NaCl, 2.5 KCl, 1.8 CaCl₂, 10.0 HEPES, pH 7.0) to a point at which small bundles of fibers were isolated for the length of the muscle. The solution was then changed to a high-K⁺ relaxing solution (in mM: 120 K-glutamate, 2 MgCl₂, 5 HEPES, 0.1 EGTA, pH 7.0), and a long segment of a single fiber was dissected and placed in a double Vaseline gap chamber. The chamber was milled from Lucite and consisted of three pools separated by two walls ~300 µm wide. The length of the middle pool was ~700 µm. A small groove in the top surface of each wall connected adjacent pools. For fiber mounting, the two grooves were partially filled with Vaseline, and the chambers were filled with excess relaxing solution so that the walls separating the pools were covered with solution. The fiber segment was mounted so as to run from one end pool to the other, spanning the length of the middle pool. The ends of the fiber segment were clamped to movable blocks in the end pools. Sliding the blocks away from the middle pool allowed the sarcomere length to be set to 3.0 µm. The space around the fiber in the grooves was filled with Vaseline, and the upper surface of the partition walls was covered with Vaseline. A Lucite covering piece was placed on each wall so that its edge was even with the side of the wall facing the middle pool. After the seals were in place, the ends of the fiber were permeabilized by exposure to 0.005% saponin in relaxing solution for 1 min. The solution in the end chambers was then replaced, after rinsing, with the experimental internal solution [in mM: 105 glutamic acid, 105 CsOH, 0.4 Ca(OH)₂, 0.5 Mg(OH)₂, 10.0 HEPES, 8.44 EGTA, 5.0 MgATP, 5.0 creatine phosphate, 5.0 glucose, pH 7.0 or 6.2]. The relaxing solution in the middle chamber was then replaced with the experimental external solution [in mM: 132.0 tetraethylammonium hydroxide, 132 methanesulfonic acid, 2.0 Ca(OH)₂, 5.0 trizma base (Tris), 1.0 3,4-diaminopyridine, 10³ tetrodotoxin, 5.0 CoCl₂, pH 7.0]. The experimental solutions were based on those described by Brum and Rios (7) and Kovacs et al. (21) and were designed to eliminate ionic currents. The chamber was then mounted on the stage of a compound Olympus microscope. Temperature was maintained at 12°C by a Cambion temperature controller via a Peltier temperature device built into the microscope stage.

The methods of voltage clamping, pulse generation, and data acquisition have been described previously (7). The technique used to measure the t-tubular charge movement can best be described in terms of a specific model of membrane currents. The total membrane current can be expressed as a sum of the terms
where I_T is the total membrane current, Q is the t-tubular charge movement, C is the capacitive current, I_I is the sum of all the time-dependent ionic currents, and I_L is the sum of all the time-independent ionic currents. Experiments were designed to either eliminate or subtract the ionic and capacitive currents while leaving Q. I_I was eliminated by removing permeant ions and including ion channel blockers in the experimental solutions. Membrane capacitance was determined by use of +40-mV voltage clamp steps from 0 mV. In this voltage range, membrane capacitance is constant, and the currents on step changes in voltage are proportional to the amplitude of the voltage step. Capacitive currents were measured before and after each series of test pulses at each pH or Ca²⁺ concentration, with ~15 min separating capacitance measurements. The average of the two measurements was used as for the subtraction of the linear capacitance. If the two measurements differed by >10%, the experiment was discarded. The capacitive current was then scaled to the amplitude of the test pulse and subtracted from the currents obtained from the test pulse. I_L was removed by fitting a straight-sloping baseline to the last portion of the difference record. The baseline was then subtracted from the difference record (18). The amount of charge moved by each test pulse was determined by integrating the area of the current associated with the initiation (on) and termination (off) of the test pulse. For small-amplitude test pulses, the on and off charges were generally equal. However, with large-amplitude test pulses, the off charge often exceeded the on charge. This was attributed to a residual ionic current. Therefore, only the on charge was analyzed.

Experiments were performed from holding potentials of 90 and 60 mV. These potentials represent the normal resting membrane potential and the resting potential seen in extreme fatigue, respectively. Test pulses were performed in 10-mV increments from the holding potential up to +20 mV, in a random order. After completion of the series of test pulses at both holding potentials at one pH, the fiber was taken out of the voltage clamp and removed from the microscope stage, and the internal solution was changed to the second pH. After 30 min at room temperature to allow for diffusion of the new solution into the fiber, the Vaseline gap chamber was mounted on the microscope stage and cooled to 12°C, the fiber was again placed under voltage clamp control, the linear capacitance was determined, and the holding potential was set to either 90 or 60 mV and held for 5 min. The series of test pulses was then repeated at both holding potentials. Diffusion of protons into and out of cut fibers is relatively slow (20-40 min in our internal solution on the basis of the considerations of Irving et al., Ref. 19); therefore, 45-50 min elapsed between solution changes and the commencement of data collection (30 min at room temperature, 10-15 min to cool Vaseline gap chamber to 12°C, and 5 min at the holding potential). In the series of experiments examining interaction of membrane depolarization and elevated extracellular Ca²⁺, test pulses to 0 mV were performed from holding potentials of 90, 75, and 60 mV. After a trial in either 2 mM (approximately resting extracellular Ca²⁺) or 10 mM (a t-tubular Ca²⁺ concentration suggested to occur in fatigued muscle; Ref. 16) extracellular Ca²⁺, the fiber was taken out of the voltage clamp, the extracellular solution was changed, and the fiber was returned to the voltage clamp. The pH internal solution in this set of experiments was 7.0.

Analysis and statistics. The dependence of the amount of charge moved (Q) on the test voltage (V) was well described by a two-state Boltzmann distribution

(1)

where Q_max is maximum amount of charge moved in nC/µF; is midpoint voltage, the voltage at which 50% of Q_max moves (in mV); and k is the steepness factor relating Q to V (in mV). Equation 1 was fitted to the data for each pH and holding potential by using a commercially available computer software program (SigmaPlot, Jandel Scientific). Carrying out experiments at both holding potentials and both pH 7.0 and 6.2 on the same fibers allowed the use of paired t-tests to test for differences in the parameters of Eq. 1. The effects of 2 and 10 mM Ca²⁺ on the amount of charge moved in response to voltage steps to 0 mV from holding potential of 90, 75, and 60 mV were tested by using analysis of variance with Student's t-tests as a post hoc analysis when indicated. Data are presented as means ± SE, with the level of significance set to P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Five fibers were studied at both pH 6.2 and 7.0 and at holding potentials of 90 and 60 mV. Typical charge movement transients are shown in Fig. 1. At a holding potential of 90 mV, the average Boltzmann parameters for all trials at pH 7.0 and 6.2, respectively, were Q_max, 27.9 ± 0.9 and 27.9 ± 1.4 nC/µF; , 45.1 ± 2.5 and 47.2 ± 3.1 mV; and k, 12.9 ± 0.5 and 12.1 ± 1.1 mV. There were no significant differences in any of these parameters. The curves fitted to these parameters are plotted in Fig. 2 and appear nearly identical.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Representative charge movement transients. Holding potential 90 mV. A and B: fiber 168E with an internal pH of 6.2 (A) and 7.0 (B). Test pulses are from 60 (bottom) to 30 mV (top) in 10-mV increments. C and D: fiber 183E with an internal pH of 6.2 (C) and 7.0 (D). Test pulses are from 60 (bottom) to 20 mV (top) in 10-mV increments. Records are the average of 5 sweeps.

View larger version (24K): [in this window] [in a new window]   Fig. 2.   Average charge-voltage relationship for fibers at holding potentials of 90 (circles) or 60 mV (squares) and pH 6.2 (solid symbols) or 7.0 (open symbols). Symbols are means ± SE. There were no significant differences in the Boltzmann parameters between the 2 pH trials at either holding potential. Inset: same data expressed as percentage of maximum charge moved to show more clearly the shift in voltage dependence.

At the holding potential of 60 mV, the Boltzmann parameters at pH 7.0 and 6.2, respectively, were Q_max, 17.0 ± 1.4 and 19.6 ± 1.4 nC/µF; , 34.1 ± 0.7 and 32.0 ± 2.3 mV; and k, 10.8 ± 0.4 and 11.0 ± 0.7 mV. There were no significant differences in any of these parameters. Curves fitted to these parameters are shown in Fig. 2. Although there appears to be a trend for Q_max to be larger at pH 6.2, the difference was not significant. These results suggest that a reduction in pH similar to that seen in severe muscle fatigue does not alter the amount or the voltage dependence of the t-tubular charge movement.

Although low pH does not appear to alter the t-tubular charge movement, as previously reported (30), prolonged depolarization moves the voltage sensor into an inactive state. Thus depolarizing the holding potential from 90 to 60 mV significantly decreased Q_max and shifted to more depolarized potentials at both pH 7.0 and 6.2 (compare Fig. 2, circles and squares). Depolarizing the holding potential reduced k at pH 7.0 but not at 6.2.

Bianchi and Narayan (5) reported that repeated activation of muscle led to an increased Ca²⁺ concentration in the t-tubular lumen. Therefore, we examined the effects of elevated extracellular Ca²⁺ on the t-tubular charge movement and the interaction between elevated Ca²⁺ and depolarization of the membrane potential. Increasing extracellular Ca²⁺ from 2 to 10 mM increased the amount of charge moved in response to test pulses in normally polarized (holding potential 90 mV), moderately depolarized (holding potential 75 mV), and severely depolarized fibers (holding potential 60 mV). Figure 3 illustrates the interaction between depolarization and elevated extracellular Ca²⁺. It is clear from Fig. 3A that depolarization of the holding potential from 90 to 60 mV reduced the magnitude of the charge movement. A comparison of the bottom trace of Fig. 3B (10 mM extracellular Ca²⁺) with the bottom trace of Fig. 3A (2 mM extracellular Ca²⁺) shows that the reduction in the amount of charge moved in depolarized fibers is lessened in 10 mM extracellular Ca²⁺. A similar preservation of charge by elevated extracellular Ca²⁺ can also be seen in an experiment using more moderate depolarization (Fig. 3, C and D). A small inward current can be seen after the on charge in a number of the records in Fig. 3, especially at elevated extracellular Ca²⁺. This may be due to incompletely blocked Ca²⁺ channels and may have led to a small underestimation of the amount of charge moved, particularly at high extracellular Ca²⁺.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Comparison of charge movement transients in 2 mM (A and C) and 10 mM (B and D) external Ca2+. Holding potentials are indicated. Test pulses were to 0 mV. A and B were recorded from the same fiber, as were C and D. Records are the average of 5 sweeps.

Figure 4 summarizes the effects of raising extracellular Ca²⁺ on the amount of charge moved by test pulses to 0 mV from 90, 75, and 60 mV. In 2 mM extracellular Ca²⁺, as the holding potential was depolarized, the amount of charge moved decreased from 24.9 ± 1.9 nC/µF at 90 mV to 19.4 ± 1.3 and 13.6 ± 2.6 nC/µF at 75 and 60 mV, respectively. However, in the presence of 10 mM extracellular Ca²⁺, significantly more charge was moved by test pulses from the three holding potentials (31.3 ± 2.0, 29.0 ± 3.1, and 18.3 ± 2.7 nC/µF at 90, 75, and 60 mV, respectively).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Summary of effects of external Ca2+ concentration on the amount of charge moved by test pulses to 0 mV from the indicated holding potentials. The amount of charge moved at each holding potential was significantly greater in 10 mM () than in 2 mM external Ca2+ (). *Significantly different from 2 mM Ca2+, P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

One of the major alterations in the internal milieu of skeletal muscle resulting from repeated muscle stimulation is a fall in intracellular pH, as low as pH 6.2 (10, 36). A pH decline of this magnitude has been shown to impair the function of isolated SR Ca²⁺ release channels (32), which may contribute to the decline in SR Ca²⁺ release and skeletal muscle fatigue. In contrast, in mechanically skinned fibers, the decline in SR Ca²⁺ release at low pH was attributed to a reduction in SR Ca²⁺ loading rather than to an effect on the depolarization-induced activation of the Ca²⁺ release channel (23). However, this study examined the effects of decreased pH on maximal stimulation of fully polarized fibers. The possibility remained that low pH might affect the voltage sensor when the resting membrane potential was depolarized or during submaximal stimulation.

Gyorke (12) evaluated the effects of repeated stimulation on excitation-contraction coupling in frog muscle fibers. Although there was a reduction in the size of the Ca²⁺ transient in these experiments, there were no alterations in the t-tubular charge movement. The primary conclusion was that alterations either in the link between the t-tubular voltage sensor and the SR Ca²⁺ release channel and/or in the channel itself was responsible for the decreased SR Ca²⁺ release. However, it is possible that intracellular pH did not decrease significantly in these fibers. The primary reason was that pH was buffered at pH 7.0. Additionally, in this preparation, the ends of the fiber are permeabilized to allow the diffusion of solution into the fiber. Because there is diffusion into and out of the fiber, it is not clear that the substrates and enzymes required for glycolysis remain within the fiber. In addition, ATP and creatine phosphate are provided in the internal solution; therefore, glycolytic energy production may be reduced. The present study focused directly on the effects of low pH on the voltage sensor of excitation-contraction coupling by buffering the internal pH either to 7.0, approximately resting pH, or to pH 6.2, a level seen in severely fatigued muscle fibers. No change in either the maximal amount of charge moved or the voltage dependence of the charge movement was observed. This was true whether the fibers were normally polarized (holding potential of 90 mV) or depolarized to an extent similar to that observed in severe muscle fatigue (holding potential of 60 mV). As can be seen in Fig. 1, A and D, the charge movement arising from test pulses in the range of 50 to 30 mV is composed of two components, an early beta and a delayed gamma charge. The origin of charge gamma remains controversial and has been attributed to the charge that triggers SR Ca²⁺ release (8) and to a charge movement resulting from SR Ca²⁺ release (31). One of our initial hypotheses was that the gamma charge might be preferentially affected by low intracellular pH. However, this does not appear to be true. Charge gamma appeared to be somewhat labile, often apparent during the first pH trial but not the second (regardless of whether the initial trial was pH 7.0 or 6.2). The charge initially contributing to the gamma component does not appear to have been lost, however, because the total amount of charge remained constant; only the kinetics changed. Therefore, we were unable to discern any effect of low intracellular pH on the charge movement, and thus it appears that the voltage sensor is relatively resistant to changes in internal pH within the physiological range.

Most studies examining the effects of intermittent stimulation on the resting membrane potential of skeletal muscle report a decrease of up to 15 mV (3, 14, 28). However, it is the potential of the t-tubular membrane that is important for the function of the voltage sensor, and it has been suggested that the t-tubular membrane may depolarize to a greater extent than the surface membrane. Due to a narrow diameter and tortuous path, the diffusion into and out of the t tubules is limited. It has been suggested that this diffusion limitation results in the accumulation of K⁺ (2, 20) and possibly Ca²⁺ (5, 16, 17) in the depths of the t tubules and leads to a greater depolarization than what occurs at the surface membrane.

Inactivation of the voltage sensor by prolonged depolarization has been well characterized (30, 31). Thus depolarization of the holding potential from 90 to 75 or 60 mV significantly reduced the amount of charge moved in response to test pulses to 0 mV. However, raising extracellular Ca²⁺ from 2 to 10 mM appears to oppose the depolarization-induced inactivation of the voltage sensors (Fig. 3). This is similar to the observation by Shlevin (35) in which raising extracellular Ca²⁺ from 1.8 to 25 mM increased the maximal charge moved. Our results also are in agreement with Schnier et al. (34), who observed that raising extracellular Ca²⁺ prolonged the plateau of tetanus activated by a voltage clamp. They attributed the prolonged tetanus to the delayed inactivation of the voltage sensor. However, the mechanism by which the delayed inactivation occurs is not clear.

There is a high-affinity Ca²⁺-binding site on the t- tubular luminal face of the voltage sensor (6, 35). Binding of Ca²⁺ to this site maintains the sensor in an activatable state. Nevertheless, it is difficult to explain the ability of 10 mM extracellular Ca²⁺ to mitigate the depolarization-induced reduction in charge movement via the high-affinity Ca²⁺-binding site. The site should be fully saturated at an extracellular Ca²⁺ concentration of 2 mM; therefore, raising extracellular Ca²⁺ to 10 mM should exert no effect via this Ca²⁺-binding site. However, it has been proposed that a second, low-affinity Ca²⁺-binding site that also maintains the voltage sensor in the active state may be present (34). An alternative explanation is the charge-screening effects of elevated extracellular Ca²⁺ (15). Ca²⁺ may interact with negative surface charges in the immediate vicinity of the voltage sensor and create a microenvironment in which the membrane potential differs from the average membrane potential. Regardless of the mechanism by which it occurs, the physiological significance remains that elevated extracellular Ca²⁺ opposes the depolarization-induced decrease in t-tubular charge movement and may help maintain in SR Ca²⁺ release in fatigued skeletal muscle.

Unfortunately, it is not clear what happens to the tubular Ca²⁺ concentration in response to repeated stimulation. The electrochemical gradient for Ca²⁺ favors Ca²⁺ influx, and the L-type Ca²⁺ current originates primarily from the t tubules (26). Indeed, the single preliminary description in the literature of direct measurements of t-tubular Ca²⁺ concentrations reported that tubule Ca²⁺ fell in single voltage-clamped fibers subjected to prolonged voltage clamp steps (13). However, the L-type current in skeletal muscle is very slow activating and contributes very little to the elevation of intracellular Ca²⁺ during a single twitch. During repeated stimulation, the activation rate of the L-type current increases, and the Ca²⁺ flux across the t tubule may become significant (26). However, Gyorke and Palade (13) found no change in t- tubular Ca²⁺ concentrations in whole muscles during prolonged low- or high-frequency stimulation. In addition to a pathway for Ca²⁺ flux from the t-tubular lumen into the myoplasm, a Ca²⁺/ATPase and Na⁺/Ca²⁺ exchange, which transports Ca²⁺ out of the myoplasm, is also present in the t tubule (33). If the activity of the Ca²⁺ transporters is sufficiently high, they may account for the t-tubular Ca²⁺ accumulation reported by Bianchi and Narayan (5). However, it is not known how the Ca²⁺ transporters and current interact to affect tubular Ca²⁺ concentrations in a fatigued fiber. If the t tubules of fatigued muscle are depleted of Ca²⁺, the voltage sensors would be moved toward an inactive state (29, 34). The inactivation of the sensors would be enhanced if Ca²⁺ depletion occurred simultaneously with depolarization. This would lead to less charge movement, a decreased SR Ca²⁺ release, and increased fatigue.

In summary, lowering intracellular pH to a level observed in extreme muscle fatigue had no effect on the amount or voltage dependence of the t-tubular charge movement, whereas depolarizing the holding potential decreased the maximal amount of charge moved and shifted the voltage dependence to more positive potentials. Raising extracellular Ca²⁺ from 2 to 10 mM increased the amount of intramembrane charge moved in response to test pulses from holding potentials near the resting membrane potential and from depolarized holding potentials. These results suggest that the decline in SR Ca²⁺ release observed in fatigued skeletal muscle fibers is not due to an effect of pH on the t-tubular voltage sensor. However, depolarization of the membrane potential similar to that seen in severe fatigue reduced the amount of charge moved and may decrease SR Ca²⁺ release. If the concentration of Ca²⁺ in the t tubules rises in fatigued muscle, it may act to stabilize the voltage sensor in the active state and help maintain SR Ca²⁺ release. However, it is not clear how tubular Ca²⁺ changes in fatigue. If Ca²⁺ falls, it would tend to move the voltage sensors into an inactive state and decrease SR Ca²⁺ release. An examination of the ionic composition of the t-tubular lumen of fatigued muscle is needed to distinguish between these possibilities.