INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

STRONG EVIDENCE INDICATES that the production of reactive oxygen species (ROS), including free radicals, is increased during strenuous exercise, chronic lung and heart disease, inflammation, and sepsis (1, 5, 6, 14, 18, 29) and may contribute to fatigue and weakness in these situations. However, the impact of ROS on contractions of intact skeletal muscle is not well understood. There is some indication that ROS may contribute to fatigue in isolated soleus (2) and diaphragm (29) preparations and to skeletal muscle lipid membrane damage (1). Paradoxically, Reid and colleagues (30, 31) have suggested that ROS may play a potentiating role in unfatigued skeletal muscle, perhaps through changes in excitation-contraction coupling. We recently reported (16) that exposure of unfatigued diaphragm fiber bundles from young adult rats to a ROS donor system [i.e., xanthine oxidase (XO)] potentiaties low-frequency contractions (30 Hz) without affecting maximal tetanic tension (P_o). These findings could indicate ROS-induced changes in diaphragm excitation-contraction coupling as proposed by Reid and Moody (31). However, more direct evidence is lacking.

High external levels of K⁺ (e.g., 130 mM) elicit a transient contracture as a result of membrane depolarization (13) and activation of voltage-dependent Ca²⁺ channels/receptors (25) found in the t-tubule membrane. The Ca²⁺ channels have been identified as dihydropyridine sensitive (8) and are thought to perform an important function in excitation-contraction coupling (25, 33) by triggering Ca²⁺ release from the sarcoplasmic reticulum (SR) (21). In mammalian muscle, K⁺ contractures are a useful probe of the "activation" and "inactivation" processes involved in excitation-contraction coupling (7, 8, 15). Therefore, if ROS affect skeletal muscle membrane excitability and/or excitation-contraction coupling, then the magnitude and dynamics (i.e., time-to-peak tension and relaxation) of K⁺ contractures could be dramatically altered. We hypothesized that application of ROS via XO would decrease K⁺ contracture time to peak and would increase K⁺ contracture peak tension in unfatigued diaphragm fiber bundles. Such findings would indicate a direct role of ROS on excitation-contraction coupling of the diaphragm.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Animals. Four-month-old Fischer 344 rats (n = 9) were housed and cared for in accordance with the American Physiological Society's policies. Rat chow and water were provided ad libitum; animals were maintained on a 12:12-h light-dark cycle. An animal use protocol was approved by the University Laboratory Animal Care Committee before the commencement of experiments.

In vitro preparation. Preparation of diaphragm fiber bundles was similar to that described by Reid et al. (29). Animals were anesthetized with 50 mg/ml pentobarbital sodium, and the whole diaphragm was then quickly excised (<30 s without blood flow) and immediately placed in ice-cold Krebs-Ringer solution aerated with a 95% O₂-5% CO₂ gas mixture. The Krebs-Ringer solution (pH adjusted to 7.40) contained the following (in mM): 118 NaCl, 4.7 KCl, 1.8 CaCl₂, 1.18 MgSO₄, 1.18 KH₂PO₄, 25 NaHCO₃, and 11 glucose. Two small fiber bundles were carefully removed from the lateral portion of the costal diaphragm with the central tendon and rib intact. The rib end of the fiber bundle was tied, by using silk 4-0 sutures, to a glass-rod holder, with another suture tied to the central tendon. Each fiber bundle was placed in a muscle bath (Harvard Apparatus) containing Krebs-Ringer solution at 36.0°C (pH = 7.40; bubbled with 95% O₂-5% CO₂), with the central tendon suture tied to an isometric force transducer (Grass FT03). Tension was processed by a direct-current amplifier and displayed on a chart recorder (Omnigraph 3000). After a 10-min equilibration period, the optimal length to achieve peak twitch tension was measured. Fiber bundles were stimulated by field generation through two platinum foil electrodes powered by a Grass stimulator (S48). All subsequent measurements were made at optimal length. Baseline twitch tension and P_o were then determined. Stimulus trains were 500 ms, pulse length = 2 ms, and frequency = 120 Hz. Stimulus strength was set at 1.5 times the voltage needed to produce peak twitch tension.

Experimental design. The content of the muscle bath medium served as the treatment. Two diaphragm fiber bundles were taken from the left and right lateral costal portion of each animal. Each fiber bundle was then randomly assigned to one of the following treatment groups: standard Krebs-Ringer solution without XO (Con group) or Krebs-Ringer solution + XO (XO group). Thus each diaphragm served as its own control. This also greatly minimized any potential differences in fiber bundle thickness and regional differences between treatments (37), which clarified interpretation of K⁺ contracture dynamics (13). ROS (primarily superoxide anion) were introduced by XO (0.01 U/ml; 1 mM hypoxanthine; Ref. 2). Production of superoxide anion in the bath medium by XO was verified by using cytochrome c reduction determined spectrophotometrically as described by Reid et al. (29). Samples were withdrawn from the muscle bath after 15 min (i.e., length of the experiments) and analyzed. Protease-free XO was purchased from Biozyme. All other chemicals were purchased from Sigma Chemical.

To determine the contractile function of the diaphragm, twitch tension and P_o (at 120 Hz) were determined in fiber bundles. After 5 min, a K⁺ contracture was induced with the introduction of 130 mM KCl, along with 1.8 mM CaSO₄, 1.18 mM MgSO₄, and 1.18 mM KH₂PO₄. The dynamics of the K⁺ contracture were determined by measuring the following: time-to-peak tension, peak contracture tension, and one-half relaxation time (RT_1/2) for the contracture. After the dynamics of the initial K⁺ contracture were taken (Pre), the fiber bundle was washed twice with normal Krebs-Ringer. Then the treatment solutions were introduced (Con or XO), and 15 min were allowed for equilibration. This allows enough time in control fiber bundles for recovery of P_o after the first K⁺ contracture (7). A second K⁺ contracture (Post) was elicited with 130 mM KCl, and contracture dynamics were again followed.

Additional experiments were performed (n = 6) with 1 mM hypoxanthine and 0.01% albumin added to the tissue medium to account for any potential effects of hypoxanthine on K⁺ contractures. The results were compared against fiber bundles with Krebs-Ringer only (n = 6) and 0.01 U/ml XO + 1 mM hypoxanthine in Krebs-Ringer (n = 6). We also performed experiments (n = 6) in which Ca²⁺ was absent from the K⁺ contracture-eliciting solution.

Statistics. One-way analysis of variance for repeated measures with Fisher's least significant difference, used post hoc where appropriate, was used to assess changes in K⁺ contracture dynamics in diaphragm fiber bundles: peak contracture tension, time-to-peak tension, and RT_1/2. The existence of mean differences for test-retest reliability was analyzed by using paired t-tests. Test-retest correlation coefficients were also calculated. Significance level was set at 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

P_o for our diaphragm fiber bundles averaged 24.2 ± 1.8 N · cm² (data not shown). Twitch tension averaged 21.1% of P_o. These results verify the patency of the preparation.

Test-retest reliability for K⁺ contracture time-to-peak tension, peak tension, and RT_1/2 was assessed in 10 additional pairs of diaphragm fiber bundles bathed in Krebs solution. No significant (P = 0.21) mean differences in time-to-peak tension for K⁺ contractures were calculated (18.8 ± 0.7 vs. 18.4 ± 0.9 s) for K⁺ contractures 15 min apart. A test-retest correlation of r = 0.88 (P < 0.001) also indicated good reproducibility. No significant mean differences (P = 0.44) in test-retest peak tension were found (9.0 ± 0.4 vs. 9.2 ± 0.5 g) for K⁺ contractures. Our values for peak K⁺ contractures represent 25% of P_o and are similar to those reported by others (5, 34) for the mammalian diaphragm. A test-retest correlation of r = 0.85 (P < 0.001) was determined for peak tension of K⁺ contractures. RT_1/2 for K⁺ contractures was less reproducible (r = 0.53); however, no significant differences (P = 0.22) in mean test-retest relaxation times (27.2 ± 1.9 vs. 25.1 ± 2.1 s) were seen.

Findings of the treatment experiments are as follows. Time-to-peak tension for K⁺ contractures was significantly decreased (P < 0.01) Post treatment in the XO group, from 19.8 ± 1.0 to 16.0 ± 0.7 s, but not in the Con group (Fig. 1). The time-to-peak tension values for K⁺ contractures were similar in both Con and XO Pre treatment groups. The effects of XO on the peak tension of K⁺ contractures are found in Fig. 2. Mean peak tension significantly increased (P < 0.001) Post in the XO group, from 5.56 ± 0.40 to 7.31 ± 0.41 N · cm², an increase of 31.5%. However, no significant alteration was found for mean peak tension Post in Con diaphragm fiber bundles. These findings indicate that XO treatment potentiates the dynamics of K⁺ contractures in fiber bundles. Cytochrome c reduction at 532 nm, a marker of superoxide anion generation, was significantly increased in XO-treated fiber bundles but not in the Con or hypoxanthine groups, indicating that ROS were increased in the XO group (32). This was verified by the application of 50 U/ml superoxide dismutase (n = 6), which inhibited 85% of XO-induced cytochrome c reduction.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Time-to-peak tension (in s) for K+ contractures in diaphragm fiber bundles pretreatment and stimulation (hatched bars; Pre) and posttreatment (solid bars; Post) for 2 treatment groups: normal Krebs-Ringer solution alone (Con) and Krebs-Ringer plus xanthine oxidase and purine substrate (XO). * Significantly different from Pre.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Peak contracture tension (in g) for K+ contractures in diaphragm fiber bundles pretreatment and stimulation (hatched bars; Pre) and posttreatment (solid bars; Post) for 2 treatment groups: Con and XO. * Significantly different from Pre.

No changes in RT_1/2 for K⁺ contractures were found in Krebs-Ringer diaphragm fiber bundles (data not shown). A trend (P = 0.094) toward a reduced RT_1/2 was observed in the XO group Post treatment and stimulation. However, this trend did not reach statistical significance for XO treatment. On closer inspection, the faster relaxation was only seen in the initial rapid phase following peak K⁺ contracture tension. We calculated the slope of the fall in tension (dP/dt) during the initial, fast phase of relaxation from a K⁺ contracture. These calculations revealed that dP/dt for the fast component of relaxation was significantly greater in magnitude following XO exposure than before exposure or in control diaphragm fiber bundles (Fig. 3). Tension of one-half of the diaphragms treated with XO did not return completely to the baseline levels after the Post K⁺ contractures (Fig. 4). Furthermore, the remaining fiber bundles treated with XO returned to baseline tension by the end of the K⁺ contractures but with a lesser rate of relaxation during the later, slower relaxation phase of the K⁺ contracture. Indeed, the calculated rates for dP/dt during the slow phase of relaxation were decreased with XO exposure (Fig. 3). Thus XO-induced oxidant challenge affected the two discrete phases of relaxation during K⁺ contractures in distinctly different manners.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Rates of relaxation (dP/dt) for diaphragm fiber bundles (in mN/s) during initial fast phase (A) and during following slow phase (B) for K+ contractures before (hatched bars) and after treatment (solid bars) with Control (Krebs-Ringer + 1 mM hypoxanthine) and XO (0.01 U/ml + 1 mM hypoxanthine). * Significantly different from Pre values. a Significantly different than Post Control.

View larger version (8K): [in this window] [in a new window]   Fig. 4.   Tracings of K+ contractures in diaphragm fiber bundles. Time and tension calibrations are indicated. A: typical K+ contractures in diaphragm fiber bundles Pre and Post Krebs-Ringer treatment only. B: K+ contractures in diaphragm fiber bundles Pre and Post XO treatment in an experiment when the Post K+ contracture did not return to baseline. C: K+ contractures in diaphragm fiber bundles Pre and Post XO treatment in an experiment when the Post K+ contracture returned to baseline. Note reduced slope of relaxation during the later, slower phase of relaxation.

The XO-induced effects on time-to-peak tension, the magnitude of peak tension, and relaxation were not affected by the absence of Ca²⁺ from the K⁺-eliciting medium (data not shown).

The dynamics of K⁺ contractures of diaphragm fiber bundles exposed to 1 mM hypoxanthine were not significantly different compared with Krebs alone (P = 0.742). Again, only XO exposure elicited an increase (+26.6%) in peak K⁺ contracture tension, with no increase with 1 mM hypoxanthine exposure (data not shown). In addition, K⁺ contracture time to peak was 4.2 s faster after XO treatment compared with 1 mM hypoxanthine alone.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Results from this investigation reveal a clear alteration in the dynamics of diaphragm K⁺ contractures with the introduction of a ROS-generating system. These are novel and exciting data that further illustrate the profound impact that ROS have on skeletal muscle function. Specifically, a number of principal findings can be drawn from our data: 1) K⁺ contractures are very reproducible in rat diaphragm fiber bundles; 2) control treatments (Krebs-Ringer alone or Krebs-Ringer + 1 mM hypoxanthine) did not alter peak tension, time-to-peak tension, or the relaxation phase of K⁺ contractures; 3) diaphragm K⁺ contractures have greater peak tensions and a more rapid onset after exposure to XO treatment; and 4) XO treatment also changes the relaxation phase of diaphragm K⁺ contractures by decreasing the initial rapid phase and prolonging the secondary slow phase. These data are currently the most direct evidence that ROS modulate excitation-contraction coupling in the intact mammalian diaphragm. A discussion of the main findings follows.

Here we have demonstrated that K⁺ contractures are very reproducible in the rat diaphragm. Therefore, any changes in the size and/or dynamics of K⁺ contractures in our model must be treatment related. Furthermore, treatment of diaphragm fiber bundles with Con or 1 mM hypoxanthine in Krebs-Ringer was similar and did not elicit any significant changes in K⁺ contracture peak tension, time-to-peak tension, or contracture relaxation. Thus the alterations in K⁺ contractures Post with XO treatment must be related to the exposure to ROS, as indicated by an increase in cytochrome c reduction.

The XO-induced decrease in time-to-peak tension and increase in peak tension of K⁺ contracture represent an enhanced activation of excitation-contraction coupling by ROS. These findings are consistent with those of Puppi et al. (26), who reported that treatment of frog rectus abdominus muscle with an oxidant (thionine) increases the redox-state potential and significantly increases the magnitude of K⁺ contractures. Indeed, the shape of our K⁺ contractures (Fig. 4) is very similar to that illustrated in single frog fibers by Puppi and colleagues (see Fig. 1 in Ref. 26). Oxidation in those experiments was associated with K⁺ contracture potentiation, whereas reduction resulted in a depressed K⁺ contracture. We have previously found that XO potentiates twitch and low-frequency tetanic tension, without a change in P_o, in unfatigued diaphragm fiber bundles from young Fischer 344 rats (16). Hu and Lawler (13) recently found that 0.01 and 0.02 U/ml XO exposures increase low-frequency tension in the diaphragm by 15 and 30%, respectively; however, higher doses (0.05 U/ml) with longer exposure times (e.g., 60 min) depress both low-frequency contractility and P_o. Reid et al. (30) noted small (<10%) but statistically significant increases in diaphragm twitch tension with hydrogen peroxide (100 µM to 1 mM) exposure. However, higher concentrations of hydrogen peroxide (10 mM) resulted in an irreversible depression of diaphragm contractile function. Reid and Moody (31) reported that dimethyl sulfoxide, a preferential scavenger of hydroxyl radicals, depresses submaximal tension development of the diaphragm in a dose-dependent and reversible manner. Reid and colleagues (30) also reported similar results when treating diaphragm fiber bundles with the antioxidants superoxide dismutase and catalase. In total, these data suggest that ROS modulate excitation-contraction coupling in unfatigued skeletal muscle in a manner dependent on the redox state of critical proteins in excitation-contraction coupling (10, 12, 15, 28, 31). The response of the intact diaphragm to ROS may also depend on a number of factors, including the contractile or fatigue state, glutathione status, type of ROS, and age (16, 17).

A direct effect of ROS on sarcolemmal and t-tubule membranes has been postulated by Shamsadeen and Duncan (36). However, disruption of normal membrane function would decrease, not increase, the response to K⁺ contractures of the unfatigued diaphragm (Figs. 1 and 2). Furthermore, our results using a Ca²⁺-free contracture medium are inconsistent with sarcolemmal Na⁺/Ca²⁺ exchange (26) playing a significant role. In contrast, ROS could directly increase the sensitivity of dihydropyridine (DHP)-sensitive voltage sensors to K⁺-induced depolarization and thus impact on excitation-contraction coupling (28). Indeed, DHP antagonists, such as nifedipine and verapamil, greatly attenuate K⁺ contractures (11). In skeletal muscle, DHPs are located in the t tubules with ~2% acting as L-type Ca²⁺ channels. Sulfhydryl groups on the DHPs are proposed to be a target of ROS action (20, 24), resulting in increased Ca²⁺ charge currents (20) that would induce the SR to release Ca²⁺ stores (23-25, 33). The XO-induced increases in diaphragm peak K⁺ contracture tension and faster time-to-peak tension in the present investigation are consistent with these postulates (Figs. 1 and 2). The initial fast phase of relaxation is not independent of changes in the rising phase of the K⁺ contracture (9). Here, continued depolarization results in DHP voltage sensors becoming inactivated or undergoing a refractory period as charge movement is reduced. It is also evident that XO treatment affected the slower phase of relaxation for K⁺ contractures, which is consistent with a lack of closing or slower inactivation of DHPs and communication with Ca²⁺-release channels in the SR (9). Puppi et al. (26) reported similar changes on the relaxation phase of single frog fibers when using the oxidant thionine.

Exposure to ROS could also increase the responsiveness of the SR to K⁺ contractures. Scherer and Deamer (35) initially described increases in Ca²⁺ permeability of SR vesicles with oxidizing agents. Favero et al. (10) recently reported that H₂O₂ augments Ca²⁺ release from SR vesicles. Furthermore, hyperreactive sulfhydryl groups, sensitive to redox state, have been proposed to be important in the opening and closing of ryanodine-sensitive Ca²⁺ release channels in the terminal cisternae of the SR (19). These findings may be consistent with the larger peak tensions for K⁺ contractures in XO-treated fiber bundles (Figs. 1 and 2). However, these factors cannot explain the ROS-induced changes in the fast and slow relaxation phases of diaphragm K⁺ contractures. The ROS-induced changes in the slower phase of relaxation (Figs. 3 and 4) of diaphragm K⁺ contractures could represent a continued "leakage" of ryanodine-sensitive Ca²⁺ channels. However, this possibility can be discounted easily as maximal caffeine contractures are still possible during the relaxation phase (early or late) of K⁺ contractures (9), reflecting the maintained responsiveness of SR Ca²⁺-release channels. Our data then are consistent with the notion that temporal changes in the rising and fast relaxation phases of K⁺ contractures are properties of DHP voltage sensors (9, 15). This conclusion is consistent with the results of Oba and Yamaguchi (24), who found that the sulfhydryl reagent N-(7-dimethylamino-4-methylcoumarinyl)maleimide inhibited twitch tension and Ag⁺ contractures, but not caffeine contractures, implicating components of excitation-coupling upstream from the SR Ca²⁺-release channels.

There is some evidence suggesting that ROS mediate oxidation of sulfhydryl groups critical to SR Ca²⁺-ATPase function with increased work demand (5). A depressed resequestering of Ca²⁺ by the SR could result in an increased peak tension of the K⁺ contractures. XO-treated fiber bundles showed a decreased rate of relaxation in the late, slow phase. These results may also indicate reduced activity of SR Ca²⁺-adenosinetriphosphatase (ATPase) pumps that resequester Ca²⁺. However, disruption of SR Ca²⁺-ATPase activity results in accumulation of intracellular Ca²⁺ concentration and a concomitant rise in resting tension. We have not seen any changes in baseline tension during XO exposure of this dosage (e.g., 0.01 U/ml). Furthermore, twitch RT_1/2 is not prolonged by XO exposure in the unfatigued diaphragm (unpublished observations). Thus these findings are inconsistent with a disruption of SR Ca²⁺-ATPase activity in our model.

It is possible that XO treatment could change the maximal response or sensitivity of contractile proteins. Røed (34) proposed that sulfhydryl groups at the actin-myosin binding sites influence diaphragm rigor induced by elevated intracellular Ca²⁺. However, we have shown that oxidative stress imposed by XO does not change P_o in diaphragm fiber bundles from young rats (16, 17). Redox change in the contractile apparatus might increase peak K⁺ contracture tension but could not alter time-to-peak tension or relaxation dynamics. However, it would be expected that oxidation of sulfhydryl groups on myosin heads would inhibit K⁺ contractures (26) instead of the potentiation observed in the present investigation. Furthermore, the decline in tension with K⁺ contractures was a direct function of a decrease in intracellular Ca²⁺ concentration (4) and thus was not related to the contractile apparatus (9). Finally, Brotto and Nosek (3) found that hydrogen peroxide does not alter tension of skinned single muscle fibers. Therefore, it is unlikely that contractile proteins contributed to the changes in diaphragm K⁺ contractures seen currently with XO exposure.

In conclusion, ROS produced via the XO system alter the dynamics of K⁺ contractures in the diaphragm. Indeed, it is evident from our data that XO exposure affects both the activation and inactivation processes of K⁺ contractures. Our data are consistent with the notion that ROS modulate DHP-sensitive voltage sensors and excitation-contraction coupling, as hypothesized by Oba et al. (23). These are novel and exciting data, which further exemplify the impact of ROS on the ability of intact skeletal muscle to generate tension.