INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IN RECENT YEARS, accumulating evidence has implicated altered intracellular Ca²⁺ regulation as a major contributor to fatigue. Studies utilizing single fibers isolated from mice, frog, and toad muscles have demonstrated impaired sarcoplasmic reticulum (SR) Ca²⁺ release with fatigue, with a consequent decline in tetanic cytosolic Ca²⁺ concentration ([Ca²⁺]) and muscle force (1, 13, 30, 54, 55). The mechanisms for impaired SR Ca²⁺ release with fatigue remain incompletely understood. However, these most likely include cytosolic accumulation of Ca²⁺, Mg²⁺, IMP, and AMP; localized depletion of glycogen and ATP; and SR luminal Ca²⁺-P_i precipitation (3, 11, 16, 18, 30, 42, 48). Despite these recent advances in understanding in other species, little is known about the effects of fatigue on SR Ca²⁺ release in human muscle.

In human skeletal muscle, it is not possible to study SR Ca²⁺ regulation in intact single fibers, and thus recent studies have utilized crude muscle homogenate preparations (6, 20, 23, 25, 27, 45). Studies investigating SR function in human muscle with exercise have demonstrated depressed SR Ca²⁺ uptake and/or Ca²⁺-ATPase activity with prolonged submaximal contractions (6, 23, 52), a decrease in SR Ca²⁺ uptake after intense contractions (20, 25, 27), and, recently, depressed SR Ca²⁺ release after 7 min of intense fatiguing knee extensor muscle contractions (27). In the latter study, the torque decline was near maximal by midexercise, although muscle biopsies for SR measures were not taken until the cessation of exercise several minutes later (27). This raises the possibility of a different time course in the decline in Ca²⁺ release and muscle torque with fatigue. Furthermore, the training status of their subjects was not specified, and this may potentially have an important bearing on the decline in SR function. Surprisingly, they found no effects of repeated maximal contractions on SR Ca²⁺-ATPase activity, despite a reduction in SR Ca²⁺ uptake (27), a dissociation not evident in earlier human exercise studies (6, 23, 52). The reason for this discrepancy is unclear and worthy of further investigation. Therefore, the first aim of this study was to investigate the effects of fatigue induced by repeated maximal contractions on each of SR Ca²⁺ release, Ca²⁺ uptake, and Ca²⁺- ATPase activity in human skeletal muscle with biopsy sampling coincident with the rapid decline in peak torque.

Resistance and endurance training induce fundamentally different physiological and muscular performance outcomes (e.g., 5, 22, 37), and their effects on SR function are therefore of interest. However, such studies are sparse and yield conflicting results. Ryanodine receptor (RyR) binding was reduced in long-term resistance-trained elderly men (32), whereas SR Ca²⁺- ATPase protein expression and Ca²⁺ uptake were reduced in hypertrophied rat muscle (31). In contrast, neither SR Ca²⁺-ATPase activity (23) nor content (32) was changed after short- or long-term resistance training in humans, whereas both SR Ca²⁺ uptake and Ca²⁺-ATPase activity were elevated with resistance training in elderly, but not young, women (28). Endurance training in rats induces changes in muscle SR function or SR proteins consistent with an increased expression of oxidative fibers [see review (39)]. Surprisingly, the effects of endurance training on SR function in young adults are unknown, and little is known about whether training protects against deteriorating SR function with fatigue. In humans, resistance training attenuated the decline in SR Ca²⁺-ATPase activity with prolonged submaximal exercise (23), and endurance-trained rats displayed a higher Ca²⁺-ATPase activity at exhaustion than untrained rats (4). No studies have investigated whether resistance or endurance training in humans alters Ca²⁺ release in resting muscle or attenuates the reduction in Ca²⁺ release with fatigue. Therefore, the second aim of this study was to determine the effects of chronic resistance and endurance training on SR Ca²⁺ release, Ca²⁺ uptake, and Ca²⁺-ATPase activity in human skeletal muscle, both at rest and after fatiguing maximal contractions.

Three hypotheses were tested in human skeletal muscle: 1) that fatiguing, maximal contractions would depress SR Ca²⁺ release and Ca²⁺-ATPase activity; 2) that chronic endurance-trained and resistance-trained athletes would exhibit lower rates of SR Ca²⁺ release, Ca²⁺ uptake, and Ca²⁺-ATPase activity in resting muscle; and 3) that the decline in muscle SR function with fatigue would be attenuated in both endurance-trained and resistance-trained athletes.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

1

1

Maximal Aerobic Power

Muscle Function Assessment

Torque-velocity relationship. The peak muscle-generated torque was measured during maximal concentric knee extension contractions at limb angular velocities of 60, 120, 180, 240, 300, and 360°/s on a Biodex isokinetic dynamometer (Biodex Medical Systems, Shirley, NY). Peak torque at each velocity was used to construct a torque-velocity relationship. Before they were tested, subjects warmed up by cycling at 50 W for 3 min and were then strapped to the Biodex chair with belts across the hips, chest, and leg to stabilize the upper body and thigh. After submaximal familiarization contractions for each velocity, subjects performed two practice maximal repetitions, had 1 min of rest, and then performed three maximal dynamic repetitions, each separated by a 2-min rest. The robust reliability of isokinetic measures of muscular strength has been demonstrated in numerous studies (57).

Fatigue test. Subjects performed two trials of a test designed to induce rapid muscular fatigue of the knee extensor muscles. The first was conducted without invasive procedures on the Biodex dynamometer, and the second was conducted on a Cybex isokinetic dynamometer (Cybex II, Lumex, Ronkonkoma, NY) ~7 days later and included both pre- and postexercise muscle biopsies. Separate dynamometers were used for the muscle fatigue and torque-velocity tests for practical reasons. The fatigue test comprised 50 repetitions of maximal concentric knee extension at 180°/s and at 0.5 Hz, with intervening passive knee flexion, as modified from Thorstensson and Karlsson (51). This test was chosen because maximal voluntary contractions would be expected to recruit the entire motoneuron pool, or close to it (19, 29), and thus repeated exhaustive contractions would be expected to maximize any fatigue effects on SR function. The peak torque during the test was calculated as the mean of the 5 strongest in the first 10 contractions. The final peak torque during the test was calculated as the mean of the 5 weakest in the final 10 contractions. The fatigue index (FI) was calculated as the percent decline in torque from peak to final contractions. The FI results were compared for the noninvasive and invasive trials and found to be reproducible (n = 24, r = 0.77, P < 0.0001), consistent with previous reports of a low method error (51). Each subject refrained from exercise, alcohol, and caffeine consumption for 24 h before the test.

Muscle Biopsy Sampling and Analyses

Muscle SR Function

Muscle Ca²+ uptake/release assay. The Ca²⁺ uptake and Ca²⁺ release rates were measured in duplicate in a standard buffer containing HEPES (20 mM, pH 7.0), KCl (150 mM), Mg-ATP (4.5 mM), indo 1 (1 µM, Calbiochem), oxalate (7.5 mM), sodium azide (10 mM), and N,N,N',N'-tetrakis (2-pyridylmethylethylenediamine (5 µM). The extravesicular starting free [Ca²⁺] was determined in the absence of muscle homogenate by using the indo 1 ratiometric data and was calculated by using standard equations (24). This value varied between 0.8 and 1.0 µM but did not differ significantly between groups or rest vs. exercise samples. Slight modifications to previously described methods (45) include a higher oxalate concentration to increase vesicle Ca²⁺ loading and deletion of dithiothreitol from the homogenate buffer to avoid any inhibition of SR Ca²⁺ release. The buffer was stirred and maintained at 37°C. All measurements were completed within 50 min after thawing of the sample. The maximal rates of Ca²⁺ uptake into and release from vesicles formed by homogenization were monitored using indo 1. The reaction was initiated by the addition of 30-50 µl of homogenate to the buffer. Ca²⁺ uptake was mediated via Ca²⁺-ATPase activity because we have previously shown uptake to be inhibited by the addition of the specific inhibitor cyclopiazonic acid (CPA) to this assay medium (45), consistent with the findings of others (53). The Ca²⁺ uptake reaction was then allowed to stabilize before Ca²⁺ release was initiated by the addition of AgNO₃ (141 µM) to the reaction buffer. It is known that Ag⁺ initiates Ca²⁺ release by oxidizing sulfhydryl groups on the RyR and that this Ca²⁺ release can be inhibited by the addition of the reducing agent dithiothreitol (45, 46, 53, 56). Although high concentrations of Ag⁺, such as used in this study, are known to depress SR Ca²⁺-ATPase activity in addition to inducing Ca²⁺ release, these effects are independent (8, 43). Furthermore, Ca²⁺-ATPase activity inhibition by Ag⁺ is beneficial through minimizing confounding Ca²⁺ uptake during the release measurements. Finally, separate experiments with and without CPA yielded no difference in the measured rate of Ag⁺-induced Ca²⁺ release, confirming that the release was not simply due to inactivation of Ca²⁺ pumps. In human muscle homogenates (n = 15 paired observations), the Ag⁺-induced Ca²⁺ release was not significantly different in the absence vs. presence of CPA (2.20 ± 0.38 vs. 1.83 ± 0.29 µmol · g¹ · min¹, respectively). Preliminary experiments with the RyR modulators ruthenium red and caffeine were unsuccessful because of marked effects observed on the indo 1 ratio in these experimental conditions (data not shown), and these were therefore not employed. Spectral changes of indo 1 were monitored by using a luminescence spectrometer (AB2, SLM-Aminco, Urbana, IL). The sample was excited by a xenon lamp at 349 nm with a band pass of 1 nm; emission was measured at 410 nm for Ca²⁺-bound and at 485 nm for Ca²⁺-free forms of dye, with 8-nm band passes. A 410-to-485 nm fluorescence ratio was collected every 1 s. Minimum and maximum ratios were determined at the completion of the assay by the addition of EGTA (3.5 mM) and CaCl₂ (5 mM), respectively. The dissociation constant for indo 1 was determined to be 164 nM, with the free [Ca²⁺] calculated with the use of standard equations (24).

Muscle SR Ca²+ ATPase activity. The SR Ca²⁺-ATPase activity was determined at 37°C in triplicate, using 50 µl of homogenate and a spectrophotometric method (47). The total ATPase activity was first measured after the addition of 10 µl of 100 mM CaCl₂, giving a final total concentration of 0.6 mM CaCl₂ with a free [Ca²⁺] of ~10 µM (47). The basal ATPase activity (Mg²⁺-ATPase) was then measured after the addition of 20 µl of 2 M CaCl₂, giving a final concentration of 40 mM CaCl₂, which selectively inhibits the SR Ca²⁺-ATPase. The SR Ca²⁺-ATPase activity was then determined from the total minus basal activities. The method is specific for SR Ca²⁺-ATPase activity, as indicated by its inhibition by both CPA (45) and a Ca²⁺-ATPase-specific antibody (47), by the close correspondence between Ca²⁺-ATPase activity measures in a homogenate preparation and purified SR, as well as the similar [Ca²⁺] and temperature dependence (47). The Ca²⁺ uptake and Ca²⁺-ATPase activity data are not used to determine turnover rates because of the markedly different free [Ca²⁺] in the two assays, of ~1 and ~10 µM, respectively.

Muscle protein and calculations. SR measures were expressed relative to muscle wet weight (µmol · min¹ · g muscle¹) and, to identify the effects of any fluid shifts, muscle protein content (µmol · min¹ · g protein¹). Muscle protein was determined spectrophotometrically in triplicate, with albumin as a standard. The decline in SR function with fatigue is denoted by  (e.g., Ca²⁺ release) and the percent decline from rest values by % (e.g., %Ca²⁺ release).

Variability of SR function measurements. We previously found no difference between repeat biopsies for SR Ca²⁺ uptake or Ca²⁺-ATPase activity in two subjects (6). To determine intrasubject variability, a vastus lateralis muscle biopsy was taken from separate incisions in the same leg in three control subjects (36.3 ± 8.3 yr, 179.0 ± 7.5 cm, 75.2 ± 8.3 kg; means ± SD), before and after a 30-min supine rest, and muscle homogenates were analyzed for SR Ca²⁺ uptake and Ca²⁺ release rates. The intra-assay coefficient of variation was 14.8% (n = 18) for Ca²⁺ release and 13.4% for Ca²⁺ uptake (n = 20). The interassay variability in rat muscle was previously reported as 8.1 and 4.3% for Ca²⁺ uptake and Ca²⁺-ATPase, respectively (45). The limited tissue sample obtained by biopsy precludes determination of the interassay variability for SR variables on the subjects reported in this study. However, the tissue sample size was sufficient in one subject who performed the muscle fatigue test to allow determination of interassay variability. The SR Ca²⁺ release, Ca²⁺ uptake, and Ca²⁺-ATPase activity were determined on separate days for both rest and fatigue measures and yielded almost identical results. Furthermore, to minimize variability, all rest and fatigue SR analyses for each subject were performed in the same assay run. It is unlikely that inclusion of the female subject in the resistance-trained group biased the SR data, on the basis of very similar values for Ca²⁺ uptake and Ca²⁺-ATPase activity for young women and men (6, 28).

Muscle metabolites. Muscle was freeze dried, dissected free of connective tissue, weighed, powdered, and extracted. Approximately 2 mg of freeze-dried muscle were extracted in perchloric acid (0.5 M) and neutralized in KHCO₃ (2.1 M). Muscle extracts were analyzed for ATP, phosphocreatine (PCr), glycogen, lactate, and creatine (Cr) contents by using fluorometric techniques, and ADP, AMP, and IMP contents were quantified by reverse-phase HPLC. Muscle metabolites excepting lactate and glycogen were corrected for total Cr content. A portion of muscle was extracted in 250 µl of 2 M HCl at 100°C for 2 h and neutralized with 75 µl of 0.76 M NaOH for measurement of glycogen using an enzymatic method. All methods for metabolite analyses were as previously described (25). Muscle homogenate pH was determined in 2-4 mg freeze-dried tissue (1 mg/100 µl) with a pH microelectrode at 37°C (MI-410 microelectrode) in a homogenate buffer containing sodium iodoacetate (5 mM), KCl (145 mM), and NaCl (10 mM).

Statistical Analyses

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Significant differences were found between groups for O_2 peak, sum of skinfolds, and FI during repeated maximal contractions (P < 0.05), with higher O_2 peak and lower sum of skinfolds, and FI in ET than in both UT and RT (Table 1, P < 0.05). A significant main effect for training status was found for peak torque during dynamic isokinetic contractions (Fig. 1) being higher in RT than both ET (P < 0.05) and UT (P = 0.05). No significant group differences were found between groups for thigh circumference (Table 1). The FI results for the noninvasive trial on the Biodex dynamometer (48.6 ± 10.7, 50.1 ± 9.3, 31.2 ± 14.2% for UT, RT, and ET, respectively; means ± SD) did not differ from those in the invasive trial on the Cybex dynamometer (Table 1).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Quadriceps muscle torque-velocity relationship during dynamic isokinetic contractions in untrained (UT; ), resistance-trained (RT; ), and endurance-trained (ET; ) subjects. Values are means ± SE; n = 8. *RT > ET (P < 0.05) and UT (P = 0.05).

Muscle Fiber Type and Metabolites

A significant exercise main effect (P < 0.05) was seen for each metabolite except ADP, with decreases in ATP, PCr, glycogen, and pH and increases in lactate, Cr, and IMP (Table 3). A significant train-by-exercise interaction was found for PCr and Cr (P < 0.05). Resting muscle PCr was higher in RT than in UT (P < 0.05), which was higher than in ET (P < 0.05). After exercise, ET had higher PCr and lower Cr than RT and UT (P < 0.05).

Muscle SR Characteristics

Exercise and training comparisons. A significant exercise main effect was found for all SR variables, each of which was depressed with fatigue (P < 0.05), whether expressed per muscle mass (µmol · min¹ · g muscle¹) or muscle protein content (µmol · min¹ · g protein¹).

1

1

1

1

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Maximal Ag+-induced Ca2+ release rates measured in crude homogenates from vastus lateralis muscle obtained at rest and at fatigue induced by 50 repeated maximal quadriceps contractions in UT, RT, and ET groups. Values are means ± SE; n = 8. *Fatigue < rest within same group;  < UT for corresponding sample;  < RT for corresponding sample (P < 0.05).

1

1

1

1

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Maximal Ca2+ uptake rates measured in crude muscle homogenates obtained at rest and at fatigue in UT, RT, and ET groups. Values are means ± SE; n = 8. *Fatigue < rest within same group;  < UT for corresponding sample (P < 0.05).

1

1

1

1

1

1

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Maximal Ca2+-ATPase activity measured in crude muscle homogenates obtained at rest and at fatigue in UT, RT, and ET groups. Values are means ± SE; n = 8. *Fatigue < rest within same group;  < UT for corresponding sample;  < RT for corresponding sample (P < 0.05).

1

1

1

1

Variability of SR function measurements. Very similar SR Ca²⁺ uptake and Ca²⁺ release rates were found in measurements from two separate resting biopsies in the same individual (n = 3). The SR Ca²⁺ uptake was 1.01 ± 0.15 vs. 0.97 ± 0.15 µmol · g¹ · min¹, whereas the Ca²⁺ release was 2.96 ± 1.19 vs. 3.32 ± 0.88 µmol · g¹ · min¹, in resting biopsy 1 and in biopsy 2, respectively. When the same muscle sample in one individual was measured on separate days, almost identical SR Ca²⁺ release, Ca²⁺ uptake, and Ca²⁺-ATPase activity for both rest and fatigue measures were found. The values (day 1 rest, fatigue; day 2 rest, fatigue; µmol · g¹ · min¹) for Ca²⁺ release were 1.35, 0.51; 1.35, 0.54; for Ca²⁺ uptake 1.38, 0.84; 1.50, 0.78; and for Ca²⁺-ATPase 10.65, 8.10; 10.74, 8.22.

Relationships Between SR Function and Fiber Type

Resting muscle. The SR Ca²⁺ release, Ca²⁺ uptake, and Ca²⁺-ATPase activity rates in resting muscle were each related to the type II fiber proportion (n = 24, P < 0.01, Fig. 5). SR function values (µmol · min¹ · g muscle¹) for type I and II fibers were estimated by extrapolation from regression equations related to the type II fiber proportion by substituting values of 0 and 100 for type I and type II fiber values, respectively. This yielded estimates of 1.13 vs. 2.54 for SR Ca²⁺ release, 1.20 vs. 2.26 for Ca²⁺ uptake, and 8.61 vs. 20.71 for Ca²⁺-ATPase activity, respectively. The rates of SR Ca²⁺ uptake and Ca²⁺-ATPase activity were highly correlated (e.g., rest plus fatigue data pooled, r = 0.81, n = 48, P < 0.0001).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Relationships between the maximal rates of SR Ca2+ release (A), Ca2+ uptake (B), and Ca2+-ATPase activity (C) measured in resting muscle and the proportion of type II muscle fibers. SR units are µmol · min1 · g muscle1; n = 24. , UT; , RT; and , ET. Regression lines are for pooled data and equations are as follows: Ca2+ release = (0.018 × type II%) + 1.134 (r = 0.62, P < 0.005); Ca2+ uptake = (0.011 × type II%) + 1.199 (r = 0.52, P < 0.01); Ca2+-ATPase = (0.121 × type II%) + 8.606 (r = 0.56, P < 0.005).

Fatigued muscle. The Ca²⁺ release (r = 0.58, P < 0.05), Ca²⁺ uptake (r = 0.40, P = 0.05), and Ca²⁺-ATPase activity (r = 0.43, P < 0.05) pooled for all subjects were also related to the type II fiber proportion (n = 24, µmol · min¹ · g muscle¹, Fig. 6). The Ca²⁺ release was related to both Ca²⁺ uptake (r = 0.53, P < 0.01) and Ca²⁺-ATPase activity (r = 0.65, P < 0.001) with the latter two also related (r = 0.47, P < 0.05).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Relationships between the fatigue-induced absolute decline in the maximal rates of Ca2+ release (A), Ca2+ uptake (B), and Ca2+-ATPase activity (C) in muscle and the proportion of type II muscle fibers. SR units are µmol · min1 · g muscle1; n = 24. , UT; , RT; and , ET. Regression lines are for pooled data and equations are as follows: Ca2+ release = (0.014 × type II%) +0.153 (r = 0.58, P < 0.005); Ca2+ uptake = (0.009 × type II%) + 0.198 (r = 0.40, P = 0.05); Ca2+-ATPase = (0.085 × type II%) +1.886 (r = 0.43, P < 0.05).

Relationships Between SR Function and Metabolites

Relationships Between SR Function, Fiber Type, and Performance

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Relationships between the fatigue index (FI) and the fatigue-induced absolute decline in the maximal rates of Ca2+ release (A), Ca2+ uptake (B), and Ca2+-ATPase activity (C) in vastus lateralis muscle during the 50 repeated maximal quadriceps contractions. SR units are µmol · min1 · g muscle1; n = 24. , UT; , RT; and , ET. Regression lines are for pooled data and equations are as follows: FI (%) = (21.41 × Ca2+ release) + 23.8 (r = 0.51, P < 0.05); FI (%) = (22.56 × Ca2+ uptake) + 26.61 (r = 0.51, P < 0.05); FI (%) = (2.63 × Ca2+-ATPase) + 25.51 (r = 0.52, P < 0.01).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study reports several unique findings on the effects of intense fatiguing contractions on skeletal muscle SR function, measured under standardized in vitro conditions, in each of UT, ET, and RT subjects. We show for the first time that SR Ca²⁺ release in human muscle is markedly depressed with fatigue in UT subjects as well as in athletes and that the depression was significantly related to the extent of muscle fatigability. These findings indicate that impaired SR function is an obligatory response to fatiguing contractions and point toward a key involvement of SR dysfunction in muscular fatigue in humans. Both fiber type and training appear to influence SR function in resting muscle. Lower SR Ag⁺-induced Ca²⁺ release and SR Ca²⁺ uptake rates were found in resting muscle in ET, consistent with their low type II fiber proportion; however, fiber type differences could not explain the lower muscle Ca²⁺ uptake found in RT than controls.

Fatigue Depresses Ag+-induced SR Ca²⁺ Release and SR Ca²⁺-ATPase Activity in Human Muscle

Ag⁺ induces Ca²⁺ release via oxidizing SH groups on the physiological release channel, the RyR, and this can be inhibited by SH reducing agents such as dithiothreitol (26, 45, 46, 53, 56). Although Ag⁺-induced Ca²⁺ release is specific for Ca²⁺ release from the RyR (14), several limitations exist in our methodology. First, neither Ag⁺ nor the process of RyR oxidation is involved in the physiological induction of SR Ca²⁺ release (34), and thus the physiological significance of our findings remains to be elucidated by further study. Caffeine is commonly used to study fatigue effects on SR Ca²⁺ release because of its specific effects on the RyR (14). However, caffeine could not be utilized here because of its marked effects on indo 1 at the high concentrations required to induce release in human muscle. Importantly, however, others have reported qualitatively similar responses to fatigue, or to conditions mimicking fatigue, in Ca²⁺ release induced by Ag⁺, caffeine, and 4-chloro-m-cresol (14, 53, 56). In particular, Ag⁺- and 4-chloro-m-cresol-mediated Ca²⁺ release rates were similarly depressed under conditions of graded fatigue (53). Hence, it is highly probable that our findings of reduced Ag⁺-induced Ca²⁺ release are valid indicators of depressed SR release in human muscle. Finally, the measured Ag⁺-induced Ca²⁺ release rate will be affected by Ca²⁺ buffering due to oxalate, EDTA, and muscle proteins within the assay medium and homogenate. Therefore, the measured Ca²⁺ release (and Ca²⁺ uptake) rates reported here should be compared only with other measures using the same experimental conditions. Because all assay conditions were constant, the comparisons between rest and exercise and between groups are valid.

We report for the first time a decline in Ca²⁺-ATPase activity in human muscle after maximal contractions and confirm proportional and significantly correlated reductions in SR Ca²⁺ uptake (43%) and Ca²⁺-ATPase activity (39%). Thus neither maximal contractions, nor prolonged submaximal contractions (6, 52) appear to differentially affect the hydrolytic and Ca²⁺ vectorial transport properties of the Ca²⁺-ATPase enzyme. Thus the recent report of reduced SR Ca²⁺ uptake but unchanged Ca²⁺-ATPase activity with fatigue (27) is inconsistent with our findings and those of other human studies. The large percent reduction in muscle SR Ca²⁺ uptake after intense contractions in the present study was similar to other reports in humans (20, 25) and in horses (10) but twice that found in humans after prolonged exercise (6). This effect most likely reflects the greater muscular fatigue and thus depression in SR function induced by intense contractions (53).

An important finding was that in vitro SR Ca²⁺ release, Ca²⁺ uptake, and Ca²⁺-ATPase activity were significantly depressed with fatigue in both ET and RT groups, as well as in the UT group. Thus chronic training did not prevent the decline in SR function with fatigue. Neither the absolute () nor the relative (%) decline in SR Ca²⁺ release differed between groups and both variables were correlated with the FI. This suggests that depressed Ca²⁺ release is an essential component of muscle fatigue during voluntary exercise in humans. The relationship between the FI and all SR variables is also consistent with an earlier finding in rat muscle that the extent of depression in SR function is dependent on the severity of fatigue (53). The relative decline with fatigue in SR Ca²⁺ release was ~9% less (not significant) in the ET than in UT, consistent with the lesser decline in SR function in subjects with a predominance of type I fibers. The lack of significance of a training effect most likely reflects the modest statistical power of the study. Together with the cross-sectional design of the study, these suggest a likely type II error with respect to training effects on the decline in SR Ca²⁺ release with fatiguing exercise. The Ca²⁺ uptake and Ca²⁺-ATPase activity (per g protein) were less in ET than in UT, consistent with their lower proportion of type II fibers. Overall, our RT data suggest that resistance training does not attenuate the SR function during repeated maximal contractions.

Role of Fiber Type and Metabolism in Depressed SR Function with Fatigue

Muscle Fiber Composition and Training Affect SR Function in Resting Muscle

Muscle SR characteristics clearly varied between the different training groups. However, the lower SR Ca²⁺ release, Ca²⁺ uptake, and Ca²⁺-ATPase activity in ET appeared to be entirely consistent with their lower proportion of type II fibers. The cross-sectional experimental design employed makes it impossible to discern whether endurance training exerts any additional effects on muscle SR characteristics beyond those differences due to muscle fiber type, but these training effects are likely to be small in humans (32, 38). Specific training effects on muscle SR are suggested in RT, in which SR Ca²⁺ uptake was significantly lower than in controls, without differences in fiber composition. This is consistent with decreased Ca²⁺-ATPase protein expression, Ca²⁺ uptake, and Ca²⁺-ATPase activity in rat fast-twitch muscle after muscle loading (31) but differs from the unchanged resting Ca²⁺-ATPase activity or content with short-term and chronic resistance training in young and elderly humans (23, 32). The reasons for this are unclear. The similar SR Ca²⁺ release in RT and UT may indicate that RyR are upregulated in proportion to muscle mass. Interestingly, decreased muscle RyR content was found in long-term RT elderly men (32), although this does not necessarily imply that maximal Ca²⁺ release rates would also be lower. On the other hand, both RyR expression and SR Ca²⁺ release were increased in resting muscle after sprint training, without change in fiber composition (40). The lack of effect of RT on Ca²⁺ release might reflect the fact that our RT group did not appear to be as highly trained as the ET group, although their RT training status was confirmed by greater dynamic torque production. Overall, our findings therefore suggest that chronic endurance and resistance training do not induce major changes in SR functional characteristics but, rather, that these are dominated by the muscle fiber composition.

Contractile and Performance Implications

In conclusion, this study advances our understanding of fatigue, training, and muscle fiber composition effects on SR function in human muscle. The maximal in vitro SR Ca²⁺ release rate was markedly reduced after voluntary fatiguing contractions, with proportional depressions in in vitro Ca²⁺ uptake and Ca²⁺-ATPase activity. These findings indicate that SR function in human muscles is similarly affected by fatigue as in other species. Human SR function is clearly dependent on muscle fiber type but may be additionally influenced by training status. ET athletes had lower SR function than controls, although this was consistent with their lower proportion of type II fibers. RT athletes had an unchanged fiber composition and thus similar SR Ca²⁺ release to controls, but lower SR Ca²⁺ uptake, which suggests additional resistance training effects. Finally, chronic training did not prevent or even significantly attenuate the depressive effects of fatiguing exercise on muscle SR Ca²⁺ release. This points to a possible causal link between SR dysfunction and fatigue in contracting human muscle.