INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE CA²⁺-ATPASE OF SARCOPLASMIC RETICULUM (SR) catalyzes the ATP-dependent translocation of Ca²⁺ from the cytoplasm into the lumen of the SR and maintains the resting cytoplasmic free Ca²⁺ ([Ca²⁺]_f) at levels below 100 nM. In skeletal muscle, the SR Ca²⁺-ATPase plays an important role in excitation-contraction coupling by replenishing SR Ca²⁺ stores and allowing Ca²⁺ to be available for release on repeated activation.

The Ca²⁺-ATPase is a 110-kDa polypeptide chain with known sequence and structure of 997 amino acid residues (20, 37). The large cytoplasmic extramembrane portion of the enzyme (headpiece) contains the phosphate acceptor Asp-351 residue (1, 20) and the binding site for ATP (36). Other functionally important residues for binding and translocation of Ca²⁺ are oriented in two sites among transmembrane helixes M4, M5, and M6 (28). The Ca²⁺-ATPase contains 24 cysteine residues (1). Consequently, the SR Ca²⁺- ATPase may be a principal target for modulation of muscle function by reactive oxygen species (16). Inactivation of the SR Ca²⁺-ATPase as a result of free radical formation has been demonstrated in numerous in vitro studies (23, 33, 40, 41). Free radicals may cause protein denaturation and damage involving the ATP binding site (41), enhanced lipid peroxidation (15, 23), and increased protein oxidation, leading to both a reduced sulfhydryl group content and increased protein aggregation (23, 33, 40).

It is well established that free radical formation plays a primary role in the etiology of ischemia-induced damage in skeletal muscle (30). With prolonged muscle ischemia, a significant loss of ATP and total adenine nucleotides occurs (2, 12, 29, 35). This leads to elevations in substrates for the enzyme xanthine oxidase, which catalyzes the reaction where both superoxide radicals and hydrogen peroxide are formed in muscle (25).

Although free radical formation would be expected during extended ischemia, it remains unclear whether the disturbance in Ca²⁺ regulation occurs during the ischemic period per se or whether ischemia simply predisposes the muscle, with the actual changes becoming manifest during reperfusion. Unexpectedly, we have found a time-dependent increase in maximal Ca²⁺-ATPase activity with ischemia in both rat soleus and extensor digitorum longus muscle (8). Because the effects were only demonstrated in homogenate preparations, it is unclear what is happening at the level of enriched SR vesicle preparations, which are free from the potential contaminating effects of other proteins and cellular ATPases. Interestingly, reductions in Ca²⁺-ATPase activity have been found after ischemia in vesicle preparations from cardiac muscle (27).

Studies of ischemia-reperfusion in cardiac muscle, and more recently in skeletal muscle, have shown that impaired SR function and Ca²⁺ homeostasis may also be involved in the etiology of ischemia-reperfusion injury (9, 14, 19). In one study, pretreatment with the oxygen free radical scavengers superoxide dismutase and catalase maintained higher Ca²⁺ uptake by the SR of skeletal muscle after 3 h of ischemia and 19 h of reperfusion in rat hindlimb (19). Thus free radical formation is likely a mechanism for impaired SR function with ischemia and reperfusion.

Because measurement of SR function in ischemia studies is generally done in vitro, under optimal conditions, ischemia-induced reductions in Ca²⁺-ATPase activity and Ca²⁺ uptake are probably due to structural alterations to the Ca²⁺-ATPase protein and/or the SR membrane. However, the nature or precise location of altered structure on the SR Ca²⁺-ATPase has not been determined in skeletal muscle samples harvested from tissue that was made ischemic in vivo.

In this study, we employed a 4-h hindlimb ischemia model in rats to characterize the alterations in SR Ca²⁺-ATPase activity and structure that occur with prolonged periods of skeletal muscle ischemia. To assess the structural alterations to the Ca²⁺-ATPase, the fluorescent probes FITC and N-cyclohexyl-N'-(dimethylamino--naphthyl)carbodiimide (NCD-4) were used to measure changes in the regions of the ATP binding site and the Ca²⁺ binding sites of the Ca²⁺-ATPase, respectively. We hypothesized that Ca²⁺-ATPase activity would be lower in ischemic SR compared with control and that the lower Ca²⁺-ATPase would be accompanied by a reduction in FITC binding, indicative of structural alterations to the ATP binding site.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal Description and Care

Experimental Protocol

Before the induction of ischemia, the rats were weighed and anesthetized. Anesthesia was initially accomplished by using an intraperitoneal injection of pentobarbital sodium (6 mg/100 g body wt) and then was maintained by using supplementary intraperitoneal injections as required. To prevent dehydration, experimental rats were administered 2 ml of saline by injection just under the skin before the induction of ischemia. Throughout the ischemic periods, body temperature was maintained between 37 and 39°C by having the rats lay in a prone position on a warm heating pad. At the end of each of the ischemic periods, a small piece of white gastrocnemius (G) muscle was rapidly sampled from each of the I and C limbs and frozen in liquid nitrogen for later analysis of muscle metabolites. The remainder of the G muscle (both red and white portions) and the entire tibialis anterior (TA) muscle were excised and placed in ice-cold buffer to be used for SR isolation by differential centrifugation. The hindlimbs were not reperfused before muscle sampling. The G and TA muscles from the CC animal were sampled and excised in the same manner immediately after anesthetization.

Analytic Procedures

Sample preparation for SR assessment in vitro. Ischemic and control muscles were prepared according to Heilmann et al. (11). Mixed G and TA muscles were diluted ~1:5 (wt/vol) in ice-cold homogenizing buffer containing (in mM) 5 HEPES (pH 7.5), 250 sucrose, 0.2% sodium azide, and 0.2 phenylmethylsulfonyl fluoride (PMSF) and mechanically homogenized with a polytron homogenizer (PT 3,100) at 16,500 rpm, for two 30-s bursts. An aliquot of this sample was quick frozen in liquid nitrogen, stored at 70° to 80°C and used for homogenate determinations of Ca²⁺-ATPase activity. To obtain an enriched SR membrane fraction, a combination of two SR isolation protocols was used (6, 11). The homogenate was centrifuged at 5,500 g for 10 min to remove cellular debris, and the supernatant was filtered through four layers of gauze to remove as much fat as possible. The supernatant was then transferred to clean tubes and centrifuged at 12,500 g for 18 min. These pellets were discarded, and the spin was repeated. Again, the supernatant was transferred to clean tubes and centrifuged at 50,000 g for 52 min. These pellets were resuspended in 10 ml homogenizing buffer plus 600 mM KCl and allowed to incubate at 4°C for 30 min. This suspension was then centrifuged at 15,000 g for 10 min to pellet all the mitochondria. The supernatant was centrifuged at 50,000 g for 52 min. The final pellet, enriched in SR membranes (no sucrose cushion), was resuspended in homogenizing buffer at a protein concentration of 2-6 mg/ml. SR isolation was carried out by differential centrifugation by use of a Beckmann ultracentrifuge with a 70.1 Ti fixed-angle rotor. The SR membrane fraction was used for measurements of Ca²⁺-ATPase activity and isoform composition and to examine structural alterations to the enzyme. All homogenates and SR membrane isolations were made with only PMSF as a proteolytic inhibitor.

SR Ca²⁺-ATPase activity measurements. Spectrophotometric (Schimadzu UV 160U) measurement of SR Ca²⁺-ATPase activity was performed on homogenates by using procedures developed by Simonides and van Hardeveld (34) and SR samples analyzed according to methods of Leberer et al. (18) with minor modifications. Total (Mg²⁺-activated)-ATPase activity was measured in the presence of the Ca²⁺ ionophore A-23187, across a range of CaCl₂ concentrations. Basal activity was measured in the presence of 40 µM cyclopiazonic acid, which completely inhibits SR Ca²⁺-ATPase activity (32). The difference between total and basal activities represents the Ca²⁺-activated ATPase activity. Maximal activity and the Ca²⁺ dependency of Ca²⁺-ATPase activity were assessed by adding 1-11 µl of 100 mM CaCl₂ in 0.5-µl additions. Ca²⁺-ATPase activity increases with [Ca²⁺]_f until a plateau occurs, once maximal activity is reached. The [Ca²⁺]_f corresponding to each CaCl₂ addition was assessed separately, on a different SR aliquot, by using dual-wavelength spectrofluorometry and the Ca²⁺-fluorescent dye indo 1. Ca²⁺-ATPase activity was then plotted against the negative logarithm of [Ca²⁺]_f (pCa), and the Hill coefficient and the [Ca²⁺]_f that gives half-maximal activity (Ca₅₀) were determined. These properties were measured through nonlinear regression with computer software (Graph Pad Software) using the following sigmoidal dose-response equation
where Y_bot is the value at the bottom of the plateau, Y_top is the value at the top of the plateau, log Ca_1/2 is the concentration that gives a response halfway between Y_bot and Y_top, and n_H is the Hill coefficient. For calculation of these properties, only a portion of the curve that corresponded to between 20 and 80% of maximal activity was used. Total protein was determined by the method of Lowry as modified by Schacterle and Pollock (31). All Ca²⁺-ATPase activities are expressed relative to total protein content. On a given analytical day, samples from all conditions were analyzed in duplicate.

SDS-PAGE and Western blotting. A suspension of 0.5 mg/ml SR protein in either 40 µl reducing buffer [1.25 M sucrose, 0.1 M dithiothreitol (DTT), 0.25 M Tris · HCl, pH 6.8, 5% SDS, 0.01% bromophenol] or nonreducing buffer (1.25 M sucrose, 0.25 M Tris · HCl, pH 6.8, 5% SDS, 0.01% bromophenol) brought to 200 µl by distilled water was heated for 10 min at 100°C, and solid DTT was added to the sample in reducing buffer to raise the final concentration of DTT to 100 mM. All samples were then sonicated for 10 s in a probe sonicator, and 5 µg of each sample was analyzed in duplicate on separate 7% polyacrylamide SDS gels (BIO-RAD Mini-PROTEAN II) with a 3.75% stacking gel.

Fluorescence measurements. Fluorescence measurements were made on an SLM-4800S spectrofluorometer (SLM Instruments, Urbana, IL) according to Lalonde et al. (17). FITC (Sigma Chemical) and NCD-4 (Molecular Probes) were stored at a concentration of 5 mM in ethanol at 20°C. FITC emission spectra (500-550 nm) were collected by exciting samples at 490 nm (see Fig. 3A). FITC labeling was done by washing the SR samples once in wash buffer (homogenizing buffer without sucrose) with no DTT, then resuspending the samples in FITC labeling buffer (wash buffer plus 2.5 µM FITC, pH 8.8) and vortexing gently in darkness for 20 min at 25°C. The SR samples were then washed again in wash buffer to remove unbound label. NCD-4 emission spectra were collected by exciting samples at 340 nm and scanning the emission intensity from 400 to 430 nm at 1-nm increments. NCD-4 labeling was done by washing the SR samples once in wash buffer with no DTT, then resuspending in NCD-4 labeling buffer (wash buffer plus 150 µM NCD-4, pH 6.2) and incubating in darkness for 3 h at 25°C. As before, the sample was washed to remove unbound label.

Muscle metabolites. Muscle metabolites were measured on samples of stored tissues (80°C) that were freeze dried and cleared of visible connective tissue and blood. After extraction of the metabolites by perchloric acid (0.5 M) and neutralization, the contents of phosphocreatine (PCr), creatine, and lactate were assessed by using fluorometric procedures according to techniques previously published by our laboratory (10). A portion of the extract was also used for the determination of ATP, ADP, AMP, IMP, inosine, and hypoxanthine by using ion-pair, reversed-phase HPLC techniques (13) as described earlier (10).

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ca²⁺-ATPase Activity

1

1

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Kinetic analysis showing the Ca2+ concentration ([Ca2+]) dependence of Ca2+-ATPase activity of control control (CC, ), control (C, ), and ischemic (I, ) sarcoplasmic reticulum (SR) vesicles prepared from mixed gastrocnemius and tibialis anterior muscles. Curves were fit by nonlinear regression by using the average values of the Hill coefficient and cytoplasmic free Ca2+ that gives half-maximal activity (Ca50) from Table 2 for each group. Maximal activity was expressed relative to CC. Compared with CC, maximal activity was lower (P < 0.05) in I, but there were no differences (P > 0.05) in Hill coefficient or Ca50 between groups.

Electrophoresis and Western Blot Analysis

View larger version (69K): [in this window] [in a new window]   Fig. 2.   Western blot analysis of Ca2+-ATPase protein with use of the SERCA1 monoclonal antibody IIH11 for a sample from CC, C, and I groups. Under nonreducing conditions (A), only 2 main bands were indicated, corresponding to the monomer form of SERCA1 (110 kDa) and a smaller molecular weight (~70 kDa) SERCA1 product. Under reducing conditions (B), 6 discrete bands were identified, corresponding to SERCA1 monomer and smaller molecular weight products.

To determine whether prolonged ischemia results in SERCA1 protein oxidation and protein aggregation, SR samples from all groups were subjected to nonreducing (no DTT) and reducing (DTT) SDS-PAGE. Under nonreducing conditions, only two main bands were detected, corresponding to the monomer form of SERCA1 (110 kDa) and a smaller molecular weight (~70 kDa) SERCA1 product (Fig. 2A). Under these conditions, the intensity of the band corresponding to the monomer form of SERCA1 was lower (P < 0.05) in I relative to CC but not to C (Fig. 3A). There were no differences between CC and C. No differences were found in relative intensity of the 70-kDa bands between groups (Fig. 3B). Similarly, when the same two bands were analyzed for relative intensity under reducing conditions, there were no differences for either band between groups. This finding suggests that higher molecular weight aggregates were present in ischemic SR, leading to a lower intensity of the monomer form of SERCA1 under nonreducing conditions but not under reducing conditions.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Relative SERCA1 protein content in CC, C, and I SR vesicles corresponding to monomer form (110 kDa; A) and a smaller SERCA1 product (~70 kDa; B). Protein contents were determined by using nonreducing (no dithiothreitol; open bars) and reducing (dithiothreitol; solid bars) SDS-PAGE and Western blot analysis. Values are means ± SE; n = 9. *Significantly different (P < 0.05) from CC.

Fluorescence Measurements

A sample emission spectrum of FITC from each group is shown in Fig. 4A. The average maximum intensity, which occurs at 520 nm, was 26% lower in I (P < 0.05) compared with both CC and C (Fig. 4B). There was no difference between CC and C. The maximum emission intensity of NCD-4 was 0.36 ± 0.2, 0.36 ± 0.2, and 0.34 ± 0.2 for the CC, C, and I groups, respectively. No differences were found between any of the groups.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Sample FITC emission spectra, 1 each from CC (), C (), and I () SR vesicles (A) and average maximum FITC fluorescence for each group (B). Values are means ± SE; n = 9. Maximum FITC fluorescence occurs at 520 nm. Significantly different (P < 0.05) from CC and C.

Ischemia resulted in extensive changes in the metabolic status of the muscle (Table 2). For the metabolites measured, ischemia induced near-complete depletion of ATP and PCr. These changes were accompanied by large relative increases in IMP, inosine, and hypoxanthine. In contrast, ADP, but not AMP, was reduced with ischemia. Total creatine was not different in any of the groups. In addition, lactate content was elevated ~12-fold over control muscles. No differences were found for any of the metabolites between the CC and C groups.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In summary, we have found that 4 h of ischemia resulted in a reduction in maximal Ca²⁺-ATPase activity in SR vesicles that was accompanied by a reduction in FITC but not NCD-4 binding. The reduction in maximal Ca²⁺-ATPase activity was not accompanied by changes in either the Hill coefficient or pCa_50. Interestingly, no changes in maximal Ca²⁺-ATPase activity were observed with ischemia in homogenates.

Differences Between Homogenates and SR Fractions

Structural Changes in Ca²⁺-ATPase

The reductions in maximal Ca²⁺-ATPase activity that we observed in this study in ischemic SR compared with CC and C paralleled the reductions in FITC binding that occurred with ischemia. On the other hand, kinetic analysis of Ca²⁺-ATPase activity shows that both the Hill coefficient and sensitivity of the enzyme to Ca²⁺ (Ca₅₀) were unaffected. These results are consistent with a problem in binding ATP, but only in a selected population of Ca²⁺ pumps. A similar effect has been postulated to occur in rabbit muscle after chronic muscle activity (18). The reduction in FITC binding is not due to reductions in Ca²⁺-ATPase protein as determined by Western blot analysis. Rather, it appears that a reduction in the number of FITC binding sites per Ca²⁺-ATPase protein occurred in response to the prolonged ischemia.

We did probe for structural alterations in the region of the Ca²⁺ binding domain of the Ca²⁺-ATPase, by using the fluorescent probe NCD-4. We found that ischemia had no effect on the maximum binding capacity of NCD-4, suggesting that the structure of the Ca²⁺ binding domain was unaltered after prolonged ischemia. This is in close agreement with our kinetic analysis of ATPase activity because the Hill coefficient, which theoretically represents the number of Ca²⁺ binding sites, was not different between groups. However, it must be emphasized that the NCD-4 probe not only binds near to Ca²⁺ binding sites but displays some nonspecific binding as well (3).

Potential Role of Free Radicals

Our results indicate that the structural changes in Ca²⁺-ATPase with ischemia are similar to those reported after exposure of the Ca²⁺-ATPase to oxidizing conditions in vitro. In the in vitro study, it was found that exposure of the Ca²⁺-ATPase to levels of hydroxyl radicals similar to that measured during postischemic reperfusion denatures the Ca²⁺-ATPase and inhibits ATPase activity by directly attacking the ATP binding site without damaging the primary structure of the enzyme (41). This would suggest that prolonged ischemia also leads to structural reorganization of the region of the nucleotide binding domain, which would likely impair ATP binding and ATPase activity.

In the in vitro study, it was also found that presaturation of the active site with ATP completely protected both cardiac and skeletal muscle SR Ca²⁺-ATPase function from hydroxyl radical-induced inhibition. This suggests that depletion of cellular ATP, in the region of the enzyme, induces the structural alteration to the nucleotide binding domain of the Ca²⁺-ATPase. In our study, depletion of muscle ATP stores, as observed with prolonged periods of skeletal muscle ischemia, not only suggests formation of free radicals in skeletal muscle but may also indicate the susceptibility of SR Ca²⁺-ATPase to free radical-induced damage, specifically to the nucleotide binding domain.

The precise nature of the structural modification to the nucleotide binding domain in ischemic SR, which resulted in a reduced FITC binding capacity, cannot be directly ascertained from this study. However, several in vitro studies have helped to characterize the free radical-induced molecular modification of the SR Ca²⁺-ATPase that is correlated with its functional properties (33, 39). Viner et al. (39) employed a peroxyl radical-generating system by using the free radical initiator 2,2'-azobis(2-amidinopropane) dihydrochloride to examine and identify oxidation-sensitive peptides within the SR Ca²⁺-ATPase of fast-twitch rabbit skeletal muscle. They identified six oxidatively sensitive peptides on the cytoplasmic side of the SR membrane. One of these peptide segments (Glu₅₅₁-Arg₆₀₄) is located in the nucleotide binding domain and was found to participate in the formation of intermolecular bityrosine cross-links with the identical Glu₅₅₁-Arg₆₀₄ peptide from a neighboring Ca²⁺-ATPase polypeptide chain. Although this peptide does not contain the FITC binding site Lys₅₁₅ (4) or the actual ATP binding site around Arg₆₀₄ (36), cross-linking and aggregation in this region may interfere with both FITC and ATP binding. Whether cysteine residues are affected by prolonged ischemia, as might be expected if aggregation occurred, was not assessed in this study.

As reported, we detected smaller molecular weight SERCA1 fragments by using reducing SDS-PAGE and Western blot analysis. However, the relative amount of each fragment was not different between groups (data not shown). Likely, the appearance of these products was due to proteolytic activity and/or oxidative fragmentation that took place during homogenization and preparation of SR vesicles, and this should be the same for all groups. Our homogenization buffer only included one proteolytic inhibitor (PMSF) and did not include the sulfhydryl reducing agent, DTT. The decision not to use DTT was purposeful because we have reported that even 5 mM DTT in the homogenization buffer can alter the effects of 4 h ischemia on SR Ca²⁺-ATPase activity (38).

Although aggregation of Ca²⁺-ATPase in ischemic SR compared with control is suggested, as indicated by the lower relative monomer form of the protein in I, the gels were not supportive. The difference in monomer levels was corrected in reducing gels. This suggests that aggregation did occur with ischemia and was due to disulfide cross-links. A similar finding was reported by Senisterra et al. (33), with exposure of the SR Ca²⁺-ATPase to the thiol-specific reagents diamide and arsenite.

In the present study, we used an ischemia model in rat skeletal muscle known to lead to elevations in oxygen free radical production and reductions in SR Ca²⁺-ATPase activity to assess the structural alterations associated with in vivo oxidation of the SR Ca²⁺-ATPase. We have shown that prolonged ischemia leads to reductions in FITC binding capacity in isolated SR preparations, which is associated with reductions in maximal Ca²⁺-ATPase activity. We suggest that the molecular mechanism is likely oxidation of one or more of the cysteine residues within the nucleotide binding domain of the Ca²⁺-ATPase by xanthine oxidase-produced superoxide and/or H₂O₂. Whether these structural alterations are manifested in vivo remains to be demonstrated.