Ca²⁺-pumping impairment during repetitive fatiguing contractions in single myofibers: role of cross-bridge cycling

Abstract

The energy cost of contractions in skeletal muscle involves activation of both actomyosin and sarcoplasmic reticulum (SR) Ca²⁺-pump (SERCA) ATPases, which together determine the overall ATP demand. During repetitive contractions leading to fatigue, the relaxation rate and Ca²⁺ pumping become slowed, possibly because of intracellular metabolite accumulation. The role of the energy cost of cross-bridge cycling during contractile activity on Ca²⁺-pumping properties has not been investigated. Therefore, we inhibited cross-bridge cycling by incubating isolated Xenopus single fibers with N-benzyl-p-toluene sulfonamide (BTS) to study the mechanisms by which SR Ca²⁺ pumping is impaired during fatiguing contractions. Fibers were stimulated in the absence (control) and presence of BTS and cytosolic calcium ([Ca²⁺]_c) transients or intracellular pH (pH_i) changes were measured. BTS treatment allowed normal [Ca²⁺]_c transients during stimulation without cross-bridge activation. At the time point that tension was reduced to 50% in the control condition, the fall in the peak [Ca²⁺]_c and the increase in basal [Ca²⁺]_c did not occur with BTS incubation. The progressively slower Ca²⁺ pumping rate and the fall in pH_i during repetitive contractions were reduced during BTS conditions. However, when mitochondrial ATP supply was blocked during contractions with BTS present (BTS + cyanide), there was no further slowing in SR Ca²⁺ pumping during contractions compared with the BTS-alone condition. Furthermore, the fall in pH_i was significantly less during the BTS + cyanide condition than in the control conditions. These results demonstrate that factors related to the energetic cost of cross-bridge cycling, possibly the accumulation of metabolites, inhibit the Ca²⁺ pumping rate during fatiguing contractions.

during the transition from rest to exercise in skeletal muscle, the demand for ATP increases several hundredfold, primarily because of the activation of both actomyosin (i.e., cross-bridge cycling) and sarcoplasmic reticulum (SR) Ca²⁺ ATPase (SERCA) pumps (6). After membrane depolarization, cytosolic Ca²⁺ concentration is transiently increased, thereby elevating actomyosin and SERCA ATPase activities. Although the Na⁺-K⁺-ATPase activity has a high contribution in the energy cost at rest [∼40% of total; (27)], it is quite small during contractions [∼1.5–7%; (6)]. Conversely, during contractions, actomyosin ATPase and SERCA account for ∼60% and 40% of the total ATP demand, respectively (5, 6, 22, 30). However, the energy cost of Ca²⁺ handling by SERCA during contractions may be as high as 80% of the total energy cost in type IIB fibers (40). Interestingly, although the high cost of SR Ca²⁺ pumping during exercise is often not appreciated, SERCA dysfunction (i.e., slowing in the Ca²⁺ uptake after contractions) has been implicated in many pathological states, including chronic obstructive pulmonary disease (13), muscular dystrophy (11), and heart failure (19). Because SERCA dysfunction may be an important factor associated with the decreased locomotor activity in these diseases (13), the study of the factors that influence SERCA activity during fatigue is very relevant.

The increased ATP demand during exercise is met by three energetic sources: phosphocreatine breakdown, anaerobic glycolysis, and mitochondrial oxidative phosphorylation (2). The contribution of each source is time- and fiber-type-dependent and is regulated by the total metabolic energy cost of the contractile work (2, 14). Also, the contribution of each energetic source during exercise has been shown to regulate the development of muscle fatigue. During high-intensity exercise, skeletal muscle fatigue develops rapidly, particularly in fast-twitch, glycolytic fibers (2). Fatigue development is highly dependent on intracellular product accumulation (e.g., P_i and H⁺), as well as substrate depletion, which may interfere directly with the energy supply metabolism and with the excitation-contraction coupling in skeletal muscle (2, 10). One hallmark of fatigue is the slowing of relaxation, which has often been associated with a decrease in SR Ca²⁺ pumping rate due to SERCA dysfunction (2, 28). The reduced ATP hydrolysis rate of SERCA at fatigue along with a reduced SR Ca²⁺ release may directly affect the total energy cost during contractions (26). Although the impairment in Ca²⁺ uptake at fatigue has been well described, the importance of the ATP demand on SERCA regulation during repetitive contractions leading to fatigue is controversial (2, 28). Furthermore, little is known concerning whether the ATP cost of Ca²⁺ handling during muscle activation is sufficient to regulate the SR Ca²⁺ uptake kinetics and function during fatiguing contractile work.

Several studies have assessed SR Ca²⁺ uptake energy cost by blocking the activation of cross-bridge cycling using the compound N-benzyl-p-toluene sulfonamide (BTS) (5, 7, 22, 30, 40). This compound inhibits the affinity between myosin and actin (23) but does not affect Ca²⁺ release and uptake in single muscle fibers (9, 30, 39). Interestingly, the role of SERCA energy requirements during repetitive contractions on real-time measures of intracellular pH (pH_i) and SR Ca²⁺ pumping function has not been investigated. The aim of the present investigation was to use BTS as a tool to test the hypothesis that the SR Ca²⁺ pumping dysfunction that occurs during repetitive contractions leading to fatigue is dependent on the accumulation of intracellular metabolites that arise from the energy cost of cross-bridge cycling.

METHODS

Animal care and single fiber isolation.

All procedures were approved by the University of California, San Diego Institutional Animal Care and Use Committee and conformed to American Physiological Society guidelines. Female adult Xenopus laevis were anesthetized (pentobarbital sodium: 120 mg/kg), double pithed, and decapitated, and the lumbrical muscles (II–IV) were removed. Single skeletal muscle fibers (n = 29) were isolated using a pair of scissors and forceps under dark-field illumination. Dissection and experiments were performed in Ringer solution (116.5 mM NaCl, 2 mM KCl, 1.9 mM CaCl₂, 2 mM NaH₂PO₄, and 0.1 mM EGTA, at pH 7.0) at room temperature (20–22°C).

Intracellular Ca²⁺ and pH detection.

Cytosolic calcium concentration ([Ca²⁺]_c) and pH_i changes were obtained by fluorescence spectroscopy using a Photon Technology International illumination and detection system (DeltaScan model), integrated with a Nikon inverted microscope with a 40× Fluor objective. After dissection and isolation, each fiber was pressure injected using a micropipette with either the ratiometric compounds Fura-2 (12 mM) for [Ca²⁺]_c detection, or 2′,7′-bis-(2-carboxyethyl)-5-(and -6)carboxyfluorescein (BCECF; 10 mM) for pH_i detection (both compounds were diluted in 150 mM KCl and 10 mM HEPES at pH 7.0). After injection, the myofiber rested for 1 h. Both indicators were obtained from Invitrogen (Carlsbad, CA).

For detection of [Ca²⁺]_c during contractions, the fibers were illuminated with two rapidly alternating (200 Hz) excitation wavelengths of 340 nm and 380 nm, and the resulting fluorescence emissions at 510 nm were divided (340 nm/380 nm) to obtain the Ca²⁺-dependent signal. Fluorescence excitation ratio (340/380 nm; R) was converted to [Ca²⁺]_c, according to Eq. 1.

$${\left[{\text{Ca}}^{2+}\right]}_{\text{c}}=K_{\text{D}}\cdot \beta \cdot \left[\left(R-R_{\min }\right)/\left(R_{\max }-R\right)\right]$$ (1)

From Eq. 1, K_D, the dissociation constant for Ca²⁺-Fura-2, was set to 224 nM (33). β is the fluorescence ratio between high and no [Ca²⁺]_c at 380 nm, and R_min and R_max are the fluorescence ratios at no cytosolic calcium and high calcium, respectively. R_min and R_max were determined using an internal in vivo calibration described by Kabbara and Allen (16) with small modifications. To determine R_min, fibers were incubated for 30 min with a no-calcium Ringer-EGTA solution (116.5 mM NaCl, 2 mM KCl, 2 mM NaH₂PO₄, 10 mM EGTA, at pH 7.3) supplied with 2 mM caffeine. Then, 100 μM BAPTA-AM, solubilized in DMSO and then in Ringer-EGTA (DMSO final concentration was 1%), was incubated for 30 min followed by washout with Ringer-EGTA solution for an additional 20 min. R_min was determined as the smallest R value observed (0.30 ± 0.04; n = 4; means ± SE). For R_max, another group of fibers were exposed to a high-calcium Ringer-MG solution (116.5 mM N-methyl-d-glucamine, 2 mM KCl, 20 mM CaCl₂, 0.1 mM EGTA, pH 7.0) for 120 min followed by a 2 mM 4-chloro-m-cresol to obtain R_max (6.97 ± 0.56; n = 4; means ± SE). β was determined (4.05 ± 0.36; n = 6 fibers, means ± SE) as described previously (3, 4). The contraction-induced fluorescence ratio was calculated by averaging the fluorescent signal in the final 100 ms of stimulation and was then used to determine peak [Ca²⁺]_c. The intracellular cytosolic Ca²⁺ before each contraction (i.e., basal [Ca²⁺]_c) was calculated by averaging the signal in the 100 ms before each stimulation.

When the pH_i changes were monitored during contractions, fibers were illuminated at 440 nm and 490 nm, and the resulting fluorescence emissions at 535 nm were divided (490 nm/440 nm) to obtain the pH-dependent signal. An in vivo calibration was performed in a different group of fibers (n = 5), according to Westerblad and Allen (32). Fibers were injected with BCECF, mounted in the experimental chamber, and incubated for 20 min in each of two different pH buffered solutions (175 mM KCl, 1.2 mM KH₂PO₄, 0.5 mM MgCl₂, 10 mM HEPES, 10 μM nigericin), previously titrated with KOH to pH 5.0 and 9.0, as the fluorescence signal was detected. Fluorescence excitation ratio (440/490 nm; F) was then converted to pH_i according to Eq. 2.

$${\text{pH}}_{\text{i}}=\text{p}{K}_{\text{A}}-\log \left(S_b/S_a\right)-\log \left[\left(F-F_b\right)/\left(F_a-F\right)\right]$$ (2)

From Eq. 2, pK_A, the logarithmic dissociation constant, was set to 7.15 (32). S_b and S_a are the fluorescence signals from 440 nm excitation at pH 9.0 and pH 5.0, respectively, and the fluorescence ratio of S_b/S_a was determined as 0.73 ± 0.06 (n = 5 fibers). F_a and F_b are the fluorescence ratios at low (acid) and high (basic) pH, respectively, and were determined as 0.59 ± 0.01 for F_a and 1.82 ± 0.12 for F_b (n = 5 fibers; means ± SE).

Determination of [Ca²⁺]_c pumping rate by the sarcoplasmic reticulum.

High-affinity Ca²⁺ probes (e.g., Fura-2 and Indo-1) have been used to investigate the kinetics of Ca²⁺ uptake by the SR during contractions (36). To explore the effects of fatigue development in the presence and absence of cross-bridge cycling on Ca²⁺ uptake kinetics, the poststimulation period (i.e., relaxation) of selected contractions was used to determine either the fast-phase rate of Ca²⁺ uptake and the function of SR Ca²⁺ pumping. Different time points of the contractile periods, such as the first contraction, contractions at 60, 120, 240, 360, 480, and 600 s, as well as the last contraction, were selected, and [Ca²⁺]_c kinetics were determined as follows.

To determine the fast-phase rate constant of [Ca²⁺]_c decay (λ₁), the poststimulation [Ca²⁺]_c signal was fitted using a double-exponential equation, as shown in 3.

$${\left[{\text{Ca}}^{2+}\right]}_{\text{c}}=A{e}^{{-\lambda }_1.t}+B{e}^{{-\lambda }_2.t}$$ (3)

From Eq. 3, A and B represent that fraction of [Ca²⁺]_c decay for each exponential phase, λ₁ and λ₂ represent the rate of Ca²⁺ decay for the fast and slow phases, respectively, and t is the time of Ca²⁺ decay.

To measure the function of SR Ca²⁺ pumping during the contractile bouts, we used a method described by Klein et al. (17), which was used in several reports by the group of Dr. Westerblad (34, 35). This method assumes that the elevated long tail of intracellular [Ca²⁺]_c decline (from 200 to 2.5 s after the stimulation period), represents a state which Ca²⁺ pumping by SERCA is proportional to the rate of [Ca²⁺]_c decline (d[Ca²⁺]_c/dt). Therefore, SERCA Ca²⁺ pumping curve was obtained by plotting -d[Ca²⁺]_c/dt during the tail of [Ca²⁺]_c decay vs. [Ca²⁺]_c (Eq. 4).

$$-\text{d}{\left[{\text{Ca}}^{2+}\right]}_{\text{c}}/\text{d}t=A\cdot {\left[{\text{Ca}}^{2+}\right]}_{\text{c}}^N-L$$ (4)

From Eq. 4, -d[Ca²⁺]_c/dt is the rate of intracellular [Ca²⁺]_c decline, and A, N, and L are three adjustable parameters representing the rate of SR Ca²⁺ pumping (in μM^N−1/s), the power function, and the SR Ca²⁺ leak, respectively. To compare directly the SR Ca²⁺ pumping function (in μM⁻³/s) between contractions [first contraction: 0.15–2.5 s; 62nd contraction: 0.35–2.5 s; fatigue time point in the control (442 ± 23 s): 0.5–2.5 s; or last contraction with BTS: 0.2–2.5 s], the curves were fitted setting N = 4 and L = 5 nM/s, with a maximum increase in least-square error of 15% (34). The prolonged tail of [Ca²⁺]_c decay after the stimulation period is dependent on the time course of the fast phase of [Ca²⁺]_c decay (λ₁), which was determined by Eq. 3. The λ₁ was slowed during the time course of the contractile bout under control conditions, as shown in Fig. 3C; therefore, the beginning of the slow phase of [Ca²⁺]_c decay was delayed between the first, the 62nd, and the fatigue time point. Moreover, the fast phase of [Ca²⁺]_c decay with BTS present was not significantly slowed (Fig. 3C), so the beginning of the tail of [Ca²⁺]_c decay was not modified from the first contraction.

Fig. 3.

Kinetics of the initial fast-phase rate of [Ca²⁺]_c decay (λ₁) at different time points of a repetitive contractile bout in the presence and absence of BTS. A and B: trace recordings of [Ca²⁺]_c decay after stimulation period (starting at 0 s) at different time points during the contractile bouts in the absence (control; A) and the presence of BTS (BTS; B). In A, the first contraction is shown (a), then the 62nd contraction (360 s; b), and the fatigue time point (c). In B, the first contraction is shown (a), then the 62nd contraction (360 s; b), and the last contraction (d). C: fast-phase rate of [Ca²⁺]_c decay (λ₁) in the absence (●) and the presence of BTS (○). n = 5 fibers. *P < 0.05 vs. initial contraction, one-way ANOVA, repeated measures; **P < 0.05 control vs. BTS, two-way repeated-measures ANOVA. Values are expressed as means ± SE. The time points shown in A and B are indicated in C as a, b, c, and d.

Experimental protocol.

Following the fiber injection and subsequent rest period, platinum clips were attached to the tendons, and each fiber was mounted in a single muscle strip myograph system [model 920CS, Danish Myo Technology (DMT), Atlanta, GA] and was placed on the stage of the inverted microscope. After mounting, the fiber length that promotes the maximal isometric tension (L₀) was determined by stretching the fiber until maximal tetanic tension (70 Hz) was developed. After setting L₀, the fiber was allowed to rest for 30 min. Tetanic contractions (70 Hz, 250-ms train duration, 2-ms monophasic pulses, 8 V) were evoked by direct stimulation, using a Grass S48 stimulator (Quincy, MA). Tension development was measured with a force transducer system (model KG4, 0–50 mN, DMT, Atlanta, GA) that was previously calibrated with weights (20 mg to 1 g) and then converted to millinewtons. A Biopac Systems MP100WSW (Santa Barbara, CA) A-D converter was used to convert the analog to digital data, and AcqKnowledgeIII 3.2.6 software (Biopac Systems) was used to analyze the data. Tension development (in millinewtons) was normalized to the cross-sectional area (in millimeters squared) determined from the diameter of the fiber (in micrometers).

Fatigue-inducing contraction bouts were utilized in the present study in the presence or absence of 12.5 μM BTS. To avoid an order effect, the fibers performed two consecutive contractile periods with and without BTS in random order separated by 60 min of recovery. It has been previously shown that this rest period allows complete recovery of the myofiber and should not interfere with the time to fatigue on a second contractile bout (24). When BTS was present during the fatigue run, maximal tension (70 Hz) was previously tested to ensure the same tension development between contractile periods, and fibers were preincubated with BTS for 10 min, which produced a strong inhibition of peak isometric tension developed at 70-Hz stimulation (∼95%) (30). The effect of BTS on fiber contractility was transient, and when this compound was washed out, the normal contractile characteristics of the fiber were restored after ∼10 min. The fatigue-inducing contraction bout consisted of a series of repeated contractions with decreases in the rest period between contraction, using increments in train frequency (one contraction each at 8, 6, 4, 3, 2 and 1 s) for 2 min each. This protocol was performed only with fast-fatiguing fibers [time to reach the fatigue time point in control conditions of less than 600 s (25)]. The bout was terminated (i.e., fatigue time point) when the initial isometric tension was reduced by 50% (control condition; when BTS was absent) or the initial Fura-2 fluorescence ratio decreased to ∼50–60% of the initial (when BTS was present).

In a separate group of fibers, after the first contractile period in the presence of BTS, a second contractile bout was carried out in the presence of BTS plus 2 mM NaCN to inhibit mitochondrial respiration and, thereby, aerobic ATP supply. In these cases, the fibers were exposed to NaCN for 2 min before the beginning of the work period and during the whole contractile period. These contractile bouts were followed by extensively washing out BTS and CN with normal Ringer solution, and a third contractile bout (control) was performed to ensure fiber viability starting 60 min after the second contractile bout.

Data analysis.

The experimental results are presented as means ± SE. For comparisons between multiple groups, one-way ANOVA followed by the Tukey test or two-way ANOVA was used, as indicated. Analyses were carried out using PRISM 4 software and P < 0.05 was considered to represent significant difference.

RESULTS

Steady-state isometric tension and [Ca²⁺]_c responses.

As illustrated in Fig. 1, incubating fibers with BTS (12.5 μM) inhibited tension development (70-Hz stimulation, which produces maximal tetanic tension development), by ∼95%, while cytosolic calcium ([Ca²⁺]_c) transients were not affected at the beginning of the contractile period. When tension development and [Ca²⁺]_c were analyzed during the contractile bouts under control conditions (no BTS), time to fatigue was 442 ± 23 s (Fig. 2A). Also, the changes in tension were accompanied by fall in peak [Ca²⁺]_c and calcium accumulated in the cytosol (i.e., increased basal [Ca²⁺]_c; Fig. 2B, closed symbols). However, at the same time points that fibers developed fatigue under control conditions, basal [Ca²⁺]_c in the BTS treatment was less (47 ± 5%, Fig. 2B; P < 0.05), and peak [Ca²⁺]_c was higher (120 ± 40%, Fig. 2B; P < 0.05). At 630 ± 33 s of contractions during the BTS treatment, both Ca²⁺ accumulation and fall in peak [Ca²⁺]_c reached similar values compared with the control fatigue time point (Fig. 2B).

Fig. 1.

Representative trace recordings for isometric tension (top) and [Ca²⁺]_c (bottom) from an intact fiber evoked by a 70-Hz stimulation. Note scale differences. The first contraction and the contraction after 380 s of repetitive stimulations are shown before (control) and after benzyl-p-toluene sulfonamide treatment (BTS). For this particular fiber, after 380 s, tension had decreased 50% from the initial contraction under control conditions. Bar indicates the period of electrical stimulation.

Fig. 2.

Intracellular Ca²⁺ transient responses during the repetitive contraction protocol in the absence of cross-bridge cycling. Peak isometric tension development at 70 Hz (A), basal [Ca²⁺]_c, and peak [Ca²⁺]_c responses (B) in the absence (closed symbols) and presence of BTS (open symbols); n = 5 fibers. *P < 0.05 BTS vs. control, two-way repeated-measures ANOVA; Values are expressed as means ± SE.

Ca²⁺ uptake kinetics and SR pumping function with and without cross-bridge cycling.

Figure 3 illustrates the time course of [Ca²⁺]_c decay after the end of the stimulation period for selected contractions and at the fatigue time point, for control conditions (Fig. 3A), or at the last contractions, for BTS conditions (Fig. 3B).

The [Ca²⁺]_c decay period (from 0 to 2.5 s after the end of stimulation) was fitted by using a double-exponential equation (see methods). This was performed to determine the fast-phase of intracellular Ca²⁺ decay (λ₁), which is a sum of several factors, such as Ca²⁺ uptake rate by SERCA, Ca²⁺ dissociation off-kinetics from troponin C, and Ca²⁺ loading rate by parvalbumin (35). At control, λ₁ was gradually slowed up to the fatigue time point, as demonstrated by the gradual decrease in λ₁ during the contractile bout (83 ± 6% decrease at the fatigue time point, Fig. 3C; P < 0.05). In the presence of BTS, the highest decrease in λ₁ was obtained by 480 s of contractions (30 ± 6% decrease, Fig. 3C; P < 0.05). However, the changes in λ₁ in the BTS conditions were significantly smaller compared with the control conditions starting at 240 s of contractions (P < 0.05; Fig. 3C).

To follow the changes in SR Ca²⁺ pumping function during the contractile period in the presence and absence of BTS, the tail of elevated [Ca²⁺]_c after the stimulation period was used. The tail [Ca²⁺]_c measurements were fitted to a double-exponential function, and this function was used to generate the rate of [Ca²⁺]_c decay during the tail (−d[Ca²⁺]_c/dt; see methods for details). The plots of the mean data for tail [Ca²⁺]_c with the mean data for rate of [Ca²⁺]_c decay are shown in Fig. 4A (for control condition) and in Fig. 4B (for BTS condition).

Fig. 4.

Analysis of SR Ca²⁺ pumping at different time points of the contractile bout in presence (control; A) and absence of cross-bridge cycling (BTS; B). The tail of [Ca²⁺]_c decay trace recordings (as shown in Fig. 3, A and B) were fitted with a sum of two exponentials, plotted against the rate of [Ca²⁺]_c decay (−d[Ca²⁺]_c/dt) at the same periods of time, and the SR Ca²⁺ pumping rate was determined from Eq. 4, assuming the power function (N) and the SR Ca²⁺ leak (L) as 4 and 5, respectively (as described in methods). Mean data from five individual fibers obtained from the first contraction (a), the 62nd contraction (b), and at the fatigue time-point (c; control), or at the last contraction with BTS (d). The curves from the mean data were fitted with Eq. 4, but N and L were not fixed. Inset: individual rates of SR Ca²⁺ pumping were normalized by the first contraction and compared between control and BTS conditions. †P < 0.05 control vs. BTS, two-way repeated-measures ANOVA; n = 5 fibers. Values are expressed as means ± SE.

At the control conditions, the rate of SR Ca²⁺ pumping (the parameter A in Eq. 4) was significantly reduced during the time course of the fatiguing contractile bout (in μM⁻³/s: 135 ± 42, 13 ± 3, and 7 ± 3, for first contraction, 62nd contraction, and at the fatigue time point, respectively, P < 0.01 vs. first contraction, one-way repeated-measures ANOVA). With BTS present, there was also a decrease in slowing in the rate of SR Ca²⁺ pumping (in μM⁻³/s: 114 ± 30, 48 ± 8, and 38 ± 13, for first contraction, 62nd contraction, and last contraction, 202 ± 30 contractions, respectively, P < 0.01 vs. first contraction, one-way repeated-measures ANOVA). However, comparing the changes in rate of SR Ca²⁺ pumping between the two conditions, the presence of BTS significantly attenuated the slowing in the rate of SR Ca²⁺ pumping (∼40% smaller slowing, Fig. 4B, inset; P < 0.05, vs. control, two-way repeated-measures ANOVA).

Ca²⁺ handling and decay kinetics in the presence of NaCN.

To evaluate whether the delayed changes in [Ca²⁺]_c when BTS was present were due to smaller accumulation of anaerobic metabolites, through a sufficient contribution of aerobic ATP supply when cross-bridge cycling was absent, mitochondrial respiration was inhibited by 2 mM NaCN (Fig. 5A). Peak [Ca²⁺]_c fell faster when NaCN + BTS were present compared with the BTS-alone condition (P < 0.05; Fig. 5A). However, basal [Ca²⁺]_c was not significantly modified when NaCN + BTS were present compared with BTS alone.

Fig. 5.

[Ca²⁺]_c transients, [Ca²⁺]_c decay kinetics, and SR Ca²⁺ pumping rate during contractions when mitochondrial respiration was blocked with NaCN. A: basal and peak [Ca²⁺]_c responses in the presence of BTS (○) and in the presence of BTS + NaCN (△). †P < 0.05 BTS + NaCN vs. BTS, two-way repeated-measures ANOVA. B: fast-phase rate of [Ca²⁺]_c decay (λ₁) in the presence of BTS (○) or BTS + NaCN (△). *P < 0.05 vs. initial contraction, one-way repeated-measures ANOVA. C: mean data of [Ca²⁺]_c dependence of the rate of [Ca²⁺]_c decline at different time points of the contractile bout in the presence of BTS (closed symbols) or BTS + NaCN (open symbols). n = 4 fibers. Values are expressed as means ± SE. Time points selected were the same as in Fig. 4 (indicated as a, b, and d). Rate of SR Ca²⁺ pumping was determined as described in methods and in Fig. 4. The curves from the mean data were fitted with Eq. 4, but N and L were not fixed.

The differences in the fast-phase [Ca²⁺]_c decay rate (λ₁) with BTS during the contractile bout were also analyzed in the presence of NaCN + BTS. Figure 5B shows that λ₁ was progressively slowed when NaCN + BTS were present (Fig. 5B; P < 0.05 vs. initial contraction).

SR Ca²⁺ pumping function was also analyzed during the contractile periods in presence of BTS or BTS + NaCN (Fig. 5, C and D, respectively). As occurred in the first group of fibers with BTS (Fig. 4B), the SR Ca²⁺ pumping was slowed in the beginning of the contractile period (i.e., 62nd contraction), but no further slowing was detected in the SR Ca²⁺ pumping in the following contractions (Fig. 5C). Interestingly with BTS + NaCN, there was also no additional slowing in SR Ca²⁺ pumping after the 62nd contraction, compared with the BTS alone condition (Fig. 5D).

pH_i changes.

The changes in pH_i were monitored during the contractile periods. As shown in Fig. 6, all control fibers showed the expected immediate alkalosis and subsequent acidosis induced by the work period (31). At the fatigue time point (390 ± 21 s), pH_i was significantly decreased to 6.37 ± 0.08 (Fig. 6A; P < 0.05). In the presence of BTS, no significant alkalosis was detected, and the pH_i was only statistically smaller from resting values by 540 s of contractions (P < 0.05). At the end of the contractile period with BTS (at 633 ± 3 s), pH_i was significantly decreased to 6.78 ± 0.05 compared with resting values.

Fig. 6.

Intracellular pH (pH_i) changes during repetitive contractions. A: pH_i was measured during the work periods in the presence (○) and the absence of BTS (●). *P < 0.05 vs. initial contraction, one-way ANOVA, repeated measures. **P < 0.05 BTS vs. control, two-way repeated-measures ANOVA; n = 5 fibers. B: pH_i responses in the presence of BTS (○) and BTS + NaCN (△). *P <0.05 vs. initial contraction, one-way repeated-measures ANOVA. †P < 0.05 BTS vs. BTS + NaCN, two-way repeated-measures ANOVA; n = 4 fibers. Dashed line represents the mean data of pH_i measurements under control conditions (no BTS and no NaCN) shown in A. Values are expressed as means ± SE.

To investigate whether the Ca²⁺ handling work during contractions was supplied by mitochondrial ATP turnover, mitochondrial respiration was blocked with NaCN. When both BTS + NaCN were present during the contractile protocol, there was a significant early alkalosis (P < 0.05), compared with the BTS condition alone (Fig. 6B). The presence of BTS + NaCN during contractions produced a significant reduction in pH_i compared with BTS alone by 420 s (Fig. 3B; P < 0.05) but significantly smaller than the control conditions (e.g., pH_i was decreased to 6.63 ± 0.03 after 600 s with BTS + NaCN compared with 6.38 ± 0.08 after 390 s under control, open triangles vs. dashed line, respectively, P < 0.05).

DISCUSSION

Our results demonstrate that inhibition of cross-bridge cycling during repetitive contractions results in enhanced maintenance of the SR Ca²⁺ pump function, possibly due to a reduced intracellular accumulation of H⁺. Specifically, BTS administration in contracting myocytes: 1) inhibited the contraction-induced slowing in both fast-phase [Ca²⁺]_c decay kinetics and SR Ca²⁺ pumping function (Figs. 3 and 4); and 2) abolished the initial intracellular alkalosis and diminished the delayed intracellular acidification (Fig. 6).

Intracellular Ca²⁺ responses during contractions.

The peak [Ca²⁺]_c transient and basal [Ca²⁺]_c during contractions are determined by the interplay of SR Ca²⁺ release and uptake and leak (2). At the contractile fatigue time point, the decrease in SR Ca²⁺ release has been shown to play a large role in decreasing peak [Ca²⁺]_c (1) and may be impaired by the accumulation of some metabolites [e.g., free Mg²⁺, ADP, IMP, ROS, P_i; (1, 20–22)]. The production rate of these metabolites is controlled by the energy requirements of the contractions, as well as the ability to supply ATP aerobically (20). In our current experiments, peak and basal [Ca²⁺]_c was 2.1 ± 0.3 × higher and 34 ± 5% smaller during BTS treatment, respectively, compared with control at 360 s of the contractile period (Fig. 2A). These changes have been reported in intact single mouse fibers (8) and were correlated to a diminished metabolite accumulation as a consequence of a smaller metabolic demand during contractions (8). In the present study, the basal [Ca²⁺]_c in the last contraction with BTS was very similar to the basal [Ca²⁺]_c obtained at the fatigue time point in the control condition. This result is different from the results obtained in single mouse fibers (8); however, single mouse fibers do not develop significant acidosis during repetitive contractions (32), and experimental protocols between the two studies were quite different (8). Although we cannot speculate which metabolites were diminished during the BTS condition, the data obtained from the [Ca²⁺]_c decay periods (discussed below in details) suggest that accumulation of free Mg²⁺ and ADP were likely reduced in the cytosol when the metabolic demand was diminished because of BTS. However, the intracellular accumulation of P_i would produce Ca-P precipitates within the SR, decreasing SR Ca²⁺ release and increasing SR Ca²⁺ leak (28). Therefore, it is reasonable to suggest that the decrease in energy cost of contractions with BTS would delay intracellular P_i accumulation, thereby delaying the changes in [Ca²⁺]_c with contractions.

Ca²⁺ pump function in intact single fibers.

It has been previously shown by our group and by others that blocking cross-bridge cycling during contractions produces a ∼60% decrease in total energy utilization, thereby implying that almost 40% of the total energy cost of contractions is due to SERCA Ca²⁺ pumping (5, 6, 22, 30). Although BTS inhibits actin-myosin interaction (23), it did not affect SERCA Ca²⁺ pumping function in the first contraction (see Fig. 4). Thus, this makes BTS treatment an excellent tool to study the kinetics of either SERCA or SR Ca²⁺ uptake function, particularly during fatigue development. During fatigue, the slowing in the relaxation period has been associated with either a reduction in cross-bridge dissociation rate or a decrease in SR Ca²⁺ pumping rate or both (for review, see Refs. 2 and 28). The present study evaluated the [Ca²⁺]_c decay rate without the confounding factors involved in cross-bridge cycling, such as the actin-myosin dissociation and the cross-bridge ATP demand. We demonstrated in the current study that the absence of cross-bridge cycling during contractions blunted the slowing in [Ca²⁺]_c decay (during the relaxation cycle of a contraction) detected in the control condition. Although this phenomenon has been reported in intact single mouse fibers (8), it has not been examined in a systematic fashion.

The initial [Ca²⁺]_c decay rate (λ₁) accounts for ∼90% and ∼20% of the relaxation time period during the unfatigued state and at the fatigue time point, respectively. Any changes in λ₁ have significant implications for the overall muscle relaxation. The λ₁ is strongly regulated by the parvalbumin-Mg²⁺ off-rate, and the increase in [Mg²⁺]_free slows this phase (15). BTS did not modify λ₁ during the unfatigued state, which suggests that parvalbumin-cation binding properties were not affected. During a repetitive contractile bout, λ₁ was progressively reduced under control conditions, but not with BTS present. Although [Mg²⁺]_free was not measured, the slowing in λ₁ when mitochondrial respiration and cross-bridge cycling were blocked together, suggests that [Mg²⁺]_free was either smaller or not increased in the absence of cross-bridge cycling due to sufficient ATP generation by mitochondrial respiration. Therefore, actomyosin ATP cost during repetitive contractions is fundamental for slowing the initial phase of relaxation.

The SR Ca²⁺ pumping function was measured during the prolonged elevated tail of [Ca²⁺]_c decay after the stimulation period (35). As expected, contractions after the onset of fatigue produced a progressive slowing in the SR Ca²⁺ pumping rate during control conditions, which was ∼40% less slowed by the last contraction with BTS (Fig. 4). Interestingly, SR Ca²⁺ pumping during contractions when BTS + NaCN were present was similarly slowed compared with BTS alone (Fig. 5, C and D).

The data from the present investigation show that the primary condition in which SERCA function was significantly inhibited during contractions was when intracellular acidification and cross-bridge cycling were present. There is a large body of literature, particularly performed in mechanically skinned single fibers (18, 28), on the negative effects of metabolite accumulation (e.g., high [ADP], [H⁺], [P_i], and [Mg²⁺]_free) on SERCA ATPase activity, SR Ca²⁺ leak, and SR Ca²⁺ uptake rate. Our data suggest that the ATP breakdown of cross-bridge cycling (together with the ATP cost of SR Ca²⁺ uptake) plays an important role in regulating intracellular metabolite accumulation, particularly H⁺, and these factors may ultimately modulate SERCA function during contractions.

The inhibition of SERCA function may also be due to nonmetabolic effects of fatigue. Several reports have shown, by using in vitro isolated SR vesicles and muscle homogenates to measure SERCA function under “nonfatiguing” conditions, that SR Ca²⁺ uptake rate and SERCA ATPase activity are persistently diminished (for hours) after fatiguing exercise (12, 29), suggesting potential protein structural changes in SERCA. However, the data from the present investigation show that SERCA function was restored rapidly between contractile periods (1-h rest periods), thereby suggesting that it was likely that the changes in SERCA function were due to the acute metabolic perturbations occurring with contractions, which were diminished in the absence of cross-bridge cycling.

Changes in intracellular H⁺ during repetitive contractions.

Intact myofibers, as whole muscle, develop marked intracellular pH (pH_i) changes during repetitive fatiguing contractions (31). Alkalosis occurs early during contractions and is then followed by acidosis (31), being driven by creatine kinase (CK) activity (H⁺ buffering) and activation of anaerobic glycolysis, respectively (31). The changes in pH_i detected in myofibers are the result of an equilibrium between H⁺ production by ATP hydrolysis and anaerobic glycolysis, intracellular H⁺ buffering, and the H⁺ extrusion from the cell (32).

The absence of cross-bridge cycling blocked the early alkalosis that occurred during the first 60 s of contractions in control conditions (Fig. 6A), indicating that the CK activity was reduced. These results suggest that the SERCA Ca²⁺-handling ATP requirements were not high enough to fully activate CK in the beginning of the contractile period. Reducing the energy cost of contractions by blocking cross-bridge cycling produced a smaller contraction-induced acidosis (Fig. 6A), possibly because of a smaller activation of anaerobic glycolysis (22). Comparing the changes in pH_i (Fig. 6A) with SERCA function (Fig. 4), SR Ca²⁺ pumping rate was significantly reduced after 360 s of contractions (62nd contraction), which was associated with ∼0.7 units of pH_i acidification. It is important to note that the acidification detected in the control condition (i.e., ∼pH 6.4–6.3) has been reported to inhibit SERCA activity in vitro (26, 38) and in vivo (34), as well as decreased relaxation rate in mammalian single fibers at low (i.e., 22°C) and high (i.e., 32°C) temperatures (37). As discussed above, the SR Ca²⁺ pumping rate was significantly less inhibited with BTS compared with the control, and pH_i was not changed. Furthermore, at the end of the contractile protocol with BTS (633 ± 3 s), there was a small, but statistically significant, acidification (∼0.26 units of pH), yet the SR Ca²⁺ pumping rate was not changed further. Interestingly, when NaCN + BTS were present, contraction-induced acidosis was significantly increased compared with BTS alone (∼0.4 units of pH decrease), but the SR Ca²⁺ pumping rate was not slowed compared with BTS alone. This suggests that there is possibly a threshold of intracellular acidification that produces SERCA Ca²⁺ pumping dysfunction, and in the current study, this was not achieved when cross-bridge cycling was inhibited. Therefore, reducing the energy cost of contractions by ∼60% avoided the contraction-induced acidification that is required to inhibit SERCA.

Perspectives and Significance

The results of this study demonstrate that the ATP cost of cross-bridge cycling during contractions is an important determinant of metabolite accumulation, particularly H⁺ in the cytosol and possibly P_i within the SR, during repetitive contractions and that these regulate peak [Ca²⁺]_c responses and SR Ca²⁺ pumping function. These changes will ultimately inhibit SERCA function during contractions, thereby slowing the relaxation period. These results provide significant information concerning the relationship between the energy cost of cross-bridge cycling and SR Ca²⁺ uptake with intracellular metabolite accumulation during repetitive contractions, as well as correlate the energy cost of contractions with the changes in SERCA function in skeletal muscle during fatigue development.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

L.N. and M.C.H. conception and design of research; L.N. and A.A.S. performed experiments; L.N. analyzed data; L.N., P.G.G., and M.C.H. interpreted results of experiments; L.N. prepared figures; L.N. and M.C.H. drafted manuscript; L.N., A.A.S., P.G.G., and M.C.H. edited and revised manuscript; L.N., A.A.S., P.G.G., and M.C.H. approved final version of manuscript.