MDL-28170, a membrane-permeant calpain inhibitor, attenuates stunning and PKCε proteolysis in reperfused ferret hearts (original) (raw)

Abstract

Objectives: This paper tests the hypothesis that calpains are activated in the ischemic (I)/reperfused (R) heart and contribute to myocardial stunning. Methods: Isolated ferret hearts were Langendorff perfused isovolumically, and subjected to 20 min of global I followed by 30 min of R in the presence or absence of 0.2 _μ_M MDL-28170, a membrane-permeant calpain inhibitor. Right trabeculae then were isolated from these hearts, skinned chemically, and pCa2+-force curves obtained. Samples of left ventricle were extracted, subjected to SDS-PAGE, and Western analyzed for PKCε and PKMε. Results: Perfused ferret hearts exhibit a 43% decline in left ventricular developed pressure during R. Pre-treatment of hearts with MDL-28170 prior to I significantly improves function during R. Trabecular myofilaments from normal hearts have a KD for Ca2+ of 6.27±0.06; I/R decreased the KD to 6.09±0.04; trabeculae from I/R hearts pre-treated with MDL-28170 have a KD of 6.28±0.04. Western analysis shows ferret hearts to contain a single ≈96 kDa species of PKCε. I/R hearts contain the native PKCε and a ≈25 kDa smaller species of PKCε, which corresponds to PKMε, the calpain proteolyzed form of PKCε. Pre-treatment of I/R hearts with MDL-28170 markedly diminishes PKMε in reperfused hearts. Conclusions: Mechanical stunning during R is sensitive to MDL-28170. Depressed mechanical function is reflected in a hyposensitization of trabecular myofilaments to Ca2+. Western analysis shows that PKMε is present in R hearts.

Time for primary review 26 days.

1 Introduction

A reversible depression in myocardial contractile performance occurs during reperfusion following brief periods of severe ischemia [3]. The molecular mechanism(s) mediating this ‘myocardial stunning’ remains unresolved. Increases in cytosolic Ca2+ are observed during both ischemia and reperfusion [12, 15, 19, 29]. It has been hypothesized that increases in cytosolic Ca2+ play a role in the pathogenesis of ‘stunning’ by activating Ca2+-dependent processes in the ischemic/reperfused heart.

Calpains, Ca2+-activated neutral proteinases [7], could be one such Ca2+-dependent process activated during ischemia/reperfusion. There are two calpain isozymes, designated mu- and m-calpain, which are ubiquitous non-lysosomal intracellular cysteine proteinases [7, 22]. Despite some uncertainty [6], it is generally accepted that both calpains exist as pro-enzymes and are activated by Ca2+-induced auto-proteolysis of an N-terminal peptide [7]. Purified mu-calpain, or calpain I, requires micromolar amounts of Ca2+ for in vitro activation; m-calpain, or calpain II, requires millimolar Ca2+ for activation [7, 22].

Calpain could affect reperfusion function directly through limited proteolysis of the sarcomeres [10, 14]. Alternatively, calpain could depress reperfusion function indirectly via limited proteolysis of proteins involved in excitation–contraction coupling [25], or proteins such as microtubules [26]and intercalated disks [33]. Another indirect effect of calpain on function during reperfusion may occur as a result of calpain-mediated limited proteolysis of protein kinase C (PKC) [16, 17]. Partially proteolyzed PKCs, designated protein kinase M (PKM) [17, 24], are constitutively active kinases that phosphorylate contractile proteins in a unique manner [23, 30].

The direct and indirect effects of calpains on cardiac function have been investigated using calpain-specific inhibitors. Since naturally occurring calpain inhibitors, such as E-64 and leupeptin, cannot readily pass through the sarcolemma, their in vivo usefulness is limited [21]. Consequently, new plasma membrane-permeant calpain inhibitors, such as the leupeptin derivative MDL-28170 (Cbz-Val-Phe-H), have been synthesized and characterized [21].

Experiments reported in this paper examine the role activated calpains play in myocardial stunning. We investigated whether MDL-28170 protects the Langendorff perfused ferret heart against ‘stunning’. Our results show that MDL-28170 significantly improves function during reperfusion. We also tested whether the sensitivity of myofilaments to Ca2+ is altered by ischemia/reperfusion, and whether alterations in the sensitivity of myofilaments to Ca2+ induced by ischemia could be prevented by MDL-28170. Our results show that skinned muscle fibers isolated from the trabeculae of stunned ferret hearts have decreased sensitivity to Ca2+, and that this decrease in sensitivity is reversed by treating hearts with MDL-28170 prior to ischemia and during early reperfusion. Finally, we investigated whether products of calpain-mediated limited proteolysis are observed in ischemic/reperfused hearts. Specifically, we determined whether PKCε, a substrate for calpain [16, 17]and the major PKC isoform in the mammalian heart [1], is partially proteolyzed during ischemia/reperfusion. We report the proteolysis of PKCε to PKMε in ischemic/reperfused ferret hearts, and that MDL-28170 essentially prevents this limited proteolysis.

2 Material and methods

2.1 Preparation of perfused ferret hearts

All investigations conform to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication no. 85-23, revised 1985), as enforced by the Institutional Animal Use and Care Central Committee of the University of Alabama at Birmingham. Three- to 4-month-old ferrets (Marshall Research Animals, North Rose, NY 14516) were anesthetized with pentobarbital sodium (60 mg/kg i.p.). The hearts were isolated, their aortas cannulated and Langendorff perfused at 30°C with Krebs-Ringer perfusate containing ([mM]) Na+ [145], K+ [4.2], Ca2+ [1.25], Mg2+ [1.2], Cl− [125], SO42− [1.2], H2PO4− [2.4], HCO3− [25] and glucose [11]. All perfusates were equilibrated with 95% O2 and 5% CO2. Perfusion of the aortic root was achieved by a servo-controlled pump that maintained a constant perfusion pressure of 100 mm Hg. Coronary flow was calibrated for each experiment by measuring the volume of perfusate forwarded by ten revolutions of the pump, and then measured continuously throughout all experiments. An on-line computer monitored flow rate by generating an output signal proportional to the pump rate which was recorded on a strip-chart recorder. A suture was placed between the central fibrous body and the crest of the interventricular septum in order to produce atrioventricular block. Bi-polar pacing leads then were inserted into the outflow tract of the right ventricle, and square wave pulses of 5 ms duration with currents 10% above threshold levels were delivered from a constant current stimulator. Hearts were paced at 100 or 120 beats per min throughout all perfusions. Left ventricular contractile performance was assessed using a water-filled latex balloon that was inserted into the left ventricle and sutured in place. The balloon was connected, via a water-filled polyethylene catheter, to a pressure transducer, and left ventricular pressures were continuously recorded. End-diastolic pressures (LVEDP) were set to 4–10 mm Hg by adjusting the balloon volume. The initial left ventricular developed pressures (LVDP) varied between 90 and 120 mm Hg. Global ischemia was achieved by turning off the servo-controlled pump.

2.2 Perfusion protocol and procedure for induction of stunning

All isolated heart preparations were perfused for an initial 10–20 min period during which heart function stabilized. In one set of experimental hearts, the stabilization period was followed by a 60 min period of normoxic perfusion. A second set of experimental hearts was subjected to 20 min of global ischemia followed by 30 min of reperfusion. The LVEDP of this latter group of hearts did not increase during the 20 min of global ischemia, and only a small increase in LVEDP, that did not exceed 20 mm Hg, was observed early after onset of reperfusion. Within 5 min of reperfusion the LVEDP returned to pre-ischemic levels and stayed unchanged throughout reperfusion. This procedure of ischemia/reperfusion, however, produced systolic dysfunction during reperfusion. The steady state LVDP achieved during reperfusion was 40 to 45% lower than that measured during pre-ischemic perfusion, or in normoxically perfused control hearts (Fig. 1). This second set of experimental hearts, therefore, exhibited functional changes defined as stunning [3].

Fig. 1

Pre-treatment with MDL-28170 preserves the mechanical function of reperfused ferret hearts: Isolated ferret hearts were Langendorff perfused, and subjected to ischemia and reperfusion as described in Section 2. LVDP was measured using an isovolumic balloon in the left ventricle of these hearts. MDL-28170 at 0.2 _μ_M, in 2.6 mM DMSO, was used to pre-treat hearts prior to ischemia and reperfusion. MDL-28170-treated hearts (●) had significantly improved mechanical function during reperfusion compared to untreated or DMSO-treated hearts (■).

To test the effect of MDL-28170 on left ventricular function during reperfusion, a stock solution was prepared by dissolving MDL-28170 in DMSO. MDL-28170 was introduced into the Krebs-Ringer perfusate at a final concentration of 0.2 _μ_M MDL-28170 and 2.6 mM DMSO. All hearts treated with MDL-28170 were perfused with Krebs-Ringer perfusate containing MDL-28170 for 10 min prior to ischemia and for the first 10 min of reperfusion.

2.3 Preparation of skinned myofilaments and measurement of myofilament sensitivity to Ca2+

Trabeculae, ≤130 _μ_m in diameter, were isolated from the right ventricles of ferret hearts that were either perfused normoxically or subjected to ischemia/reperfusion in the presence or absence of MDL-28170. All trabeculae were skinned chemically using 1% Triton X-100, as previously described [31]. The solutions for skinning, relaxing, and activating these trabeculae contained ([mM]) K+EGTA [10], free Mg2+ [3], BES (N,N bis [2-hydroxymethyl]-2-aminoethane sulfonic acid) [30], ATP [5], Na+phosphocreatine [12], leupeptin [0.5], dithiothreitol [1], creatine phosphokinase [15 U/ml], and sufficient KCl to adjust the ionic strength of the solution to 180 mM. The final concentrations of the metal, ligand, and metal-ligand complexes present in all solutions were determined using the computer program of Fabiato [9]. Stability constants were corrected for a temperature of 22°C, the temperature at which myofilament function was measured. The pH of each solution was titrated to 7.0 with 0.1 M KOH. In addition, the relaxing solution contained a pCa (-log free [Ca2+]) of 9.0.

pCa2+-force curves then were obtained as previously reported [31]. The experimental data for these measurements were fit to a Hill equation of the form:

formula

using a non-linear least-square method. F is the force expressed in g/mm2 obtained at various concentrations of Ca2+; N is the Hill coefficient; _F_max is the measured maximum Ca2+-activated force; and KD is the [Ca2+] required to achieve half-maximal force.

2.4 Western analysis of PKCε/PKMε in stunned ferret hearts

Langendorff-perfused ferret hearts were subjected to either normoxic perfusion or global ischemia/reperfusion in the presence and absence of MDL-28170 as outlined in Section 2.1. At the end of perfusion, all hearts were freeze-clamped with Wollenberger tongs, and stored at −90°C. For Western analysis, approximately 100 mg of these frozen hearts were recovered, pulverized under liquid nitrogen, and homogenized in 9.5 M urea, 3% sodium dodecyl sulfate, 0.7 M _β_-mercaptoethanol, and 0.08 M Tris (pH 6.8) using a Tekmar Tissumizer followed by homogenization with loose- and tight-fitting glass homogenizers. Individual homogenates from normoxic or ischemic/reperfused hearts were heated at 95°C for 3 to 5 min, cooled, and their protein content determined using the biuret method. Approximately 50 _μ_g of protein from extracts of control and ischemic/reperfused hearts were electrophoresed through a 10% SDS-polyacrylamide gel, and transferred at 4°C to PVDF paper for 16–20 h [4]. The PVDF paper was recovered, blocked at room temperature, and probed with a polyclonal antibody for PKCε which recognizes amino acids 726–737 of the C-terminus of PKCε (Gibco/BRL). Enhanced chemiluminescence labelling was used to detect PKCε-reactive proteins (Amersham Co.).

2.5 Statistical analysis of data

All measurements are reported as the mean±1 S.D. Significance of differences between the means was assessed either with paired Student's _t_-test or one-way ANOVA. Values were considered statistically significant at P<0.05.

3 Results

3.1 Ferret hearts exhibit mechanical stunning during 30 min of reperfusion following 20 min of ischemia

Ferret hearts (n = 5) perfused with normoxic Krebs-Ringer perfusate have LVDPs of 105±10 mm Hg and LVEDPs of 8±2 mm Hg. After 20 min of global ischemia and 5 min of reperfusion their LVDPs are 59±11 mm Hg and their LVEDPs are 9±2 mm Hg. Over the ensuing 25 min of reperfusion, LVDP remains significantly depressed (Fig. 1), while the LVEDP remains unchanged and similar to control values. In separate experiments, after 20 min of global ischemia and 5 min of reperfusion in the presence of 2.6 mM DMSO, the vehicle for MDL-28170, ferret hearts (n = 4) have LVDPs of 61±13 mm Hg and LVEDPs of 8±1 mm Hg. During the ensuing 25 min of reperfusion, LVDPs remain significantly depressed in hearts treated with DMSO, while their LVEDPs remain normal. Since the time course and the degree of contractile dysfunction during reperfusion following 20 min of global ischemia are not significantly affected by DMSO, data from hearts made ischemic and reperfused in the presence or absence of 2.6 mM DMSO were pooled, and are reported as the MDL-28170-untreated group of ischemic/reperfused hearts (Fig. 1).

The mechanical function and coronary flows of hearts (n = 7) perfused normoxically for 60 min are not significantly different from the function and flow measured during pre-ischemic perfusion in either MDL-28170-treated or -untreated hearts (data not shown). The mechanical performance and coronary flows of ferret hearts (n = 3) also are not affected during 60 min of normoxic perfusion by the presence of 2.6 mM DMSO.

3.2 Pre-treatment with MDL-28170 minimizes myocardial stunning

The general hypothesis that calpains are activated during ischemia/reperfusion and mediate mechanical stunning during reperfusion was tested by measuring the effect of MDL-28170, a membrane-permeant calpain inhibitor, on myocardial stunning in reperfused ferret hearts.

Recovery of LVDP during reperfusion is significantly improved in MDL-28170-treated hearts compared to untreated hearts (Fig. 1). Compared to pre-ischemic LVDPs of 107±9 mm Hg and LVEDPs of 8±2 mm Hg, MDL-28170-treated hearts have LVDPs of 97±7 mm Hg and LVEDPs of 8±1 mm Hg after 30 min of reperfusion following 20 min of ischemia.

Pre-ischemic coronary flows of MDL-28170-untreated (n = 9) and -treated (n = 5) hearts average 37±5 and 36±7 ml/min, respectively. Restitution of flow following 20 min of ischemia causes a prompt ‘hyperemic’ response that peaks within 2 min. This ‘hyperemic’ flow is 105±20 ml/min in MDL-28170-untreated hearts; MDL-28170-treated hearts have ‘hyperemic’ flows of 94±15 ml/min. These two responses are not significantly different. Following 30 min of reperfusion, the coronary flow in both the MDL-28170-untreated and -treated hearts returns to pre-ischemic levels of 38±5 and 34±6 ml/min, respectively. With heart weights of 6.9±0.3 g for MDL-28170-untreated and 6.8±0.4 g for MDL-28170-treated hearts, the coronary flow per gram tissue are identical in MDL-28170-treated and -untreated hearts during normoxic perfusion and reperfusion.

3.3 Myofilament sensitivity to Ca2+ Is decreased in trabeculae from stunned hearts and normalized by pre-treatment with MDL-28170

Mechanical stunning during reperfusion of ischemic ferret hearts was found to be sensitive to MDL-28170 (Fig. 1). Subsequent experiments had two additional goals. First, to determine whether the characteristics of myofilaments isolated from stunned hearts were similar or different from those of myofilaments isolated from normal hearts. And second, to determine whether differences in the character of myofilament isolated from normal and stunned hearts were sensitive to MDL-28170.

To accomplish these goals, trabeculae were isolated from the right ventricle of ferret hearts that either had been perfused normoxically or had been subjected to 20 min of ischemia followed by 30 min of reperfusion in the presence or absence of MDL-28170. These three sets of trabeculae then were skinned chemically, and the Ca2+ responsiveness of their myofilaments characterized.

The maximum Ca2+-activated force measured in skinned trabeculae (n = 8) isolated from normoxically perfused ferret hearts (n = 7) was 11.8±1.2 g/mm2 (A in Fig. 2a). The maximum Ca2+-activated force measured in trabeculae isolated from hearts (n = 5) made ischemic and reperfused in the absence of MDL-28170 was 8.8±1.1 g/mm2 (C in Fig. 2a). This Ca2+-activated maximum force was significantly lower than the maximum force measured in control trabeculae. The maximum Ca2+-activated force measured in skinned trabeculae isolated from hearts (n = 5) made ischemic and reperfused in the presence of MDL-28170 was 10.8±1.4 g/mm2 (B in Fig. 2a). This value was not significantly different from that measured in trabeculae isolated from normoxically perfused hearts.

Fig. 2

(a) Maximum Ca2+-activated force, depressed in skinned trabeculae isolated from ischemic/reperfused ferret hearts, is largely preserved by MDL-28170 pre-treatment: Trabeculae from right ventricles of ischemic/reperfused ferret hearts were isolated and chemically skinned as previously described [31]. Force-pCa2+ curves were obtained from these preparations [31]. A in panel a represents Ca2+-activated force in normoxic trabeculae (A; ▴); pre-treatment of isolated ferret hearts with MDL-28170 (B; □) significantly improves maximum Ca2+-activated force compared to untreated hearts (C; ●). (b) Normalization of Ca2+-activated force demonstrates that ischemia/reperfusion induces myofilament hyposensitivity which is MDL-28170-sensitive: Data from panel a were normalized to the average maximal Ca2+-activated force (_F_max) for normoxic conditions (A; ▴), MDL-28170-treated (B; □) or -untreated (C; ●) hearts. Normoxic skinned trabeculae had an _F_max of 11.8g±1.2g/mm2. The _F_max of MDL-28170-treated skinned trabeculae was 10.8±1.4 g/mm2; the _F_max of MDL-28170-untreated skinned trabeculae was 8.8±1.1 g/mm2. Ischemia/reperfusion induces hyposensitization of the trabeculae, which is reversed by MDL-28170 pre-treatment.

The KD for Ca2+-induced contraction of skinned trabeculae isolated from normoxically perfused hearts was 6.27±0.06 (A in Fig. 2b). This KD was significantly different from the KD of 6.09±0.04 measured for Ca2+-induced contraction in skinned trabeculae isolated from hearts made ischemic and reperfused in the absence of MDL-28170 (C in Fig. 2b). The KD of skinned trabeculae isolated from hearts treated with MDL-28170 prior to ischemia/reperfusion was 6.28±0.04 (B in Fig. 2b). This value was not significantly different from that obtained from hearts perfused normoxically.

The Hill coefficient derived from the force-pCa2+ curves of trabeculae from normoxically perfused hearts was 4.4±1.0. The Hill coefficient derived from the force-pCa2+ curves of skinned trabeculae isolated from hearts that had undergone ischemia/reperfusion in the absence of MDL-28170 was 3.2±0.7. This value, while lower than that of control hearts, was not significantly different from it. Both of these values were also not significantly different from the Hill coefficient of 4.3±0.7 that was obtained in trabeculae isolated from hearts that had undergone ischemia/reperfusion in the presence of MDL-28170.

3.4 PKMε is produced during ischemia/reperfusion in an MDL-28170-sensitive manner

PKCs are good substrates for both in vitro and in vivo limited proteolysis by activated calpains [16, 17]. We determined, therefore, whether PKCε, the main PKC in heart [1], is partially proteolyzed in ischemic/reperfused ferret hearts, and whether this proteolysis is sensitive to MDL-28170.

Ferret hearts were subjected to 20 min of global ischemia followed by 10 min of reperfusion in the presence (n = 4) or absence (n = 4) of MDL-28170. Four other hearts were perfused normoxically and served as controls. These 12 hearts were freeze-clamped at the end of perfusion and stored at −90°C. Hearts were prepared for analysis by extracting with SDS-urea, followed by SDS-PAGE electrophoresis [4]. Electrophoretically resolved samples were then probed with an antibody for PKCε.

Protein samples from normoxically perfused ferret hearts typically contained one PKCε species with a molecular weight of ≈96 kDa (Fig. 3; right lane (C)). This protein corresponds to cardiac PKCε[1]. Samples from ferret hearts made ischemic and reperfused in the absence of MDL-28170 contained two PKCε species; one at ≈96 kDa corresponding to PKCε, and a second ≈25 kDa smaller (Fig. 3; left lane (I/R)). This decrease in apparent molecular weight is consistent with that reported for other PKMs [16, 17, 24]. This second smaller protein observed in reperfused hearts is likely to be PKMε, the calpain-proteolyzed form of PKCε. The fact that MDL-28170 diminished PKMε in reperfused hearts supports this assertion (Fig. 4).

Fig. 4

Pre-treating ferret hearts with MDL-28170 prior to ischemia/reperfusion reduces their content of PKMε: Isolated ferret hearts were subjected to ischemia/reperfusion in the absence (lane C) or presence (lane A) of MDL-28170. Hearts were recovered, extracted, and 50 _μ_g of protein were electrophoresed and Western analyzed as in Fig. 3. Pre-treatment with MDL-28170 (lane A) decreases levels of PKMε, the partially proteolyzed form of PKCε, in reperfused ferret hearts. Lane B contained molecular weight markers used to obtain the apparent weights of PKCε and PKMε.

Fig. 3

PKCε undergoes limited proteolysis to PKMε during ischemia/reperfusion: Ferret hearts were perfused normoxically or subjected to ischemia/reperfusion as described in Section 2. Frozen hearts were extracted, and 50 _μ_g of protein from normoxically perfused or ischemic/reperfused hearts were electrophoresed through a 10% SDS-PAGE gel. Samples were probed for their PKCε content using a polyclonal antibody as described in Section 2. Normoxically perfused ferret hearts (lane C) contain a single species of PKCε at a molecular weight of ≈96 kDa. Ischemic/reperfused ferret hearts (lane [I/R]) contain two PKCε reactive species; one at ≈96 kDa, and a second, PKMε, at ≈70 kDa.

4 Discussion

This paper addresses the hypothesis that activated calpains play a role in the pathophysiology of the post-ischemic systolic dysfunction known as ‘myocardial stunning’. To test this hypothesis, isolated ferret hearts were perfused in the presence or absence of MDL-28170, a membrane-permeant inhibitor specific to calpains, and subjected to ischemia and reperfusion. Four results from these experiments suggest that activated calpains are indeed present in ischemic/reperfused hearts, and contribute significantly to the phenomenon of stunning.

First, treatment of ferret hearts with MDL-28170 prior to ischemia and during early reperfusion markedly reduces post-ischemic contractile dysfunction. This observation extends previous reports in which leupeptin both decreased proteolysis in ischemic myocardium and protected against myocardial stunning [2, 20]. Second, treatment of hearts with MDL-28170 prevents the decrease in the Ca2+ sensitivity of their trabeculae when compared to trabeculae isolated from untreated hearts subjected to ischemia/reperfusion. Third, the maximum Ca2+-activated force generated by skinned trabeculae isolated from ischemic/reperfused hearts tends to be preserved by MDL-28170. In this context it is useful to note that we observed no positive inotropic action when ferret hearts were exposed to 0.2 _μ_M of MDL-28170 for 30 min (unpublished results). Finally, PKCε, a substrate for calpain [16, 17, 24], undergoes limited proteolysis following reperfusion, which is diminished by treatment with MDL-28170.

4.1 Contractile performance and myofilament sensitivity to Ca2+ decrease, while the level of PKMε increases in reperfused hearts

Our results indicate that mechanical function in the isolated ferret heart is depressed during reperfusion following 20 min of ischemia (Fig. 1). Ischemia/reperfusion also increases the concentration of Ca2+ required to generate half-maximal steady state developed tension in skinned trabeculae (C in Fig. 2b). Such decreases in myofilament sensitivity to Ca2+ have been reported by others [13]. This myofibrillar hyposensitivity to Ca2+ is accompanied by a decrease in the maximum Ca2+-activated force (_F_max) generated by skinned trabeculae isolated from ischemic/reperfused hearts (C in Fig. 2a). While depression of _F_max in myofilaments isolated from post-ischemic hearts have been noted by some [5, 18], others report no such effect [8, 13]. The reason for this discrepancy is unclear, although the use of different preparations and species could be important factors. Our results also show that the slopes of the force-pCa2+ curves of skinned trabeculae isolated either from normoxically perfused hearts or post-ischemic hearts without treatment with MDL-28170 were similar to those of trabeculae isolated from hearts that underwent ischemia/reperfusion in the presence of MDL-28170. The Hill coefficients in our experiments are similar to those reported by others [13], and demonstrate that cooperativity is little affected in the stunned ferret heart.

Ischemia followed by reperfusion, therefore, decreases myofibrillar sensitivity to Ca2+ and the maximum contractile force generated by myofibrils during reperfusion. These changes in myofibrillar function could result from either reversible or irreversible changes to the myofibrils which are stable and survive the process of chemical skinning.

Reversible depression of contractile function during in situ ischemia/reperfusion may be mediated by phosphorylation/dephosphorylation of cardiac proteins [11]. For example, in vitro phosphorylation of contractile proteins by protein kinase A and PKCα decreases myofibrillar sensitivity to Ca2+, and depresses activity of myofibrillar actomyosin ATPase, respectively [32].

Another possible cause of reversible contractile dysfunction during ischemia/reperfusion may be the activation of mu-calpain brought about by elevations in cytosolic Ca2+ in ischemic/reperfused hearts [15, 29]. Specific (mu-calpain:PKCα) complexes exist in muscle [27]. Activation of mu-calpain, therefore, could generate partially proteolyzed PKCs, known as PKMs, which are constitutively active protein kinases that do not require the known lipid or ion co-factors of classical or novel PKCs for activity [16, 17]. PKMs exhibit a unique pattern of substrate specificity [23]. Hence, PKMε, produced via calpain proteolysis as a result of elevated cytosolic Ca2+, would not only occur in the setting of ischemia/reperfusion, but would act to phosphorylate multiple substrates, including myofibrils.

In keeping with this viewpoint, our results demonstrate the existence of a protein in ischemic/reperfused hearts that cross-reacts with a PKCε specific antibody and possesses a molecular weight anticipated for PKMε (Fig. 3). These results, therefore, support the possibility that activated calpains exist in ischemic/reperfused hearts, and act to partially proteolyze cardiac proteins, notably PKCε.

It is noteworthy that our Western analyses demonstrate that the PKCε in ferret ventricles has an apparent molecular weight of ≈96 kDa. This is in agreement with a previous report for the apparent molecular weight of PKCε obtained using Western analysis [1]. They contrast, however, with the molecular weight of PKCε expected from the sequence of cDNA clones for PKCε[28]. One interpretation of this result is that significant post-translational modification of PKCε occurs in the heart cell. The possible nature(s) of these modification(s) is still unclear.

4.2 MDL-28170 diminishes myocardial stunning, maintains myofilament sensitivity to Ca2+, and reduces the level of PKMε in reperfused hearts

The presence of PKMε in stunned ferret hearts (Fig. 3) strongly indicates that calpains are activated in ischemic/reperfused hearts. Thus, MDL-28170, a membrane-permeant calpain inhibitor, was tested as an agent to ameliorate the effects of ischemia on heart function during reperfusion. Treatment with MDL-28170 provides significant protection against ischemia/reperfusion-induced depressions in both mechanical and myofilament functions (Figs. 1 and 2). These results, therefore, lead us to conclude that activated calpains could be involved in and, at least in part, mediate post-ischemic myocardial dysfunction. Furthermore, MDL-28170 significantly decreased the level of PKMε in ischemic/reperfused hearts (Fig. 4). One possible mechanism of short-term reversible stunning, therefore, may be PKMs, generated by mu-calpain-mediated partial proteolysis of PKCs during ischemia and/or reperfusion, which phosphorylate proteins critical to contraction and excitation–contraction coupling.

While results reported in this paper suggest the existence and the importance of activated calpains in myocardial stunning, two important facts remain to be ascertained. First, the presence of activated calpains in ischemic/reperfused hearts must be directly measured by biochemical and immunological means. Second, the precise molecular site(s) at which calpains contribute to ‘myocardial stunning’ must be resolved. Since activated calpains have multiple protein substrates, any of these proteins could contribute to stunning. These substrates include proteins involved in the regulation of Ca2+ movement in the heart cell, including the L-type Ca2+ channel [25], proteins essential to the generation of force, including troponin I and C [10], and proteins that can directly or indirectly alter the functional properties of the contractile apparatus itself, including the PKCs [16, 17].

Acknowledgements

This work was supported by an NHLBI SCOR on Ischemic Heart Disease P50 HL17667 (F.U., P.E.W., S.B.D., K.D.H., A.A.W.), and Grants-in Aid from the Alabama Affiliate of the American Heart Association Al-G-950012 (F.U.) and Al-G-940023 (P.E.W.).

References

Characterization of protein kinase C isotype expression in adult rat heart. Protein kinase C-ε is a major isotype present, and it is activated by phorbol esters, epinephrine, and endothelin

Circ Res

1993

757

767

Role of cellular proteinases in acute myocardial infarction. I. Proteolysis in non-ischemic and ischemic rat myocardium and the effects of antipain, leupeptin, pepstatin, and chymostatin administered in vivo

J Am Coll Cardiol

1983

671

680

The stunned myocardium: prolonged, postischemic ventricular dysfunction

Circulation

1982

1146

1149

‘Western Blotting’: Electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A

Analyt Biochem

1981

112

195

207

Decreased myofilament responsiveness in myocardial stunning follows transient calcium overload during ischemia and reperfusion

Circ Res

1992

1334

1340

Effects of monoclonal antibodies specific for the 28-kDa subunit on catalytic properties of the calpains

J Biol Chem

1993

268

25740

25747

Purification and characterization of calcium-dependent proteases from rat heart

J Biol Chem

1983

258

5660

5665

Stunning does not change the relation between calcium and force in skinned rat trabeculae

J Mol Cell Cardiol

1993

541

549

Fabiato A. Computer programs for calculating total from specified free of free from specified total ionic concentrations in aqueous solutions containing multiple metals and ligands. In: Fleischer S, Fleischer B, editors. Methods in enzymology, Vol. 157, Biomembranes part Q ATP-driven pumps and related transport: calcium, proton, and potassium pumps. New York: Academic Press, 1988:Chapter 31, 378–417.

Degradation of troponin I in ischemic/reperfused myocardium: molecular basis of myocardial stunning?

Biophys J

1995

A112

(Abstract)

Effect of protein kinase C activation on sarcoplasmic reticulum function and apparent myofibrillar Ca2+ sensitivity in intact and skinned muscles from normal and diseased human myocardium

Circ Res

1990

744

752

[Ca2+]i transients in ischemic Langendorff ferret hearts

J Mol Cell Card

1995

A20

(Abstract)

Altered calcium sensitivity of isometric tension in myocyte-sized preparations of porcine postischemic stunned myocardium

Circ Res

1993

Calcium activated neutral protease. Its localization in the myofibril especially at the Z-band

J Biochem

1980

343

346

Direct measurement of changes in intracellular calcium transients during hypoxia, ischemia, and reperfusion of the intact mammalian heart

Circ Res

1989

1029

1044

Proteolytic activation of calcium-activated, phospholipid-dependent protein kinase by calcium-dependent neutral protease

J Biol Chem

1983

258

1156

1164

et al.

Limited proteolysis of protein kinase C subspecies by calcium-dependent neutral protease (calpain)

J Biol Chem

1989

264

4088

4092

Pathophysiology and pathogenesis of stunned myocardium: depressed Ca2+ activation of contraction as a consequence of reperfusion-induced cellular calcium overload in ferret hearts

J Clin Invest

1987

950

961

Quantification of [Ca2+]i in perfused hearts: critical evaluation of the 5F-BAPTA and nuclear magnetic resonance method as applied to the study of ischemia and reperfusion

Circ Res

1990

1255

1267

Protective effect of the protease inhibitor leupeptin against myocardial stunning

J Cardiovasc Pharmacol

1993

135

142

Cell-penetrating inhibitors of calpain

Trends Biochem Sci

1991

150

153

Canine cardiac calcium-dependent proteases: resolution of two forms with different requirements for calcium

FEBS Lett

1980

109

129

133

Catalytic fragment of protein kinase C exhibits altered substrate specificity toward smooth muscle myosin light chain

FEBS Lett

1991

294

144

148

Identification of the proteolytically activated form of protein kinase C in stimulated human neutrophils

Proc Natl Acad Sci U. S. A.

1990

3705

3707

Calpastatin and nucleotides stabilize cardiac calcium channel activity in excised patches

Pflügers Arch

1991

418

Reperfusion after brief ischemia disrupts the microtubule network in canine hearts

Circ Res

1993

361

375

Isolation and identification of a mu-calpain-protein kinase C alpha complex in skeletal muscle

FEBS Lett

1995

359

Proteolytic activation of Z protein kinase C_ε_

Eur J Biochem

1990

191

431

435

Elevation in cytosolic free calcium concentration early in myocardial ischemia in perfused rat heart

Circ Res

1987

700

707

Takahashi K. Calpain substrate specificity. In: Mellgren RL, Murchi T, editors, Intracellular calcium-dependent proteolysis, Boca Raton, CRC Press, 1990:Chapter 4, 65–74.

Minimal deterioration of maximal calcium activated tension in skinned cardiac trabeculae of ferrets

J Mol Cell Cardiol

1992

P47

(Abstract)

Protein kinase C-mediated phosphorylation of troponin I and C-protein in isolated myocardial cells is associated with inhibition of myofibrillar actomyosin MgATPase

J Biol Chem

1993

268

2705

2711

et al.

Reperfusion of rat heart after brief ischemia induces proteolysis of calspectin (nonerythroid spectrin or fodrin) by calpain

Circ Res

1995

603

610