INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE MAMMALIAN MYOCARDIUM EXPRESSES µ-, -, and -opioid receptors. Expression of µ-opioid receptors decreases postnatally to undetectable levels by day 7, whereas the - and -opioid receptors continue to be expressed in the adult (46). Myocardial -opioid receptor activation causes a transient increase in twitch amplitude followed by a negative inotropic effect in adult rats (43). Opioid receptor activation before ischemia can also protect the heart from postischemic contractile dysfunction (33) and necrosis (23, 34, 35, 36) via a protein kinase C (PKC)-dependent pathway. The overall goal of the present study was to investigate the cellular mechanism(s) responsible for the effects of -opioid receptor activation on the heart. An understanding of the basic underlying mechanism will help in the characterization of a possible therapeutic role for -opioid receptor stimulation in diseased myocardial states.

The observed positive inotropic effects of -opioid receptor stimulation of the heart (43) might involve one or several mechanisms. These include an increase in intracellular [Ca²⁺] (43), activation of the sarcolemmal Na⁺/H⁺ exchanger [leading to intracellular alkalosis and a resulting increase in myofilament force on contraction at a given Ca²⁺ concentration (42)], and/or a direct increase in the Ca²⁺ sensitivity of myofilament tension generation. A direct effect of -opioid receptor activation on myofilament Ca²⁺ sensitivity has not been previously investigated. Therefore, the first objective of the present study was to examine the effects of -opioid receptor stimulation on the relationship between [Ca²⁺] and isometric tension in agonist-treated and subsequently skinned ventricular myocytes. -Opioid receptor activation may improve postischemic myocardial function (33) by decreasing the sensitivity of the myofilaments to ischemia-induced acidosis. As such, the second aim of this study was to determine if the Ca²⁺ sensitivity of tension and maximum isometric tension are equally influenced by decreased pH in -opioid receptor agonist-treated compared with untreated and subsequently skinned ventricular myocytes.

A PKC-dependent decrease in maximum actin-myosin ATPase activity was previously observed in hearts pretreated with a -opioid receptor agonist (33). However, this decrease in actin-myosin ATPase could be due to the combined effects of -opioid receptor activation and ischemia. Alternatively, a PKC-dependent decrease in actin-myosin ATPase may be an effect due solely to -opioid receptor activation of the heart. Thus the third objective of the present study was to determine if nonischemic perfused hearts treated with a -opioid agonist demonstrate 1) increases in whole heart [ATP], 2) myofibrils with decreased maximum Ca²⁺-dependent actin-myosin ATPase, and 3) a PKC dependency on any observed -opioid-induced changes.

Finally, -opioid receptor activation of ventricular myocytes has been shown to activate PKC (42). Thus the fourth objective of the present study was to establish which PKC isoforms, if any, decreases maximum actin-myosin ATPase in cardiac myofibrils.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Enzymatic isolation of cardiac myocytes. Ventricular myocytes were obtained by enzymatic digestion of hearts from female Wistar rats according to the method of Lester et al. (19). In brief, the heart was excised, and the aorta was cannulated for suspension on a Langendorff perfusion apparatus. The heart was perfused with oxygenated Ca²⁺-containing Ringer solution (Ca²⁺-Ringer solution) to rinse out residual blood and then with Ca²⁺-Ringer solution containing collagenase (1 mg/ml; Worthington). During collagenase perfusion, three sequential additions of CaCl₂ were done to yield a final [Ca²⁺] of 0.75 mM. The ventricles were cut into small pieces and agitated in a flask containing recycled enzyme solution with CaCl₂. Ventricular pieces were dissociated by gentle aspiration through a large-bore pipette tip, and cells were washed in Ca²⁺-Ringer solution. Washed cells were incubated for 2 min with norbinaltorphimine (norBNI; 1 µM, a -opioid receptor antagonist) or Ca²⁺-Ringer solution (control). Cells were then treated for 5 min with various receptor agonists/antagonists dissolved in Ca²⁺-Ringer solution (see Agonist/antagonist/inhibitor dosage rationale). After centrifugation and the decanting of the supernatant, myocytes were exposed to a relaxing solution (see Solutions) containing 0.6% Triton X-100 for 5 min to chemically remove lipid membranes. Cells were then washed three times in a relaxing solution without Triton X-100 and stored on ice.

Effects of isometric tension as a function of [Ca²+]. Isolated cardiac myocytes were attached via glass micropipettes to a force transducer (model 403, Cambridge Technology; Watertown, MA) and piezoelectric translator (model 173, Physik Institute; Waldbronn, Germany) with Great Stuff adhesive (Insta-Foam; Marrietta, GA). Sarcomere length was adjusted to 2.1-2.3 µM, and cell length and width were measured. A tension-pCa relationship was obtained by initially measuring force during maximal activation (pCa 4.5), followed by contractions at randomly chosen submaximal pCa solutions, and again at pCa 4.5 to assess any decline in the performance of the cell. Active tension was calculated as the difference in measured total tension (P) and resting tension (RT) obtained in a pCa 9.0 solution. For each submaximal contraction, active tension was normalized to maximum active tension (P₀) generated by the cell, i.e., (P  RT)/P₀.

Langendorff-perfused heart preparation. Hearts were removed from female Wistar rats anesthetized by Metofane inhalation. The isolated hearts were mounted on a Langendorff perfusion apparatus and paced at 300 beats/min, and a balloon was inserted in the left ventricle and inflated until end-diastolic pressure (EDP) was 5-15 mmHg (33).

Myofibrillar isolation. Myofibrils were isolated from cells or left ventricles according to a modified protocol described by Murphy and Solaro (25). Left ventricles were cut from Langendorff-perfused hearts, homogenized in standard phosphate buffer (see Solutions), and pelleted. Isolated cells were treated (or untreated), resuspended in standard phosphate buffer, and pelleted. The resulting pellets from hearts and myocytes were dissolved in ice-cold resuspension solution containing Triton X-100 (see Solutions), placed on ice for 30 min, and centrifuged to obtain a pellet. The pellet was washed and resuspended in ice-cold standard phosphate buffer plus 100 nM calyculin A. The protein concentration was determined with a Biuret assay, and myofibrils were diluted to a final concentration of 4-8 mg protein/ml.

Myofibrillar ATPase measurements. ATPase buffers with [Ca²⁺] of pCa 4.0 and 9.0 were used (see Solutions). Myofibrils containing regulated actin were added to the 32°C buffers. After 2 min of incubation, the reaction was quenched with 2 ml of 20% trichloroacetic acid. Inorganic phosphate levels were determined according to the method of Fiske and SubbaRow (12). Inorganic phosphate production was found to be linear with respect to time under conditions of 32°C with a final protein concentration of 1.0-2.0 mg/ml (data not shown).

Ventricular ATP. ATP was quantified using the luciferin-luciferase enzyme technique (21). Hearts were perfused as described under Langendorff-perfused heart preparation and were removed after 25 min. The ventricles were cut from the hearts, quickly frozen in liquid nitrogen, and homogenized in a modified Krebs-Henseleit solution with a pestle and cold mortar. The homogenate was used to measure ventricular ATP with a luciferin-luciferase assay kit (Sigma; St. Louis, MO). The light produced by ATP plus luciferin is used to calculate unknown ATP concentrations of samples. A small amount of homogenized ventricle was used to determine protein concentration with a Biuret assay. Ventricular ATP levels were expressed as nanomoles of ATP per milligram of protein in the homogenate.

Exogenous PKC treatment. Myofibrils from isolated ventricular myocytes were treated with exogenous PKC according to a modified protocol of Noland and Kuo (28). Briefly, myofibrils from isolated ventricular myocytes were incubated for 5 min at 37°C in a reaction mixture (see Solutions) plus recombinant human PKC- or - (PanVera; Madison, WI). The amount of recombinant PKC added was equal to the myofibrillar Ca²⁺-independent PKC activity previously measured (data not shown).

Solutions. The standard phosphate buffer contained 60 mM KCl, 30 mM imidazole (pH 7.0), 2 mM MgCl₂, 4 µM aprotoninin, 15 µM pepstatin A, and 20 µM leupeptin hemisulfate. The resuspension buffer contained 10 mM EGTA, 8.2 mM MgCl₂, 14.4 mM KCl, 60 mM imidazole (pH 7.0), 5.5 mM ATP, 12 mM creatinine phosphate, 10 U/ml creatinine phosphokinase, 100 nM calyculin A, and 1% Triton X-100. The reaction mixture contained 50 mM Tris · HCl (pH 7.5), 30 mM -mercaptoethanol, 0.9 mM CaCl₂, 10 mM MgCl₂, 0.5 mM EGTA, 1 mM ATP, 100 nM calyculin A, and 50 mM KCl. The pCa 4.0 buffer contained 23.48 mM KCl, 5 mM MgCl₂, 3.22 mM ATP, 2 mM EGTA, 20 mM imidazole, and 2.15 mM CaCl₂ (pH 7.0). The pCa 9.0 buffer contained 25.96 mM KCl, 5.13 mM MgCl₂, 3.16 mM ATP, 2 mM EGTA, 20 mM imidazole, and 4.86 µM CaCl₂ (pH 7.0). The free [Ca²⁺] was calculated using the program of Fabiato (10). The modified Krebs-Henseleit solution was composed of 4.7 mM KCl, 118 mM NaCl, 1.2 mM MgSO₄, 1.3 mM CaCl₂, 25 mM NaHCO₃, 11 mM glucose, 1.2 mM KH₂PO₄, 0.05 mM EDTA, and 2 mM lactic acid (pH 7.4).

Agonist/antagonist/inhibitor dosage rationale. The concentration of 1 µM for U50,488H was chosen to selectively activate -opioid receptors (7). Concentrations of 10 µM phenylephrine plus 3 µM propranolol have previously been shown to produce maximal increases in intracellular [Ca²⁺] and twitch amplitude in myocardium (3). Phenylephrine plus propranolol was included as a positive control for PKC activation. The concentration of 100 nM isoproterenol maximally increases troponin I phosphorylation (29) and was included as a positive control for protein kinase A (PKA) activation. PMA (1 µM) activates the conventional and novel PKC isoforms found in the rat heart (39).

Statistical analysis. All values are reported as means ± SE, and P  0.05 was chosen to indicate statistical significance. For tension-pCa²⁺ relationships, a two-way analysis of variance and a Student's t-test were used to determine significance. All other data were analyzed by two-way analysis of variance and Fisher's least-significant difference post hoc test.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Characterization of ventricular myocytes. Photomicrographs of cardiac myocytes attached to micropipettes from control and agonist-treated myocytes were indistinguishable (Fig. 1). Sarcomere lengths of myocytes were not significantly different in relaxing solution and during contraction between any of the groups tested. Average sarcomere lengths, total cell lengths between micropipette tips, and myocyte widths are given in Table 1.

View larger version (129K): [in this window] [in a new window]   Fig. 1.   Light photomicrographs of skinned cardiac myocytes while relaxed in a pCa 9.0 solution and during maximum activation at pCa 4.5 solution at pH 7.0. Average distance between striations are 2.10 and 2.16 µM for all photos.

Receptor agonist effects on isometric tension as a function of [Ca²+] at pH 7.0. Cumulative tension-pCa relationships at pH 7.0 for control and agonist-treated cardiac myocytes are shown in Fig. 2A. Maximum tension was not significantly affected by any of the agonist treatments (Table 1). The -adrenergic receptor agonist isoproterenol induced a significant decrease in the Ca²⁺ sensitivity of tension for pCa values between 5.6 and 6.2. Tensions at submaximal [Ca²⁺] were 10-15% lower in myocytes treated with isoproterenol compared with untreated myocytes. The -opioid receptor agonist U50,488H induced a significant increase in the Ca²⁺ sensitivity of tension for pCa values between 5.8 and 6.2. Tensions at submaximal [Ca²⁺] were 5-8% higher in myocytes treated with U50,488H compared with untreated myocytes. -Adrenergic receptor stimulation with phenylephrine plus the -adrenergic receptor antagonist propranolol did not alter the Ca²⁺ sensitivity of tension. The pCa values of half-maximum tension generation, i.e., pCa₅₀, for all treatments are shown in Table 1. The slopes of the tension-pCa relationships were not significantly different between any of the agonist-treated and control groups.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Mean relative tensions ± SE as a function of pCa at pH 7.0 (A) and pH 6.6 (B) in ventricular myocytes. Myocytes were treated with 100 nM isoproterenol (Iso), 10 µM phenylephrine (Phen) plus 3 µM propranolol (Prop), and 1 µM U50,488H or were untreated (control). After 5 min of agonist treatment, myocytes were skinned. For each preparation, the tension values were normalized to the maximum tension developed at pCa 4.5, pH 7.0, or pH 6.6. Maximum tension, pCa50, and the Hill coefficient can be found for these cells in Table 1. Active tension was calculated as the difference in measured total tension (P) and resting tension (RT) obtained in a pCa 9.0 solution. For each submaximal contraction, active tension was normalized to maximum active tension (P0) generated by the cell, i.e., (P  RT)/P0.

Receptor agonist effects on isometric tension as a function of [Ca²+] at pH 6.6. The cumulative averages of maximum tension at pH 6.6 for the various agonist treatments are presented in Table 1. Maximum tension at pH 6.6 in all myocytes was lower than at pH 7.0. The cumulative tension-pCa relationships for cardiac myocytes at pH 6.6 are shown in Fig. 2B. For control myocytes at pH 6.6 compared with data obtained at pH 7.0, the pCa₅₀ decreased by 0.59 pCa units and shifted the tension-pCa relationship rightward. At pH 6.6, neither the -adrenergic receptor agonist isoproterenol nor -adrenergic receptor stimulation with phenylephrine plus propranolol significantly altered the Ca²⁺ sensitivity of tension compared with control myocytes at pH 6.6. The -opioid receptor agonist U50,488H induced a significant increase in the Ca²⁺ sensitivity of tension compared with control myocytes at pH 6.6.

Effects of -opioid receptor antagonist on U50,488H-dependent changes in isometric tension as a function of [Ca²+] at pH 7.0. Cumulative tension-pCa relationships at pH 7.0 are shown in Fig. 3. Maximum tension was not significantly affected by any of the agonist or antagonist treatments (Table 2). The pCa₅₀ was significantly higher in U50,488H-treated myocytes compared with control myocytes. The -opioid receptor antagonist norBNI abolished the U50,488H-dependent increased in the Ca²⁺ sensitivity of tension. norBNI had no effect on Ca²⁺ sensitivity of tension by itself.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Mean relative tensions ± SE as a function of pCa at pH 7.0 in ventricular myocytes treated with -opioid receptor agonist and antagonist. Myocytes were treated with 100 nM Iso, 1 µM U50,488H, 1 µM norbinaltorphimine (norBNI), a combination of U50,488H and norBNI, or untreated (control). After treatment, myocytes were skinned, and tension-pCa relationships were determined. For each preparation, the tension values were normalized to the maximum tension developed at pCa 4.5; pH 7.0. Maximum tension and pCa50 can be found for these cells in Table 2.

Effect of PMA on the tension-pCa relationship at pH 7.0. Additional experiments were done to determine the effects of receptor-independent activation of PKC. For these experiments, enzymatically isolated myocytes were exposed to either 1 µM PMA plus 1% DMSO or to 1% DMSO alone (paired control). After exposure, the cells were chemically skinned, and isometric tension as a function of [Ca²⁺] was determined at pH 7.0. Figure 4 presents the cumulative tension-pCa relationships for paired control and PMA-treated myocytes. Maximum tension was not affected, whereas the Ca²⁺ sensitivity of tension increased after PMA exposure compared with controls (Table 2). PMA exposure caused a significant 10-17% increase in isometric tension for pCa values between 5.6 and 6.2. The slope of the tension-pCa relationships were not significantly different between control and PMA-treated myocytes.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Mean relative tensions ± SE as a function of pCa at pH 7.0 of ventricular myocytes in the presence and absence of phorbol 12-myristate 13-acetate (PMA). For each preparation, the tension values were normalized to the maximum tension developed at pCa 4.5; pH 7.0. Maximum tension and pCa50 can be found for these cells in Table 2.

Effects of PKC inhibitor on U50,488H-dependent changes in isometric tension as a function of [Ca²+] at pH 7.0. Cumulative tension-pCa relationships at pH 7.0 are shown in Fig. 5. Maximum tension was not significantly affected by either treatment (Table 2). Treatment of myocytes with the PKC inhibitor chelerythrine chloride before U50,488H exposure inhibited the -opioid receptor-dependent increase in the Ca²⁺ sensitivity of tension. Chelerythrine chloride alone did not alter Ca²⁺ sensitivity.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Mean relative tensions ± SE as a function of pCa at pH 7.0 in ventricular myocytes treated with -opioid receptor agonist and protein kinase C (PKC) inhibitor. Myocytes were treated with 2 µM chelerythrine (Chel) or 2 µM chelerythrine chloride plus 1 µM U50,488H (U50). After treatment, myocytes were skinned, and tension-pCa relationships were determined. For each preparation, the tension values were normalized to the maximum tension developed at pCa 4.5; pH 7.0. Maximum tension and pCa50 can be found for these cells in Table 2.

Actomyosin Mg²+-ATPase activity. Maximum Ca²⁺-dependent actomyosin Mg²⁺-ATPase activity was determined from myofibrils isolated from whole hearts transiently treated with agonists, antagonists, and/or PKC inhibitors (Fig. 6). -Opioid or -adrenergic receptor agonists had significantly lower mean Ca²⁺-dependent actomyosin Mg²⁺-ATPase activity compared with myofibrils from untreated control hearts (Table 3). norBNI abolished the effects of -opioid receptor activation but had no effect by itself. Mean Ca²⁺-dependent actomyosin Mg²⁺-ATPase activity was significantly increased after treatment with the -adrenergic agonist isoproterenol. The receptor-independent PKC activator PMA reduced Ca²⁺-dependent actomyosin Mg²⁺-ATPase, whereas -PMA, the inactive form of PMA, had no effect on Ca²⁺-dependent actomyosin Mg²⁺-ATPase activity.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Langendorff-perfused heart protocol. Hearts were perfused for a total of 25 min. Some hearts underwent 2 min of agonist perfusion, 4 min norBNI perfusion, and/or 15 min perfusion with PKC inhibitors. After treatment, hearts were perfused with modified Krebs-Henseleit solution to wash out all drugs.

Ventricular ATP. Hearts treated with -opioid, -adrenergic, or -adrenergic agonists had significantly higher ventricular ATP compared with untreated control hearts (Fig. 7). Both norBNI (a -opioid receptor antagonist) and chelerythrine chloride (a PKC inhibitor) abolished the effects of -opioid receptor activation.

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Ventricular ATP levels of Langendorff-perfused hearts. Ventricular ATP levels were measured after 25 min of perfusion by a Langendorff perfusion apparatus. Hearts were pretreated with 1 µM U50,488H, 100 nM Iso, 10 µM phenylephrine (Phe) plus 3 µM Pro, 1 µM norBNI, and/or 2 µM chelerythrine chloride. Ventricles were cut from hearts, and ATP was determined using a luciferin-luciferase assay. Values are expressed as means ± SE. *P  0.05 vs. control.

Effect of recombinant PKC on actomyosin Mg²+- ATPase. Maximum Ca²⁺-dependent actomyosin Mg²⁺-ATPase activity of myofibrils from isolated untreated ventricular myocytes (control) was 164.0 ± 18.9 nmol P_i · min¹ · mg protein¹ (Fig. 8). Treatment of myocytes with U50,488H before myofibril isolation significantly reduced the maximum Ca²⁺-dependent actomyosin Mg²⁺-ATPase activity to 129.4 ± 15.9 nmol P_i · min¹ · mg protein¹. The maximum Ca²⁺-dependent actomyosin Mg²⁺-ATPase activity of control myofibrils incubated with recombinant PKC- was 136.4 ± 8.0 nmol P_i · min¹ · mg protein¹. Incubation with PKC- resulted in a maximum Ca²⁺-dependent actomyosin Mg²⁺-ATPase activity of 174.3 ± 5.9 nmol P_i · min¹ · mg protein¹. This was not significantly different than untreated myofibrils. Myofibrillar actomyosin ATPase was unaffected by incubation with heat-inactivated PKC- or PKC-.

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Effects of exogenous PKC on maximum Ca2+-dependent actomyosin Mg2+-ATPase activity. Enzymatically isolated ventricular myocytes were treated with 1 µM U50,488H or were untreated controls. Some myofibrils isolated from control myocytes were incubated with recombinant PKC-, PKC-, heat-inactivated PKC-, or heat-inactivated PKC-. Values are expressed as the means ± SE. *P  0.05 vs. control (Con).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In the present study, stimulation of -opioid receptors increased the Ca²⁺ sensitivity of isometric tension compared with untreated control ventricular myocytes at pH 7.0. This increase in the Ca²⁺ sensitivity of tension was similar to that observed with direct activation of PKC with PMA. Furthermore, the U50,488H-dependent increase in the Ca²⁺ sensitivity of isometric tension was abolished by the -opioid receptor antagonist norBNI and the PKC inhibitor chelerythrine chloride. These results are consistent with our hypothesis that the positive inotropy associated with -opioid receptor activation (33, 43) is due in part to a PKC-mediated increase in the myofilament Ca²⁺ sensitivity of tension. Decreasing pH from 7.0 to 6.6 resulted in decreased maximum tension and decreased Ca²⁺ sensitivity of isometric tension in control myocytes. Stimulation of -opioid receptors attenuated the pH-dependent decrease in the Ca²⁺ sensitivity of tension compared with untreated controls.

-Opioid receptor activation induces cytosolic alkalization (42) and increases intracellular calcium (30, 42). It has been suggested that either or both of these effects may account for the initial positive inotropy associated with -opioid receptor stimulation. Although the present study did not address the -opioid receptor-dependent effects on intracellular calcium or acid-base management, the results suggest that the -opioid receptor-dependent increase in tension development may be due in part to changes in myofilament Ca²⁺ sensitivity. The use of chemically demembranated myocytes allows for the experimental control of myofilament calcium and pH. Thus, in the absence of a -opioid receptor-dependent decrease in intracellular pH and/or increase in [Ca²⁺], isometric tension development at submaximal [Ca²⁺] was higher in myocytes treated with U50,488H compared with untreated control myocytes. These findings suggest that U50,488H-induced changes in the myofilaments contribute to the -opioid receptor-dependent increase in myocardial contractility.

Previous studies have established that acidosis is a useful tool in accentuating differences in myofilament Ca²⁺ sensitivity between experimental groups. For example, an acid challenge more clearly demonstrated changes in the Ca²⁺ sensitivity of tension due to pathological (24) and development (22) changes in the myofilaments. In the present study, we examined the effects of acidosis on the Ca²⁺ sensitivity of isometric tension and found that differences in the Ca²⁺ sensitivity of isometric tension between control and U50,488H-treated cells were not increased at pH 6.6. These findings strongly support the conclusion that a U50,488H-dependent change in cardiac myofilaments can, at most, account for a 8-10% increase in force of contraction at submaximal [Ca²⁺]. In addition, troponin I is an important pH-responsive protein in myocardium (44). We (33) have previously shown that -opioid receptor activation increases troponin I phosphorylation levels. Data from our current study show little difference in the pH-induced changes in myofilament Ca²⁺ sensitivity of tension with and without U50,488H treatment. These results are consistent with the hypothesis that the troponin I sites phosphorylated after -opioid receptor stimulation are not functionally connected to the pH-sensitive domain of troponin I.

Conflicting reports exist regarding the effect of stimulation of ₁-adrenergic receptors on Ca²⁺ sensitivity of isometric tension in ventricular myocytes. Puceat et al. (31) observed an increase, whereas Strang and Moss (38) saw no change in the Ca²⁺ sensitivity of tension after phenylephrine stimulation of ventricular myocytes. Under the experimental conditions of the present study, we observed no effect of stimulating ₁-adrenergic receptors on Ca²⁺ sensitivity of isometric tension in ventricular myocytes. It is well established that stimulation of the -adrenergic-PKA pathway decreases the Ca²⁺ sensitivity of isometric tension in myocardium (15, 31, 38). Our current observations are consistent with these past studies. -Adrenergic-dependent phosphorylation of troponin I is thought to account for the decrease the Ca²⁺ sensitivity of tension (14).

Our observation of differential effects on Ca²⁺ sensitivity of tension by stimulation of ₁-adrenergic receptors, -opioid receptors, and PMA is of interest because activation of PKC is the probable second messenger pathway utilized by each of these agents. This raises the possibility of PKC isoform functional specificity. Others (9, 32) have reported that activation of various neurohormonal receptors in cardiomyocytes selectively induce an increase in activation of different PKC isoforms. Furthermore, it has been demonstrated that isoforms of PKC can serve discrete functions within a cell (4, 13, 17). Our finding of an increased Ca²⁺ sensitivity of tension with some but not all purported activators of PKC is consistent with the hypothesis that specific isoforms of PKC have differential effects in cardiac myocytes.

One concern in the present study was the difference in control myocyte pCa₅₀ values between studies. The average pCa₅₀, pH 7.0, for control cells in Fig. 2 was 5.92 ± 0.03 (n = 12). The average pCa₅₀, pH 7.0, for control/DMSO cells in Fig. 3 was 5.71 ± 0.02 (n = 5). One possible cause of the difference is in the collagenase used to isolate myocytes. Data presented in Fig. 2 used cells isolated with type IV collagenase, whereas the myocytes in Fig. 3 were isolated with type I collagenase. Problems with Ca²⁺ contamination of the solutions probably do not account for the differences in control pCa₅₀, because the three sets of pCa solutions, pH 7.0, made with deionized water and salts from three different chemical suppliers all gave pCa₅₀ values of ~5.70 for myocytes isolated with the type I collagenase. In addition, the presence or absence of DMSO did not affect the pCa₅₀ values. Cells isolated with type I collagenase and treated with 1% DMSO had an average pCa₅₀ of 5.71 ± 0.02 (n = 5) compared with 5.74 ± 0.02 (n = 12) for non-DMSO-treated control cells. It should be emphasized the shift in control pCa₅₀ values between studies does not alter our findings of relative increases in the Ca²⁺ sensitivity of isometric tension with -opioid or PMA treatment compared with control myocytes from the same hearts.

In the present study, the -opioid receptor agonist U50,488H also decreased maximum actomyosin Mg²⁺-ATPase activity. This effect was abolished by the -opioid receptor antagonist norBNI. The U50,488H-dependent decrease in maximum actomyosin Mg²⁺-ATPase activity was mimicked by the known PKC activators phenylephrine and PMA and abolished by the PKC inhibitors chelerythrine chloride and bisindolylmaleimide. Exogenous PKC- was also able to reduce maximum actomyosin Mg²⁺-ATPase activity. The reduction in actomyosin Mg²⁺-ATPase activity was associated with an increase in whole heart ventricular ATP. This effect was also abolished with PKC inhibition. Together, these results suggest that the -opioid receptor-dependent reduction in maximum actomyosin Mg²⁺-ATPase activity is mediated through PKC--dependent myofibrillar alterations and that the PKC-dependent slowing of actomyosin Mg²⁺-ATPase activity slows the depletion of intracellular ATP stores.

Administration of exogenous PKC- did not reduce actomyosin Mg²⁺-ATPase activity. Jideama et al. (16) have previously demonstrated PKC- decreases actomyosin Mg²⁺-ATPase. This apparent inconsistency may be due to two methodological differences. First, myofibrils isolated from ventricular myocytes were used for the present studies, whereas Jideama et al. (16) used reconstituted myofibrils consisting of troponin I that had been phosphorylated by PKC- before reconstitution. Thus it is possible that the PKC- phosphorylation site on isolated troponin I is not readily accessible in intact myofilaments. Second, the amount of PKC- used in the present study was chosen to be approximately equal to the amount of myofibrillar Ca²⁺-independent PKC activity found in rat ventricular myocytes. Jideama et al. (16) make no mention of the amount of PKC- used in their study. A difference in the amount of PKC- and the subsequent level of troponin I phosphorylation and/or the myocardial preparations may explain the disparate results.

Clement et al. (8) have reported that exogenous PKC treatment does not alter maximum actomyosin Mg²⁺-ATPase activity. This finding contradicts the results of the present study. One possible reason for this discrepency may be the types of PKC utilized. Clement et al. (8) used PKC isolated from bovine brains, whereas the present study used only PKC- or PKC-. It is possible that the PKC isoforms in addition to PKC- and - found in the bovine brain may phosphorylate different myofilament proteins. The effects of PKC on actomyosin Mg²⁺-ATPase activity depends on which myofilament protein is phosphorylated by PKC (27, 28).

A variety of neurohormonal agents and transient ischemic protocols protect the heart against postischemic dysfunction or necrosis concomitant with an attenuated decline in intracellular ATP (26). Phosphocreatine reserves may slow ATP depletion, but the rapid decline of their stores in the early stages of myocardial ischemia minimizes the contribution of this mechanism to preserving ATP levels (2). Given that glycolytic abatement is a well-defined characteristic of ischemia, increased ATP production through increased glycolysis is also a doubtful consideration. The results of the present study indicate that in the whole heart a reduction in actomyosin Mg²⁺-ATPase activity is associated with higher levels of ventricular ATP. It is generally accepted that myocardial ATP levels remain unchanged in the normoxic heart. Our findings of increased ventricular ATP levels may be dependent on the protocol used. In the Langendorff-perfused heart preparation, glycogen levels may be depleted over the course of the experiment, thereby reducing ATP levels (11). It is conceivable that in the studies reported here ATP levels were reduced during the 20-min perfusion period and treatment with the -opioid receptor agonist slowed this ATP depletion (Fig. 7). It should be noted, however, that mean ATP levels from all hearts were within the normal range.

The cardiac myofibrillar protein that may mediate the effects of the receptor agonist-PKC pathway is difficult to identify. Activation of PKC has been associated with in vitro increased phosphorylation of a 15-kDa sarcolemma protein, a 28-kDa cytosolic protein (40), myosin light chain 2 (8, 41), C-protein (20), troponin I (18), and troponin T (18). Myosin light chain 2 phosphorylation levels have been associated with an increase in maximum actomyosin Mg²⁺-ATPase activity (27). However, myosin light chain 2 may not be an in vivo substrate for activated PKC (26).

The present study demonstrates that -opioid receptor activation of normoxic myocardium increased the Ca²⁺ sensitivity of isometric tension and decreased maximum Ca²⁺-dependent actomyosin Mg²⁺-ATPase. Strong support was obtained in indicating that these effects are mediated by PKC, with PKC- as the potential isoform involved. The implications of these findings on whole heart function in normal and diseased states include a -opioid-dependent increase in contractility at a given [Ca²⁺] under normal and acidotic conditions and an improved energy state of the heart through modulation of myofilament function. Thus we propose that -opioid receptor agonists and other neurohormonal agents that are cardioprotective act in part by decreasing actomyosin ATPase activity, which increases or conserves ATP levels such that critical ATP-dependent pumps and channels remain more fully active during and after ischemia.