INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

POSTCONTRACTILE DEPRESSION (PCD) has been characterized as a sustained reduction in muscle force after intermittent tetanic contractions (41). In this type of fatigue, as contrasted to low-frequency fatigue in which the depression in force is specific to frequencies <20 Hz (12), force is also depressed at high frequencies of stimulation (41). A number of intracellular sites may be mechanistically linked to PCD, causing failure of the sarcolemma and T-tubular membranes to conduct a repetitive action potential; an uncoupling of the signal linking the T-tubule with the Ca²⁺-release channel of the sarcoplasmic reticulum (SR); an impairment in Ca²⁺ cycling, both Ca²⁺ release and Ca²⁺ uptake; and an inability of the myofibrillar apparatus to translate a free Ca²⁺-activating signal in the cytosol, {cytosolic-free calcium [Ca²⁺]_f}, into an expected force response (15). One model that has proven useful for allowing insight into the role of specific sites is the single-fiber preparation (1, 25). With this preparation, it is possible using fluorescent probes to measure [Ca²⁺]_f during and after different stimulation protocols (1). The results obtained from experiments employing this model indicate that PCD is associated with a depression in activating [Ca²⁺]_f (25).

The depression in [Ca²⁺]_f observed during PCD, which would appear to be at least partly responsible for the delayed recovery in force, could be due to a direct disturbance in SR function per se or to some site more peripheral. Evidence presented to date has implicated primarily excitation-contraction (EC) coupling (13, 25, 38) and, specifically, the transmission of the excitation signal from the T-tubule to the Ca²⁺-release channel (37) as the possible mechanism responsible for the low [Ca²⁺]_f observed with stimulation during PCD.

The SR and, in particular, the Ca²⁺-sequestration properties of the SR, may be indirectly linked to the lower-activating [Ca²⁺]_f observed with PCD by affecting the processes involved in T-tubule-Ca²⁺-release channel activation. There is accumulating evidence to suggest that elevated [Ca²⁺]_f levels may impair signal transmission between these processes, resulting in a depression in Ca²⁺ release from the SR Ca²⁺-release channel (37). A depression in SR Ca²⁺-ATPase activity could conceivably prolong the Ca²⁺-transient during activation and/or result in an elevated resting [Ca²⁺]_f, both of which appear important in the failure of T-tubule-SR communication (37).

However, to implicate the SR, it must be demonstrated that changes occur in SR function, either Ca²⁺ release or Ca²⁺ uptake, in association with PCD. There is only limited evidence of this nature. In frog skinned muscle fibers, it has been found that there are intrinsic alterations in the SR Ca²⁺ uptake and Ca²⁺ release associated with activity-induced PCD (42). In humans, alterations in SR Ca²⁺ pump function during PCD are particularly inviting given the depressing effect that exercise in general has on Ca²⁺ uptake and Ca²⁺-ATPase activity measured under optimal conditions in vitro (17). Of the two studies that have examined SR function in humans in conjunction with PCD, one reported that maximal voluntary contractile force and Ca²⁺ uptake recovered substantially, but still remained depressed at 30 min (14), whereas the other study found that by 60 min of recovery, force was normalized, but Ca²⁺ uptake and Ca²⁺-ATPase remained depressed (5). However, these studies may not be comparable because of the fundamentally different exercise protocols (isometric vs. dynamic) employed.

It is conceivable, given the intensity and nature of the repetitive isometric contractions needed to induce PCD, that disturbances in metabolism, such as accumulation of by-products such as H⁺ and P_i, might be the cause of the depression in force observed (15, 44). Although disturbances in the intracellular milieu may explain the fatigue observed during the exercise protocol, they would not be expected to associate with PCD. Previous studies (25) found muscle energetic status to be normal during recovery after the exercise, whereas PCD remained depressed.

In this study, we hypothesized that the sustained PCD in human skeletal muscle induced by intermittent tetanic contraction performed voluntarily would be associated with depressions in Ca²⁺ uptake and in Ca²⁺-ATPase activity. Moreover, PCD and the impairment in Ca²⁺ sequestration would occur independent of changes in muscle energy state and metabolic by-product accumulation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. Eight healthy females, averaging 20.6 ± 0.75 yr and 60.8 ± 2.5 kg (means ± SE), participated in the study. All subjects were active but untrained and free from any prescribed drugs or related medical problems. This study was approved by the Office of Human Research at the University of Waterloo, and all subjects were fully informed of all experimental procedures before obtaining written consent.

Experimental protocol. To investigate the relationship between SR Ca²⁺ pump function and PCD, the subjects were asked to perform isometric knee extension contractions for 30 min at 60% maximal voluntary contraction (MVC) and 50% duty cycle with a contraction and relaxation schedule set at 5 s contraction/5 s relaxation. Before the fatiguing protocol, at 15 and 30 min of exercise and at 60 min of recovery, the mechanical properties of the quadriceps muscle were measured using both maximal voluntary and electrically induced contractions. To examine SR function and metabolic behavior, tissue samples were obtained at rest prior to exercise, immediately after 15 and 30 min of exercise, and at 60 min of recovery from the vastus lateralis muscle using the needle biopsy technique (4). Before exercise, tissue samples were obtained from the left leg in all subjects. During exercise and recovery, all samples were obtained from the right leg. The fatigue regimen and the mechanical measurements were obtained from the right leg. Tissue sampling sites and preparation, which was completed at least 30 min before the initial mechanical measurements, have been previously described by our group (19). Two biopsies were extracted from each site at each time point. The initial biopsy was quickly plunged into liquid N₂ and stored for later analysis of muscle metabolites. The second sample was used to prepare homogenates for Ca²⁺ pump measurements. During the exercise protocol, the tissue was extracted as fast as possible after cessation of the activity. No dietary supplements were provided during the exercise or during recovery.

Approximately 3 min after the initial contractile measurements, the subjects began the fatigue protocol. An oscilloscope display screen was clearly marked to indicate the target force (60% MVC) that each subject was required to produce during the repetitive activity. Every 5 min during the activity, a 10-s force record was used to verify that the subjects were meeting the target force. At 15 min, the activity was stopped for ~5 min to allow the biopsies to be performed and the muscle contractile characteristics to be measured.

Transcutaneous electrical muscle stimulation. The experimental setup for measuring muscle contractile characteristics was described previously (18). Briefly, for all force measurements, both voluntary and involuntary, the participant sat upright in a straight-backed chair with the knee at 90° to the thigh and with the arms folded across the chest. A strap was used to secure the hips and thigh to minimize any extraneous movement that could affect force production. Twitches and tetani were delivered to the quadriceps from a Grass model S48 stimulator. A 5-cm-wide plastic cuff placed around the right leg just proximal to the malleoli was tightly attached to a linear variable differential transducer (LVDT). The LVDT output was amplified by a Daytronic carrier preamplifier and recorded on a two-channel Hewlett Packard 7402A recorder. Positioning of the LVDT was such that an angle of pull at 180° was achieved for each participant. Calibration was performed before each test session with weights of known amounts. Two brass electrodes (13 × 8 cm) coated with warm electrode gel were used to deliver the electrical impulse to the quadriceps muscle. The ground electrode was placed centrally on the anterior aspect of the thigh just above the patella, whereas the active electrode was placed toward the hip on the belly of the vastus lateralis. Each electrode was secured firmly with rubber straps wrapped around the leg and over the top of the electrode to ensure good contact between the skin and electrode.

Twitches were evoked using a single supramaximal (~150 V) impulse of 50-µs duration while tetani at low (10 and 20 Hz) and high (50 and 100 Hz) frequencies were induced using a voltage that elicited ~60% MVC (in prefatigued muscle at 100 Hz) with a pulse duration of 50 µs and a train rate and duration of 1 s and 1 s. Tetanic force, regardless of frequency of stimulation, was taken as the peak force recorded. The average of two trials was used for all contractile properties. At each measurement point, a standardized protocol was employed that consisted of measuring the twitch, the tetany (in order of increasing frequency), and MVC. To assess MVC, 5-s contractions were performed. The average of two trials was used to represent MVC before the fatiguing protocol. However, only one trial was taken at all other time points during exercise and recovery so as not to induce any further fatigue.

One week before starting the experiment, participants were familiarized with all the electrical stimulation procedures. This session was also used to determine the voltage that elicited ~60% of the participant's MVC at 100 Hz.

Measurements of SR function. Measurements of maximal Ca²⁺-ATPase activity and maximal Ca²⁺ uptake were performed on muscle homogenates prepared using 11:1 (vol/wt) dilution of buffer containing (in mM) 200 sucrose, 40 L-histidine, 1 EDTA, 10 NaN₃, and 1 DTT (pH 7.8), using a hand-held glass-glass homogenizer (Kontes, Duall 20) (30). Small aliquots of homogenate were then quick frozen in liquid N₂ and stored at 80°C for later analysis of protein content and SR function. It has been shown that homogenates are stable to freezing and that Ca²⁺ uptake (33) and Ca²⁺-ATPase activity are minimally altered (10, 33).

Maximal Ca²⁺-ATPase activity was measured using the spectrophotometric assay developed by Simonides and van Hardeveld (36) and validated for human muscle by Ruell et al. (33) with minor modifications. The reaction buffer contained (in mM) 200 KCl, 20 HEPES, 15 MgCl₂, 1 EGTA, 10 NaN₃, 5 ATP, and 10 of phosphoenolpyruvate (PEP; pH 7.0). The pH was adjusted at 37°C. Just before starting the reaction, 18 U/ml lactate dehydrogenase (LDH), 18 U/ml pyruvate kinase (PK), 33 µl homogenate, 1 µM Ca²⁺ ionophore A23187 (Sigma C-7522), and 0.3 mM NADH were added to a cuvette containing 1 ml reaction buffer. Assays were performed in duplicate at 37°C and 340 nm (Shimadzu UV 160) with 1 mg wet wt tissue per assay. After the recording of baseline absorbance and fluorescence of NADH for ~1 min, the reaction was initiated by adding 8 µl of 100 mM CaCl₂ and monitored for 1-2 min. At the end of this period, four additional 1-µl additions of 100 mM CaCl₂ were made to ensure that the activity was maximal. The additions of Ca²⁺ were estimated to produce a [Ca²⁺]_f between 6.6 and 16.1 µM (35), the range that elicits maximal Ca²⁺-ATPase activity (36). Basal or Mg²⁺-ATPase activity was determined by adding 10 µl of 2 M CaCl₂ to produce a [Ca²⁺]_f of 17 mM, the level necessary to ensure full back inhibition of the Ca²⁺-ATPase enzyme activity (36). The SR Ca²⁺-stimulated ATPase activity is based on the difference between the total and basal ATPase activities.

Oxalate-supported Ca²⁺ uptake was measured using the Ca²⁺ fluorescent dye indo 1 according to methods of O'Brien and colleagues (30, 31) and as previously reported from our group (10). Fluorescence measurements were made on a spectrofluorometer (RatioMaster system, Photon Technology International) equipped with dual-emission monochromators. The measurement of [Ca²⁺]_f using the indo 1 procedure is based on the difference in the maximal emission wavelengths between the Ca²⁺-bound form of indo 1 and the Ca²⁺-free form. The emission maxima were 485 and 405 nm for Ca²⁺-free (G) and Ca²⁺-bound (F) indo 1, respectively. Photon counts per second were recorded simultaneously for both emission wavelengths. The Ca²⁺-independent (background) fluorescence was measured in the reaction medium (without indo 1) at each emission wavelength before starting the experiment. Background fluorescence was automatically corrected prior to starting each assay using the Felix software (Photon Technology International).

The reaction buffer contained 200 mM KCl, 20 mM HEPES, 10 mM NaN₃, 5 µM N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine, 5 mM oxalate, 15 mM MgCl₂ and 10 mM PEP (pH 7.0). Before the collection of emission spectra, 18 U/ml each of LDH and PK and 1.5 µM indo 1 were added to a cuvette containing 2 ml of reaction buffer. Immediately after data collection was initiated, 100 µl of homogenate was added to the cuvette. Shortly after the addition of homogenate, 5 mM ATP was added to initiate Ca²⁺ uptake. Contaminate levels of [Ca²⁺]_f were sufficient for the Ca²⁺-uptake measures. Consequently, addition of Ca²⁺ to the reaction buffer was not necessary.

As Ca²⁺ decreases because of active SR Ca²⁺ uptake, F decreases, G increases, and the ratio of F to G decreases. The ratio (R) is used to calculate [Ca²⁺]_f. With the use of Felix software, the ionized [Ca²⁺] was calculated by the following equation (22)

(1)

where K_d is the equilibrium constant for the interaction between Ca²⁺ and indo 1, R_min is the minimum value of R at addition of 250 µM EGTA, G_max is the maximum value of G at addition of 250 µM EGTA, G_min is the minimum value of G at addition of 1 mM CaCl₂, and R_max is the maximum value of R at addition of 1 mM CaCl₂. The K_d value for the Ca²⁺-dye complex is 250 nM (Grynkiewicz et al., Ref. 22). For all Ca²⁺ uptake trials, R_min and R_max were not determined until Ca²⁺ uptake had plateaued, which occurs at ~100 nM [Ca²⁺]_f.

Before the rate of Ca²⁺ uptake was calculated, the generated curve from equation 1, [Ca²⁺]_f versus time, was smoothed over 14 points using the Savitsky-Golay algorithm. The rate of Ca²⁺ uptake was then analyzed at 1.5 µM [Ca²⁺]_f. First, linear regression was done on a range of values between 1.4 and 1.6 µM [Ca²⁺]_f. The rate of Ca²⁺ uptake was then determined by differentiating the linear fit curve. Calcium uptake rates in homogenates of human muscle are maximal at this [Ca²⁺]_f (unpublished). These procedures represent a modification from those previously employed in our laboratory (20).

For both the Ca²⁺-ATPase activity and the Ca²⁺ uptake, protein was determined by the method of Lowry as modified by Schacterle and Pollock (34). On a given day, all samples for a given variable and for a given individual were analyzed in duplicate

Muscle metabolites. The metabolites measured included the high-energy phosphates (ATP, PCr), creatine (Cr), P_i, lactate, and glycogen (Glyc). The analyses were performed on freeze-dried tissue after extraction using fluorometric techniques identical to those used previously in our laboratory (21). Metabolite concentrations were corrected for total creatine (TCr) content using the average muscle TCr calculated from all biopsies for each individual. The experimental protocol was found not to alter TCr (P > 0.05). This procedure is designed to correct for contamination by blood, fat, and connective tissue. As with the SR measurements, all samples for a given subject were performed in duplicate during the same analysis session.

Data analysis. To determine the effect of the fatigue protocol on the muscle contractile and biochemical characteristics, a one-way analysis of variance with repeated measures was employed. Where significant differences were found, Duncan's post hoc tests were used to compare specific means. For all comparisons, the level of significance was set at P < 0.05. All data are expressed as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mechanical measurements. MVC force was depressed (P < 0.05) ~25% by 15 min of exercise (Table 1). No further changes in MVC occurred during the final 15 min of exercise. After 60 min of recovery, MVC force remained depressed (P < 0.05) by ~21%.

Reductions in involuntary force were also observed at both low (10 and 20 Hz) and high (50 and 100 Hz) frequencies of stimulation (Table 1) after 15 min of exercise. Further reductions (P < 0.05) in force occurred at each of the stimulation frequencies by 30 min of exercise. After 60 min of rest, force was not significantly different from that observed immediately after 30 min exercise. The fatigue that was observed was not specific to stimulation frequency, because absolute force was reduced to a similar extent regardless of frequency (Fig. 1). In contrast to tetanic force, twitch force (P_t) was not reduced at 15 min of exercise (Table 1). By 30 min of exercise, P_t was reduced (P < 0.05) ~23% and showed no recovery after 60 min of rest.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Force-frequency curves of quadriceps muscle at rest, immediately after isometric exercise (Exercise 30), and 60 min after exercise (Recovery 60). All values are means ± SE (n = 8). * Significantly different from rest (P < 0.05).

Muscle metabolites. After 15 min of isometric exercise, no changes were observed in any of the metabolic parameters examined (Table 2). Muscle ATP contents remained constant throughout exercise and recovery. By 30 min of exercise, PCr was depressed 23%. The increase in Cr and P_i (P < 0.05) was near stoichiometric to the decrease in PCr. All parameters were fully recovered by 60 min after exercise. Glycogen was also depressed (P < 0.05) after 30 min of exercise by ~24% and was not different from resting levels 60 min after the completion of exercise. Muscle lactate content was not significantly different from rest during either exercise or recovery.

SR function. The isometric exercise protocol led to reductions (P < 0.05) in both maximal Ca²⁺-ATPase activity and maximal Ca²⁺ uptake of ~50 and 44%, respectively (Fig. 2). The reductions in Ca²⁺-ATPase activity and Ca²⁺ uptake were fully manifested by 15 min of exercise, with no further change by 30 min of exercise. A 60-min recovery period after exercise failed to restore either Ca²⁺-ATPase activity or Ca²⁺ uptake (P > 0.05). The changes in Ca²⁺-Mg²⁺ ATPase (total) mirrored the changes in SR Ca²⁺-ATPase, whereas no changes occurred to Mg²⁺-ATPase (basal) either during exercise or recovery (Table 3). Alterations in Ca²⁺-ATPase activity with exercise paralleled changes in Ca²⁺ uptake so that coupling between the two was not altered throughout exercise or during recovery (Fig. 3).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Ca2+ uptake rates (measured at 1.5 µM [Ca2+]f) and maximal sarcoplasmic reticulum Ca2+-ATPase activity measured in vastus lateralis at rest (0E), 15 (15E,) and 30 (30E) min of exercise, and 60 (60R) min after isometric exercise. Values are means ± SE (n = 8). * Significantly different from rest (P < 0.05).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Coupling ratios between Ca2+ uptake and Ca2+-ATPase in vastus lateralis before and during postcontractile depression. Values are means ± SE, n = 8. Coupling ratio is defined as Ca2+-uptake/Ca2+-ATPase activity.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

To test the hypothesis, namely, that PCD would be associated with prolonged reductions in SR Ca²⁺-ATPase pump function, it was first necessary to demonstrate that the exercise regimen employed caused a persistent fatigue across the spectrum of activation levels. Using a protocol consisting of isometric contractions at 60% MVC and 50% duty cycle (5 s contraction/5 s relaxation) for 30 min, we were successful in achieving this goal. Regardless of whether the muscle was voluntarily activated (MVC) or induced to contract by electrical stimulation, the exercise-induced reduction in force was not reversed with 60 min of rest. As a consequence, we have been able to demonstrate the existence of PCD in human subjects after voluntary exercise. Unlike previous studies in humans in which PCD was manifested primarily at low frequencies of stimulation (12, 19), our fatigue protocol elicited PCD that was observed at both low and high frequencies of stimulation.

For the fatigue protocol that we employed to induce PCD, the absolute amount of force loss was comparable both during and after exercise across the different stimulation frequencies employed. The fact that both high- and low-frequency force was depressed equally probably reflects the severity of our exercise protocol (25). Although the percent force loss is more pronounced for the low frequencies of stimulation (10 and 20 Hz), particularly at the end of exercise and recovery, this must be interpreted in the context of the smaller absolute forces elicited at these frequencies of stimulation. The fact that the P_t did not display as great a force loss as that observed for the low frequencies, particularly at 15 min of exercise, is probably due to the existence of a residual amount of posttetanic potentiation (18). The twitches were measured after the repetitive contractions and before the other contractile measures. As fatigue becomes more profound, the effects of potentiation can become masked. For MVC, it is unclear to what extent PCD is affected by changes in neural drive and muscular factors. Because the force generated by MVC was well in excess of that observed during stimulation, the percent reduction would not be expected to be as great.

Our results lend support to the major hypothesis, namely, that during PCD, reductions in both maximal Ca²⁺ uptake and maximal Ca²⁺-ATPase activity would occur when measured in vitro. Our findings are consistent with a previous study in humans that showed that both MVC and Ca²⁺ uptake were reduced after brief (2.8 min) high-intensity dynamic exercise (14). However, because only an MVC was used to monitor fatigue, it is unclear if the deficit is peripheral. In our protocol, unlike the work of Gollnick et al. (14), neither Ca²⁺ uptake nor evoked force showed any recovery by 60 min postexercise. In addition, we also demonstrated that the reduction in Ca²⁺-uptake activity is accompanied by a reduction in Ca²⁺-ATPase activity.

A relatively recent study found a reduction in Ca²⁺ pump function, both Ca²⁺-ATPase and Ca²⁺ uptake, after prolonged cycle exercise (5). Interestingly, 60 min after exercise, maximal voluntary force and electrically induced force at low frequencies had completely recovered, whereas both homogenate Ca²⁺ uptake and Ca²⁺-ATPase activity remained depressed. Surprisingly, the cycle exercise failed to produce PCD at high frequencies (100 Hz). Collectively, our study in conjunction with the studies of Gollnick et al. (14) and Booth et al. (5) illustrates the role of the nature of the task in the recovery responses that are observed. With prolonged submaximal exercise, force recovers but Ca²⁺ pump function does not; with brief, maximal exercise, force and Ca²⁺ pump function both recover but not completely; and with sustained, isometric exercise, no recovery in force or Ca²⁺ pump function is observed. An explanation for the differing responses in recovery remains unclear. What is consistent, at least in humans, is that exercise, regardless of task characteristics, leads to a reduced ability of the SR to sequester Ca²⁺.

When SR function is measured in vitro, as in this study, standardized and supposedly optimal conditions are employed. These conditions are quite different from intracellular conditions that would be expected in vivo with exercise. Any exercise-induced changes in Ca²⁺-ATPase activity or Ca²⁺ uptake measured in vitro must be due to intrinsic changes such as structural alterations in either the Ca²⁺-ATPase protein, the phospholipid membrane of the SR, or both. On the basis of our exercise protocol, we can conclude that these structural alterations persist for at least 60 min after exercise. It would be expected that the structural abnormalities causing a reduced Ca²⁺-ATPase activity and Ca²⁺ uptake in vitro also existed in vivo and presumably affected SR function in vivo. In fact, SR Ca²⁺-ATPase function is probably reduced to a greater extent in vivo with exercise than observed in vitro because of additional local metabolic effects that would be present in vivo (1).

Allen et al. (1) proposed that SR pump function can remain depressed for at least 30 min postexercise in intact single fibers due to local metabolic factors. Accumulation of metabolic by-products, such as P_i, ADP, and H⁺, have been shown to adversely affect Ca²⁺ pump function in skinned fibers (44). However, other factors appear to be involved in PCD. We observed a depressed Ca²⁺-ATPase activity and Ca²⁺ uptake despite complete metabolic recovery after intermittent isometric exercise. Our results are in agreement with Williams (42), who observed a decrement in Ca²⁺ uptake in skinned fibers with fatigue independent of an altered intracellular environment. It is possible that prolonged exposure to one or more metabolites during the exercise caused a structural alteration in the SR that was not reversed 60 min after exercise. However, this appears unlikely because the metabolic stress measured at 15 and 30 min of exercise was unimpressive.

To date, the only structural disturbance that has been shown with repetitive activity is to the enzyme itself and specifically to the nucleotide-binding domain of the Ca²⁺-ATPase enzyme (26, 28). The Ca²⁺-ATPase enzyme consists of base piece, stalk, and headpiece components (6, 27). The large cytoplasmic extramembrane portion of the enzyme (headpiece) contains a phosphorylation domain, a nucleotide domain for the binding of MgATP, and a transduction domain or -strand domain (27). The competitive inhibitory effect of fluorescein isothiocyanate (FITC) bound to the Lys 515 residue on ATP binding indicates a close relationship between this domain and the nucleotide binding site (8). It would be reasonable to expect that a similar mechanism may explain the reduction in Ca²⁺-ATPase activity and Ca²⁺ uptake observed in this study. Studies with humans using fluorescent probes, such as FITC, to test for structural alterations of the Ca²⁺-ATPase enzyme with exercise would be problematic given the small amount of tissue obtained by biopsy and the need to obtain an enriched SR fraction. To the authors' knowledge, no study, regardless of species, has examined the role of the other specialized domains, in either the headpiece or stalk (Ca²⁺-binding domain), to repetitive exercise.

Other intracellular stimuli that could potentially alter the structure of the SR membrane or Ca²⁺-ATPase in exercise are Ca²⁺ and O₂-derived free radicals. Both of these are considered primary factors responsible for ischemia-reperfusion injury in skeletal muscle (32). Ischemia and reperfusion also lead to reduced Ca²⁺-ATPase activity in vitro in rats (20).

It has been shown that fatiguing stimulation in single mouse muscle fibers can lead to an increased resting intracellular [Ca²⁺]_f (39, 40). The effect that the isometric exercise protocol that we employed in this study had on resting intracellular [Ca²⁺]_f is unknown. However, because of the nature of the exercise (i.e., sustained 5-s contractions followed by 5 s of relaxation), the muscle cells would be subjected to repeated elevations in intracellular [Ca²⁺]_f throughout exercise, even if Ca²⁺ levels during contraction were reduced with fatigue.

Although the effects of raised intracellular [Ca²⁺]_f on SR Ca²⁺-ATPase activity and Ca²⁺ uptake have not been assessed directly, it has been proposed that Ca²⁺ may cause cell membrane damage by activating phospholipase A₂ (23). Calcium may also indirectly lead to protein damage due to its activation of the neutral protease calpain (3). Membrane damage would not apply to the reductions in Ca²⁺-ATPase activity observed in this study, however, because coupling ratios were normal and unchanged during PCD. We observed similar reductions during both exercise and recovery in Ca²⁺-ATPase activity and Ca²⁺ uptake after isometric exercise (see Fig. 3). This would suggest that reductions in Ca²⁺ uptake were due to a reduced Ca²⁺-ATPase turnover and likely not to problems with passive Ca²⁺ leakage through a damaged SR membrane or the Ca²⁺-release channel itself.

Free radicals are produced in exercise and may contribute to skeletal muscle fatigue and tissue damage (2, 11). Damage to the SR in exercise has been associated with the accumulation of free radicals (11). Not only is the SR a principal site of attack by free radicals (24), but it has been shown that when isolated SR is exposed to hydroxyl radicals in vitro, Ca²⁺-ATPase function is inhibited through direct attack on the ATP binding site (43). This mechanism would be consistent with the known effects of exercise on Ca²⁺-ATPase structure and function (26).

Our results suggest that disturbances in Ca²⁺ uptake can be implicated in PCD. Because PCD appears to be due, in large part, to reductions in activating [Ca²⁺]_f, it would appear that the mechanism must be explained by reductions in Ca²⁺ release. We used a similar exercise protocol with male participants, and found that the administration of caffeine immediately after exercise greatly attenuated postcontractile force depression (16). Caffeine is known to stimulate Ca²⁺ release via direct action on the Ca²⁺-release channel (29). Alternatively, the problem may not be with the Ca²⁺-release channel per se but with EC coupling (25, 38). Both hydrogen peroxide (7) and Ca²⁺ (9, 42) have been shown to impair Ca²⁺ release. It is thought that elevated levels of free radicals or Ca²⁺ are capable of damaging the Ca²⁺-release channel or other proteins of the EC coupling process to produce a reduction in Ca²⁺ release and contractile function (7, 42). It is possible that the same mechanism, namely an elevation in O₂ free radicals and/or intracellular [Ca²⁺], was responsible for the loss in SR Ca²⁺-sequestration function and force with 30 min of intermittent isometric exercise.

In general, the intracellular [Ca²⁺]_f at any given time represents a balance between Ca²⁺ release and Ca²⁺ uptake by the SR and the binding of Ca²⁺ by cytosolic proteins and other intracellular organelles. In one context, a reduction in Ca²⁺ uptake with fatigue may be viewed as a protective response. A reduced Ca²⁺ uptake may help to offset any reduction in Ca²⁺ release and thus help to preserve intracellular [Ca²⁺]_f and force. Alternately, a depression in Ca²⁺ uptake may reduce Ca²⁺ loading into the SR, resulting in less Ca²⁺ available for release (44). An additional, particularly inviting hypothesis is that the disturbance in Ca²⁺ pump function with exercise results in an inability to restore resting [Ca²⁺]_f to prefatigue levels. The increase in [Ca²⁺]_f could conceivably account for the structural abnormalities that occur to the Ca²⁺-release channel and/or the EC coupling mechanisms.

It should be noted that the calculation of the maximal Ca²⁺ uptake in vitro is based on the reduction in [Ca²⁺]_f in the homogenate. Because some of the Ca²⁺ is bound to various binding sites in the homogenate, and particularly the oxalate, the release of Ca²⁺ from these sites would increase as Ca²⁺ is sequestered by the SR (30). This would result in a greater Ca²⁺ uptake than reported and alter the coupling ratios. However, our basic conclusions regarding the effect of the fatiguing protocol on Ca²⁺ uptake remain valid, because the concentration of oxalate remained constant throughout the experiment.

In summary, intermittent isometric knee extension exercise for 30 min at 60% MVC and 50% duty cycle resulted in parallel depressions in homogenate Ca²⁺ uptake and Ca²⁺-ATPase activity. Moreover, no recovery in SR Ca²⁺ sequestration function or force occurred during a 60-min period after exercise. Our results suggest that PCD is due to more than an impairment in Ca²⁺ release and that a common mechanism, namely elevated O₂ free radicals and/or Ca²⁺, may be responsible for both the disturbances in the Ca²⁺-ATPase pump and the Ca²⁺-release channel. Further studies are needed to establish the exact mechanism responsible for impaired pump function and whether an impairment in pump function impairs Ca²⁺ release, either directly or via changes in EC coupling processes secondary to elevated [Ca²⁺]_f.