INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CLINICALLY (26, 31) and experimentally (1, 8), chronic tachycardia can depress cardiac function and lead to congestive heart failure. Shorter periods (24 h) of rapid pacing (RP) can depress contractile function without overt signs of heart failure (40). These early changes in excitation-contraction (E-C) coupling may be important in the development of myocardial stunning. Classically, myocardial stunning is a reversible depression in contractile function that follows short periods of ischemia (6, 25). However, termination of atrial (20) or ventricular (23, 26) tachyarrhythmias also is followed by a reversible depression in contractile function, which is considered to be another form of myocardial stunning. The mechanisms by which short periods of tachyarrhythmia elicit myocardial stunning are much less understood than those responsible for chronic tachycardia-induced cardiomyopathy or ischemia-induced myocardial stunning. To date, experimental studies of RP-induced dysfunction have focused primarily on chronically paced in vivo heart preparations in an attempt to understand the complex changes that lead to congestive heart failure. Research into cellular mechanisms have used myocytes isolated from in vivo paced heart preparations (27, 34, 37). This approach, however, makes it difficult to distinguish between the direct effects of RP per se from the secondary in vivo changes that can affect contractile function. For example, in vivo models of chronic RP have documented changes in sympathetic nerve activity, alterations in the renin-angiotensin system and secretions of atrial natriuretic factor, reduced coronary perfusion, and ventricular remodeling (34). The present study indicates that a brief period of RP of ventricular myocytes depresses contractile function at the myofilament level via stimulation of a Ca²⁺/protein kinase C (PKC)-dependent signaling mechanism. These findings may be relevant to the inhibitory effects of paroxysmal ventricular tachyarrhythmia on cardiac contractile function and the development of pacing-induced cardiomyopathy. Portions of this work have been presented in abstract form (2).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Single cell isolation procedure. The methods used to isolate cardiac myocytes have been reported in detail (29). Briefly, adult cats are anesthetized with pentobarbital sodium (80 mg/kg ip). Isolated hearts were mounted on a Langendorff perfusion apparatus for enzymatic (0.07% type II collagenase; Worthington) cell isolation. After enzyme treatment, tissue obtained from the endocardial-midwall region of the left ventricular free wall was cut into small pieces and incubated in fresh enzyme solution. Isolated ventricular myocytes were stored in a solution of HEPES-Tyrode plus 0.1% albumin until use on the same day.

Recording techniques. Ventricular myocytes selected for study were elongated and relaxed and exhibited regular striations and normal action potential configurations when stimulated. Action potentials and ionic currents were recorded in the whole cell configuration (15) using a perforated (nystatin; 150 µg/ml)-patch method (17). Cells were superfused at 35 ± 1°C with external solutions containing (in mM) 137 NaCl, 5.4 KCl, 2.0 CaCl₂, 1.0 MgCl₂, 5 HEPES, and 11 glucose, which were titrated with NaOH to pH 7.35 and bubbled with 100% O₂. The internal pipette solution contained (in mM) 120 potassium glutamate, 20 KCl, 1.0 MgCl₂, 3 Na₂ATP, and 5 HEPES; pH 7.2. Isolation of the L-type Ca²⁺ current (I_Ca,L) was accomplished by replacing intrapipette K⁺ with cesium (Cs⁺) and adding 5 mM CsCl to external solutions to block K⁺ currents. Access resistance stabilized at 10-15 M within 5-10 min of forming a gigaseal.

RP protocol. The RP protocol consisted of the following: cells were electrically stimulated through the recording pipette at a cycle length of 1,000 ms (60 beats/min) until APD and cell shortening reached steady state, paced at progressively shorter cycle lengths (700, 500, and 400 ms) for steady-state times (60-90 s), and then held at 300-250 ms (200-240 beats/min) for 5 min. Progressive shortening of the pacing cycle length was used to allow APD to accommodate to the changing cycle length. If electromechanical alternans appeared, the pacing cycle length was increased a few milliseconds to a cycle length at which alternans did not appear. After pacing at the shortest cycle length, we progressively lengthened the pacing cycle length in reverse order back to control (1,000 ms) (see Fig. 1). At least 3 min of pacing at cycle lengths  300 ms was required to elicit consistent RP-induced inhibition of cell shortening. Measurements of peak cell shortening were determined as an average of the last five beats at each pacing cycle length using a custom software program (LabView). The RP protocol was always accomplished by stimulated action potentials. Because RP induces sustained changes in cell shortening, in some experiments a single cell could not serve as its own control. In this case, RP-induced inhibition of cell shortening was studied in two groups of cells from the same hearts: control and test group cells.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Short-term rapid pacing (RP) inhibits contraction of ventricular myocytes. A: starting at a pacing cycle length (CL) of 1,000 ms, CL was progressively decreased (descending), held at 250 ms for 5 min, and then progressively lengthened (ascending) back to control (1,000 ms). After RP, cell shortening was inhibited at each pacing CL. B: original recordings of action potentials (AP) and cell shortening recorded from a myocyte stimulated at 1,000 ms before (control) and after RP. After RP, peak cell shortening was decreased without changes in resting cell length.

Measurement of intercellular Ca²⁺ concentration. Single ventricular myocytes were loaded with fluorescent Ca²⁺ indicator by exposure to 5 µM acetoxymethyl esters (AM) of indo 1 (indo 1-AM; Molecular Probes, Eugene, OR) for 20 min at 20°C. For fluorescence measurements, a coverslip of cells was mounted on the stage of an inverted microscope. Indo-1 fluorescence was excited at 360 nm, and the Ca²⁺-dependent changes of fluorescence emitted from single cells was recorded simultaneously at 405 nm (F₄₀₅) and 485 nm (F₄₈₅), respectively, with photomultiplier tubes. Changes of the intracellular Ca²⁺ concentration ([Ca²⁺]_i) are expressed as changes in the ratio (R = F₄₀₅/F₄₈₅).

Analysis of PKC isoenzyme expression by Western blotting. Acutely isolated cat ventricular myocytes were centrifuged (1,000 g for 10 min) and resuspended in lysis buffer [20 mM Tris · HCl (pH 7.5) containing 0.5 mM EGTA, 0.5 mM EDTA, 10 mM mercaptoethanol, 0.5% Triton X-100, 1 mM sodium vanadate, 10 µg/ml leupeptin, 10 µg/ml aprotonin, and 1 mM Pefabloc]. After sonication and repeat centrifugation, we assessed the protein content of the supernatant fraction using a bicinchoninic acid assay (Pierce, Rockford, IL), and 100-300 µg of extracted protein were separated by SDS-PAGE and Western blotting. Separated proteins were probed with specific monoclonal antibodies to PKC-, -, -, and - (Transduction Laboratories, Lexington, KY). Protein bands were visualized using an enhanced chemiluminescence method (Amersham, Arlington Heights, IL). Similarly prepared tissue extracts of the rat brain (50 µg) and cat brain (50 µg) served as positive controls.

Drugs. The drugs used included the following: chelerythrine (Alexis, San Diego, CA), calpain inhibitor I (Calbiochem, La Jolla, CA), L-N⁵-(1-iminoethyl)-orthinthine (L-NIO) (Alexis, San Diego, CA), N^G-monomethyl-L-arginine (L-NMMA) (Alexis), and LY-379196 (generously provided by Dr. Chris Vlahos, Eli Lilly Research Laboratories). Generally, cells under study were exposed to drugs for ~5 min before the RP protocol was performed unless stated otherwise.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Figure 1, A and B, shows the effects of electrical pacing on unloaded cell shortening (contraction) of single ventricular myocytes. In Fig. 1A, a myocyte was initially paced at a cycle length of 1,000 ms (60 beats/min). Cycle length was progressively decreased to 250 ms (240 beats/min), maintained there for 5 min, and then progressively returned to 1,000 ms (see METHODS). After RP at 250 ms for 5 min, peak (systolic) cell shortening was decreased at each pacing cycle length. Diastolic cell length was not significantly different before and after RP. Figure 1B shows an original recording of action potential configuration and cell shortening before and after RP in the same ventricular myocyte. RP-induced inhibition of cell shortening were typically accompanied by a decrease in APD₉₀. In 26 cells, RP decreased APD₉₀ by 9 ± 3% (P < 0.05). However, RP-induced changes in APD₉₀ were variable among cells. In other words, it was not uncommon for RP to exert little to no effect on APD₉₀ at the same time that it markedly decreased cell shortening. This is reflected in the finding that RP-induced changes in peak cell shortening and APD₉₀ did not show a significant correlation (r = 0.17; P = 0.406). Experiments described below will show that RP-induced inhibition of cell shortening was, in fact, independent of membrane voltage. In a total of 43 cells obtained from eight hearts, RP decreased peak cell shortening by 44 ± 4% (P < 0.001).

RP-induced shortening of APD₉₀ could result from activation of ATP-sensitive K⁺ channels, indicating cellular hypoxia. However, exposure to 5 µM glibenclamide, a specific inhibitor of ATP-sensitive K⁺ channels (11), failed to prevent RP-induced shortening of APD₉₀ [control 16 ± 7% (n = 6) vs. glibenclamide 14 ± 6% (n = 7)]. This result is consistent with reports that RP of unloaded cardiac myocytes does not cause hypoxia (45). RP-induced changes in cell shortening and APD₉₀ may result from a nonspecific time-dependent rundown in cell viability. This possibility was examined by stimulating cells at 1,000 ms without RP for the same time period as the RP protocol (~15 min). There were no discernable changes in cell shortening or APD₉₀.

To determine whether changes in membrane voltage induced by RP (such as shortening of APD) may be responsible for inhibition of cell shortening, we used constant depolarizing voltage-clamp pulses to trigger cell shortening before and then after RP in the same myocytes. During voltage clamp, each cell was held at its respective resting membrane voltage (75 to 80 mV) and stepped to +10 mV for 200 ms at 1 Hz. Cell shortening was measured before and after RP when triggered by either action potentials or voltage-clamp pulses at 1 Hz. The graph in Fig. 2 summarizes the results obtained in a total of eight cells obtained from two hearts. Generally, the amplitude of cell shortening under control conditions was smaller, although not significantly (P = 0.5), when triggered by a voltage-clamp pulse than by an action potential. Nevertheless, RP inhibited cell shortening to a similar extent when triggered by action potentials (52 ± 8%) or by depolarizing voltage-clamp pulses (48 ± 10%). These findings indicate that potential changes in membrane voltage induced by RP are not responsible for the inhibition of contractility induced by RP.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   RP-induced inhibition of contraction is independent of changes in membrane voltage. Cell shortening was elicited by either AP or constant voltage-clamp pulses (clamp) before and after RP in the same cells. After RP, cell shortening was inhibited when stimulated by either AP or clamp pulses, showing that RP-induced inhibition of contraction is not due to changes in membrane potential induced by RP.

I_Ca,L plays a critical role in triggering cell shortening and may influence APD₉₀. We therefore used voltage clamp to determine whether RP inhibited I_Ca,L. I_Ca,L was measured by voltage clamping the same cell before and then after imposing the RP protocol. As shown in Fig. 3, in cells obtained from three hearts, RP had no affect on peak I_Ca,L density (control 8.9 ± 1.8 vs. after RP 9.0 ± 1.9 pA/pF; n = 13) or the rapid (₁) and slow (₂) time constant of I_Ca,L inactivation (control ₁ = 5 ± 0.8 and ₂ = 44 ± 3 ms vs. after RP ₁ = 6 ± 0.8 and ₂ = 45 ± 5 ms; n = 4).

View larger version (9K): [in this window] [in a new window]   Fig. 3.   L-type Ca2+ current (ICa,L) is unchanged after RP. Original recordings of ICa,L before (left) and after RP (right) in the same ventricular myocyte. RP had no discernible effects on holding current, peak Ca2+ current amplitude, or time course of inactivation.

A key step in cardiac E-C coupling is Ca²⁺ release from the sarcoplasmic reticulum (SR). Therefore, fluorescent microscopy and the Ca²⁺-sensitive dye indo 1-AM were used to determine whether RP altered the handling of intracellular Ca²⁺. Figure 4 shows typical measurements of [Ca²⁺]_i obtained before, during, and after RP. This cell was rapidly paced at 225 beats/min for 6 min. During RP, baseline and peak [Ca²⁺]_i were prominently elevated compared with control, resulting in an increase in mean [Ca²⁺]_i. After RP baseline, [Ca²⁺]_i and the peak Ca²⁺ transient amplitude returned to control values. Moreover, there was no change in diastolic [Ca²⁺]_i or the time course of relaxation of the Ca²⁺ transient, suggesting that SR Ca²⁺ uptake was unchanged. In five cells obtained from two hearts, during RP baseline, [Ca²⁺]_i increased from 0.33 ± 0.03 (control) to 0.41 ± 0.03 (+24%; P < 0.05). These results indicate that, during RP, time-averaged [Ca²⁺]_i is significantly elevated, and, after RP, SR Ca²⁺ release and reuptake are not significantly changed. In addition, neither during nor after RP did we observe any signs that cells were Ca²⁺ overloaded or damaged. In other words, RP never induced Ca²⁺-mediated afterdepolarizations, Ca²⁺ waves, or spontaneous Ca²⁺ transients. Also, cells did not exhibit blebbing or become granular as a result of RP.

View larger version (47K): [in this window] [in a new window]   Fig. 4.   Intracellular Ca2+ concentration ([Ca2+]i) measurements recorded from a ventricular myocyte before, during, and after RP. Top: stimulated AP. Bottom: [Ca2+]i transients. During RP, baseline and peak [Ca2+]i increased. After RP, baseline and peak [Ca2+]i returned to control values. Vm, membrane voltage; F405 and F480, fluorescence at 405 and 480 nm, respectively.

RP-induced increases in Ca²⁺ influx could, in principle, activate a number of Ca²⁺-mediated signaling pathways that may underlie RP-induced inhibition of cell shortening. Therefore, we tested the role of Ca²⁺ influx by lowering the extracellular Ca²⁺ concentration ([Ca]_o) from 2 to 0.5 mM during the RP protocol. Measurements of cell shortening were obtained in two groups of cells (control and low [Ca]_o) from the same two hearts before and after RP when the [Ca]_o = 2 mM. In control cells, RP elicited a typical decrease in cell shortening (3.6 ± 0.8 vs. 1.8 ± 0.4 µm; 48%) (n = 5). Figure 5, A-C, shows selected recordings of action potentials and cell shortening before, during, and after RP, when the [Ca]_o was reduced to 0.5 mM during RP. Comparing cell shortening before and after RP, it is evident that RP failed to decrease cell shortening (4.7 ± 0.6 vs. 4.6 ± 0.7 µm; 3 ± 4%) (n = 6). In addition, measurements of cell shortening during RP in normal [Ca]_o (2.9 ± 0.5 µm) and low [Ca]_o (1.3 ± 0.4 µm) (P < 0.05) indicate that cell shortening was significantly smaller in cells paced in low [Ca]_o. This is consistent with a reduced Ca²⁺ influx during RP in low [Ca]_o. Moreover, the records (Fig. 5B) show that lowering [Ca]_o did not interfere with excitation at high frequencies of stimulation. These results, in conjunction with those in Fig. 4, indicate that RP-induced increases in Ca²⁺ influx and subsequent elevation of [Ca²⁺]_i is an essential factor in the mechanism by which RP inhibits cell shortening.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   RP-induced inhibition of contraction is dependent on Ca2+ influx. AP and cell shortenings recorded before (A), during (B), and after (C) RP at a CL of 300 ms. During RP, the extracellular calcium concentration ([Ca]o) was reduced from 2 to 0.5 mM. Lowering [Ca]o during RP prevented RP-induced inhibition of cell shortening.

In ischemic myocardial stunning, elevation of [Ca²⁺]_i is thought to activate the Ca²⁺-sensitive protease calpain, resulting in degradation of contractile myofibrillar protein (9, 13). To determine whether a similar mechanism may operate in the contractile dysfunction induced by RP, we performed the RP protocol in cells exposed to calpain inhibitor I, a cell-permeant inhibitor of calpain (30). The experiments were performed by either exposing cells to 1 µM calpain inhibitor I acutely (5-10 min) or by incubating cells in the inhibitor for up to 1 h before the RP protocol was initiated. Because similar results were obtained with the two methods, the data have been pooled. Calpain inhibitor I failed to affect RP-induced inhibition of cell shortening (control 37 ± 11% vs. calpain inhibitor I 39 ± 7%; n = 6 cells from 2 hearts).

In cardiac myocytes, constitutive nitric oxide (NO) synthase activity is stimulated by RP-induced elevation of [Ca²⁺]_i (21), and NO can inhibit cardiac contractility (5, 21). To assess the potential role of NO, we performed two series of experiments to test the effects of 25 µM L-NIO and 100 µM L-NMMA, both inhibitors of NO synthase (28). Previous work from this laboratory (43, 44) have shown that both L-NIO and L-NMMA inhibit NO-mediated processes in cat atrial myocytes. In cells from the same three hearts, L-NIO (control 47 ± 10% vs. L-NIO 48 ± 8%; n = 11) or L-NMMA (control 69 ± 22% vs. L-NMMA 60 ± 21%; n = 6) both failed to prevent RP-induced decreases in cell shortening, suggesting that NO is not mediating the inhibitory effects of RP on contractility.

Activation of PKC is an important Ca²⁺-dependent signaling mechanism, which is known to be associated with negative inotropic effects in the heart (38, 41, 42). We therefore examined the role of PKC signaling by performing the RP protocol in two groups of cells from the same four hearts in the absence and presence of 2-4 µM chelerythrine, a specific nonselective PKC inhibitor (16). In the test cell group, chelerythrine alone had no significant effect on peak cell shortening. When compared with control cells, however, chelerythrine significantly attenuated RP-induced decreases in cell shortening (control 59 ± 7%, n = 8, vs. chelerythrine 14 ± 3%, n = 9) (P < 0.001).

In the transgenic mouse, overexpression of PKC- decreases myofilament Ca²⁺ responsiveness and contractility, possibly via phosphorylation of troponin I (38). Therefore, to assess more specifically the role of the Ca²⁺-dependent isoenzyme PKC-, we tested the effects of 30 nM LY-379196, a selective inhibitor of PKC- (33). At 30 nM, the specificity of LY-379196 for inhibition of PKC- is more than an order of magnitude greater than for inhibition of other PKC isoenzymes (33). Figure 6, A and B, shows original recordings of action potentials and cell shortenings obtained from two different myocytes. In a control cell (Fig. 6A), RP markedly decreased cell shortening (81%) and slightly shortened APD₉₀ (7%). In the test cell group, LY-379196 alone elicited a small increase in basal cell shortening (+18 ± 9%; n = 7). In a cell exposed to LY-379196 (Fig. 6B), RP-induced inhibition of cell shortening was abolished, whereas APD₉₀ still shortened (13%). In cells obtained from the same two hearts, LY-379196 blocked RP-induced inhibition of cell shortening (control 69 ± 11%, n = 5, vs. LY-379196 9 ± 4%, n = 7) (P < 0.001). To further evaluate the role of PKC activation, we directly stimulated phorbol ester-sensitive PKC isoenzymes by exposure to 0.2 µM phorbol 12-myristate 13-acetate (PMA) for 5 min. PMA consistently and significantly decreased peak cell shortening (46 ± 7%; P < 0.01; n = 5) much the same as RP (data not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Inhibition of protein kinase C (PKC)- abolishes RP-induced inhibition of contraction. A: control recordings of AP and cell shortening before and after RP. RP markedly inhibited cell shortening. B: recordings of AP and cell shortening from another myocyte in the presence of LY-379196, a selective inhibitor of PKC-. RP-induced inhibition of cell shortening was abolished.

Exactly which PKC isoenzymes are expressed in the heart is somewhat controversial (39), and feline cardiac PKC isoenzyme expression has not been previously evaluated. Therefore, cellular extracts of cat ventricular myocytes were probed with monoclonal antibodies specific for PKC-, -, -, and - , the major PKC isozymes reported to be present in ventricular myocytes of several different species, including humans (37). As shown in Fig. 7, PKC- (A), PKC- (C), and PKC- (D) were all detected in cell extracts of freshly isolated myocytes obtained from four hearts. In addition, as shown in Fig. 7B, an 80-kDa band was detected, which comigrated with PKC- isolated from both the cat and rat brain. These results are consistent with the presence of PKC- (as well as PKC-, -, and -) in adult cat ventricular myocytes.

View larger version (58K): [in this window] [in a new window]   Fig. 7.   PKC isoenzyme expression in cat ventricular myocytes. Detergent-extracted cellular proteins from isolated cat ventricular myocytes (100-300 µg) and cat brain (50 µg) and rat brain (50 µg) were separated by SDS-PAGE and Western blotting. The resulting blot was probed with mouse monoclonal antibodies generated against human PKC- (A), PKC- (B), PKC- (C), and PKC- (D). Bands were detected by an enhanced chemiluminescence protocol. The position of molecular weight standards is depicted to the left of each blot.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main finding of the present study is that RP of ventricular myocytes for just a few minutes elicits depression of systolic contractile function via activation of a Ca²⁺/PKC-dependent signaling mechanism. Moreover, the depression in contractile function that follows brief RP cannot be explained by alteration in the basic handling of intracellular Ca²⁺ during E-C coupling. More specifically, RP-induced inhibition of contraction was not associated with changes in 1) the trigger for SR Ca²⁺ release, i.e., Ca²⁺ influx via I_Ca,L; 2) SR Ca²⁺ release as determined from peak intracellular Ca²⁺ transient amplitude; or 3) reuptake of SR Ca²⁺, as determined from relaxation of the intracellular Ca²⁺ transient. Baseline levels of [Ca²⁺]_i also were unchanged after RP, indicating that diastolic [Ca²⁺]_i was unaffected. This is consistent with our finding that RP did not significantly affect diastolic cell length. Although RP generally shortened APD₉₀, there was no significant correlation between changes in APD₉₀ and peak contraction. In addition, the use of voltage-clamp pulses to elicit contraction also directly demonstrated that RP-induced inhibition of contraction is independent of RP-induced changes in action potential configuration. These results lead to the conclusion that RP-induced inhibition of contractile function occurs at the level of the contractile myofilaments. In contrast to the present results, chronically paced hearts, with (1, 8) or without (40) signs of heart failure, exhibit contractile dysfunction, which is associated with significant alterations in [Ca²⁺]_i regulation and E-C coupling. At the cellular level, ventricular myocytes isolated from hearts chronically paced in vivo exhibit more positive resting membrane potentials, triangular action potential configurations, prolonged APD, decreased I_Ca,L density, and alterations in cytoarchitecture; all of which are thought to contribute to contractile dysfunction (34). In contrast, various models of myocardial stunning (10, 12, 13, 32) exhibit impaired contractile function that is not associated with changes in basic E-C coupling mechanisms. For example, in ischemic myocardial stunning, I_Ca,L density and SR Ca²⁺ release and reuptake are unchanged, whereas the responsiveness to Ca²⁺ is attenuated, indicating that contractility is affected at the myofilament level.

The present results indicate that elevation of [Ca²⁺]_i during RP is an essential factor underlying the inhibitory effects of RP. Thus lowering [Ca]_o, and thereby Ca²⁺ influx, specifically during RP prevented RP-induced inhibition of contraction. In a variety of experimental models, elevated [Ca²⁺]_i is a key factor underlying myocardial stunning (7, 12, 22-24). For instance, ischemic stunning is attenuated when [Ca]_o is lowered in the reperfusing solutions (24). Conversely, exposure to high [Ca]_o, even in the absence of ischemia, elicits contractile dysfunction (22). Moreover, in a perfused heart preparation, elevated [Ca²⁺]_i induced during ventricular fibrillation, again in the absence of ischemia, is also thought to be responsible for the contractile dysfunction that follows termination of this arrhythmia (23). Elevated [Ca²⁺]_i can affect a variety of substrates and signaling mechanisms that could potentially result in contractile dysfunction. In ischemic-induced stunning, elevated [Ca]_i is thought to activate the Ca²⁺-sensitive protease calpain, which degrades troponin I, thereby reducing myofilament Ca²⁺ responsiveness (13, 14, 39). In the present study, calpain inhibitor I failed to prevent RP-induced inhibition of contraction, suggesting that stimulation of Ca²⁺-activated calpain is not an underlying mechanism here. Likewise, inhibition of Ca²⁺-dependent NO synthase by either L-NIO or L-NMMA failed to prevent RP-induced inhibition of contraction, indicating that pacing-induced stimulation of NO (21) also is not a key factor.

Several of the present findings, however, indicate that RP depresses contractile function by activating a PKC-dependent signal transduction pathway. Thus RP-induced inhibition of contraction was blocked by both nonselective inhibition of PKC (chelerythrine) as well as selective inhibition of the Ca²⁺-dependent PKC- isoenzyme (LY-379196). Western blots also demonstrated that cat ventricular myocytes express the PKC- isoenzyme. In addition, acute stimulation of phorbol ester-sensitive PKC isoenzymes with PMA also inhibited contraction. This latter finding, however, must be interpreted cautiously because PMA nonselectively activates a broad range of PKC isoenzymes, which could affect contraction via a variety of mechanisms. The present results indicate that RP-induced inhibition of contraction is dependent on both Ca²⁺ and PKC-dependent signaling. Perhaps the simplest explanation is that Ca²⁺ influx during RP activates Ca²⁺-dependent PKC isoenzymes, possibly PKC-. However, Ca²⁺ influx may also stimulate phospholipases to activate other Ca²⁺-independent PKC isoenzymes. For example, we recently demonstrated that brief RP (~5 min) by field stimulation of cultured neonatal rat ventricular myocytes activated the novel, Ca²⁺-independent PKC isoenzymes PKC- and PKC-, without activating the Ca²⁺-dependent PKC isoenzyme PKC- (36). It should be noted that neonatal rat ventricular myocytes do not express PKC- (35). Nevertheless, these results support the present findings that brief periods of RP activate PKC-dependent signaling.

With regard to our initial explanation, increased expression of PKC- is consistently associated with the contractile dysfunction of cardiomyopathic and failing hearts. For instance, in streptozotocin-induced myopathic hearts, the expression of PKC- is preferentially increased over other PKC isoenzymes (18). In failing human hearts, expression of PKC- was significantly increased compared with nonfailing hearts (3). Moreover, blocking PKC- with LY-333531, a selective PKC- blocker and close analog of LY-379196, indicated that PKC- contributed the greatest amount to the total increase in PKC activity in failing hearts. Transgenic expression of PKC- in adult mice hearts caused mild and progressive ventricular hypertrophy (4, 42). In addition, troponin I is a substrate for PKC phosphorylation, and phosphorylation of troponin I decreases myofilament Ca²⁺ responsiveness (19, 41). In transgenic mouse hearts, the specific overexpression of PKC- decreases myofilament Ca²⁺ responsiveness and contractility, presumably via phosphorylation of troponin I (38). The present results therefore suggest that by raising [Ca²⁺]_i, RP activates a Ca²⁺/PKC-dependent signaling mechanism, possibly via PKC-, which subsequently depresses myofilament Ca²⁺ responsiveness. Moreover, the early stimulation of PKC signaling by RP may set in motion further downstream signaling mechanisms that lead to the pathological changes characteristic of chronic tachycardia-induced cardiomyopathy. Further studies, however, involving direct measurements of PKC activation and/or translocation as well as protein phosphorylation of troponin I will be required to prove our hypothesis.