Introduction

Top Abstract Introduction Methods Results Discussion References

The early work of Trautwein and Dudel (1958) revealed that atrial, but not ventricular, myocytes were sensitive to the neurotransmitter acetylcholine. ACh decreases heart rate, slows atrial-ventricular conduction, reduces atrial contractility and attenuates catecholamine-stimulated effects (Loffelholz and Pappano, 1985). All these effects predominately arise from membrane hyperpolarization via an activation of I_K(ACh). Patch-clamp studies have demonstrated the direct activation of I_K(ACh) by application of ACh in mammalian heart cells (Sakmann et al., 1983). Boyett and his colleagues (Boyett et al., 1988; McMorn et al., 1993) reported that ACh has a direct chronotropic and negative inotropic effect on ferret and rat ventricular myocytes. They also found that I_K(ACh) is present in both atrial and ventricular myocytes. Further, Koumi and Wasserstrom (1994) identified and characterized I_K(ACh) in isolated cat, guinea pig and human ventricular myocytes. These data indicate that I_K(ACh) is present in ventricular myocytes and may play important roles in cardiac function.

Cocaine abuse has presented a major health problem in many countries for more than a century. In North America, the incidence of cocaine use has climbed rapidly in recent years, especially in young adults. Serious medical consequences of cocaine cardiotoxicity, including myocardial ischemia and infarction, ventricular fibrillation and sudden cardiac death, are serious health issues. Existing data show that reduced cardiac vagal tone increases susceptibility to ventricular arrhythmias (Corr and Gillis, 1974; Vanoli et al., 1991). Early studies demonstrated (Wilkerson, 1989) that atropine significantly enhanced cocaine-induced cardiovascular toxicity. In humans cocaine also suppressed cardiac vagal tone in Newlin's recent study (1994). Furthermore, cocaine inhibited ion flux controlled by the ACh receptor in membrane vesicles of Torpedo californica, Electrophorus electricus and in PC-12 cells, a sympathetic neuronal cell line (Karpen et al., 1982; Karpen and Hess, 1986). Binding studies indicate that cocaine acts as an antimuscarinic agent, particularly at higher doses, in heart and brain (Sharkey et al., 1988). A recent report shows that cocaine interacts with primary and allosteric recognition sites on muscarinic receptors in membrane homogenates from postmortem human brainstem (Flynn et al., 1992). The hypothesis of cocaine as an antagonist of cardiac muscarinic receptors is further confirmed by other studies (Ritz and George, 1993; Tan and Costa, 1994).

Cardiac intoxication with cocaine has been linked to the inhibition of cardiac neuronal uptake of norepinephrine and its local anesthetic effects on Na⁺ channels. The effects of cocaine on the cholinergic system of the heart have not been well investigated, however, especially at the cellular and molecular levels. Because muscarinic receptors play a crucial role in modulation of heart rate and stabilization of membrane electrical excitability (Billman and Hoskins, 1989), it is important to evaluate the effects of cocaine on cardiac I_K(ACh). Here, we report that cocaine strongly suppresses the K⁺ current activated by carbachol, adenosine and GTPS in adult ferret atrial and left ventricular myocytes. The results suggest that cocaine-induced inhibition of I_K(ACh) involves block of the channel pore or the receptor binding site, as well as an inhibition of the G-protein (G_K). Some of the data in this manuscript have been presented in brief abstract form (Xiao and Morgan, 1996).

	Methods

Top Abstract Introduction Methods Results Discussion References

Isolation of single myocytes. Single atrial and left ventricular myocytes were enzymatically isolated from adult ferret (male, 6-10 weeks of age, Marshall Farm, North Rose, NY) hearts by a method similar to that described previously (Xiao and McArdle, 1994). Animal care and experimental procedures were performed according to the guidelines of Institutional Animal Care and Use Committee in compliance with the US Public Health Service Policy as stated in The Guide for the Care and Use of Laboratory Animals (HHS, NIH Publication No. 85-23, 1985). After a ferret was deeply anesthetized with chloroform, the heart was rapidly removed and washed in ice-cold, oxygenated Tyrode's solution containing (in mM): NaCl,137; CaCl₂, 2; KCl, 5; MgCl₂, 1; CaCl₂, 1.8; HEPES, 10; glucose, 10, pH 7.4. The aorta was quickly connected to a modified Langendorff system. This perfusion system had a hydrostatic pressure of 80 cm and a flow rate of 8 to 10 ml/min. The heart was initially perfused with the oxygenated 37°C Ca⁺⁺-free Tyrode's solution for 6 min. The heart was then perfused and recirculated for 38 to 45 min with 50 ml Ca⁺⁺-free Tyrode's solution containing 45 to 50 mg collagenase (CLS 2, Worthington Biomedical Corporation, Freehold, NJ), 1 to 2 mg protease Type XIV and 0.1% bovine serum albumin (Sigma Chemical Company, St. Louis, MO). After the enzyme treatment, the heart was washed sequentially with 50 ml 0.2 mM Ca⁺⁺ and 50 ml 0.4 mM Ca⁺⁺ Tyrode's solution plus 1 mg/ml bovine serum albumin. When the enzymatic solution was completely washed out, several pieces of atrial and left ventricular tissue were cut off and placed separately into two Petri dishes (60 × 15 mm) containing 0.4 mM Ca⁺⁺ Tyrode's solution plus 1 mg/ml bovine serum albumin. The cardiac tissue was further sliced into finer pieces and gently agitated for 1 to 2 min. The dispersed cells and tissues were filtered through a 250-µm polypropylene mesh. The effluent containing dissociated myocytes was kept in 0.4 mM Ca⁺⁺ Tyrode's solution plus 1 mg/ml bovine serum albumin at 22-23°C room temperature for 1 to 2 hr before beginning patch-clamp experiments.

Recording of ACh-activated whole-cell currents. A small volume (25 µl) of the effluent solution containing cardiac myocytes was pipetted into a chamber (0.5 ml) with a coverglass bottom mounted on an inverted microscope and superfused (1-2 ml/min) with the Tyrode's solution. Recording pipettes were made from glass tubes (World Precision Instruments, Inc., Sarasota, FL) by a two-stage pull on a David Kopf vertical puller (model 700D, Tujunga, CA). The heat-polished or unpolished electrode with a resistance of 2 to 4 megohm was connected via a Ag-AgCl wire to an Axopatch 1D amplifier (Axon Instruments, Foster City, CA). After forming a conventional "gigaseal" the capacitance of the electrode was compensated. Additional suction was used to form the whole-cell configuration. With the nystatin-perforated patch technique no further suction was applied after forming a conventional high-resistance seal. The whole-cell recording configuration was formed within 15 min (8.7 ± 1.1 min, n = 24) after the gigaseal. A current representing the membrane capacitance was recorded by application of a 5-mV hyperpolarizing pulse from a holding potential of 80 mV (Xiao and McArdle, 1994). Correction of cell capacitance and series resistance was then performed before application of the experimental voltage-clamp protocol. External solutions were exchanged with a fast perfusion system (Xiao et al., 1995). I_K(ACh) was evoked by extracellular application of 1 µM carbachol and stored on the hard disk of a personal computer running the Pclamp 5.5.1 data acquisition programs. All experiments were carried out in the 2 mM Ca⁺⁺ Tyrode's solution at 22-23°C. The pipette solution for the nystatin-perforated patch method (Horn and Marty, 1988) contained (in mM): KCl, 140; EGTA, 0.5; HEPES, 5, pH 7.3; and 100 µg/ml nystatin which was freshly added before each experiment. The intracellular solution for the classic whole-cell recording method (Hamill et al., 1981) contained (in mM): KCl, 80; KOH, 60; MgCl₂, 1; CaCl₂, 1; EGTA, 10; HEPES, 10; MgATP, 5; and pH 7.3.

Recording of the action potential. Electrically stimulated action potentials of ferret atrial cells were recorded with the current-clamp method described elsewhere (Xiao and McArdle, 1995). The resting membrane potential was approximately 80 mV after obtaining the classical whole-cell recording configuration. Action potentials were elicited by intracellular injection of 30 to 50 pA depolarizing current for 10 ms.

Materials. Cocaine and MEG were obtained from Sigma (St. Louis, MO) and dissolved weekly in deionized water at a concentration of 10 mM and stored at 20°C before use. The experimental concentrations of cocaine and MEG were obtained by dilution of the stocks. Carbachol, nystatin, adenosine, ATP and GTPS were obtained from Sigma (St. Louis, MO).

Data analysis and statistics. I_K(ACh) was recorded at different holding potentials and measured as the amplitude of the carbachol-activated current minus the steady-state holding current. All myocytes were held at 80 mV for 1 to 2 min before changing holding potential and reapplication of carbachol. Recordings of I_K(ACh) were made from the same myocytes before, during and after drug treatment. A washout was conducted to determine whether I_K(ACh) returned toward the pretreatment value and to ensure that changes of I_K(ACh) were not caused by functional damage. The current density was calculated as the current amplitude divided by the cell membrane capacitance to exclude the effect of different cell size. Data are presented as the mean ± S.E. The Student's t test was used to evaluate the difference between two values. One-way analysis of variance and Dunnett's test for critical difference were used for data derived from more than two groups. A difference of P < .05 was considered statistically significant.

	Results

Top Abstract Introduction Methods Results Discussion References

I_K(ACh) of ferret cardiomyocytes. Both atrial and left ventricular myocytes isolated from ferret hearts have clear striations, but atrial cells appear spindle-shaped and narrower in width than left ventricular myocytes, which are rod-shaped. Almost all cells examined in this study were quiescent before and after forming a whole-cell configuration when the holding potential (V_h) was set at 80 mV. The resting membrane potential measured by the zero-current clamp method was 80 ± 1 and 81 ± 3 mV for the atrial (n = 13) and left ventricular (n = 6) myocytes, respectively. I_K(ACh) was evoked by rapid application of carbachol and recorded via a glass electrode with the nystatin perforated-patch technique. With this technique I_K(ACh) was observed in each patch after application of 1 µM carbachol to either atrial or left ventricular myocytes.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Comparison of IK(ACh) between ferret atrial and left ventricular myocytes. Original current traces recorded from an atrial (A) and a left ventricular (B) myocyte. IK(ACh) was activated by extracellular application of 1 µM carbachol and recorded at a holding potential of 0 mV with the nystatin-perforated patch-clamp method. (C) Comparison of the current density between atrial and ventricular myocytes. The current density is calculated by the amplitude of IK(ACh) divided by the cell membrane capacitance. ****P < .0001; between the atrial (open bar, A. cell, n = 21) and left ventricular (hatched bar, V. cell, n = 12) myocytes.

Fade of I_K(ACh). The classic whole-cell recording technique can cause "wash-out" of the muscarinic response to ACh (Horn and Marty, 1988). Therefore, the effects of the nystatin perforated-patch method and the classic whole-cell recording technique, with or without intracellular dialysis 5 mM ATP, were compared. Figure 2 (open circles) shows that I_K(ACh) evoked by repeated application of carbachol did not significantly fade in the nystatin perforated-patch myocytes (n = 8) during 60 min. By use of the classic whole-cell recording method, the amplitude of I_K(ACh) decreased by 23 ± 5% (n = 9) of the control within 30 min when the intracellular solution contained 5 mM ATP (open triangles). In contrast, I_K(ACh) was significantly reduced (P < .01, n = 5) within 5 min and only left 6 ± 4% of the control current within 30 min in the myocytes intracellularly dialyzed with ATP-free solution (closed triangles).

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Fade of IK(ACh) in ferret atrial cells. Currents were recorded at 0 mV holding potential and normalized by the value of the first exposure to 1 µM carbachol. IK(ACh) was elicited by repeated application of 1 µM carbachol and recorded by use of the nystatin-perforated patch technique () and the classic whole-cell recording method with () and without () intracellular dialysis of 5 mM MgATP.

Cocaine blockade of I_K(ACh). Binding data have shown that cocaine blocked muscarinic cholinergic receptors in heart and brain (Sharkey et al., 1988). Furthermore, atropine enhances the cardiovascular effects of cocaine (Wilkerson, 1989). It is known that cocaine exacerbates catecholamine-induced ventricular fibrillation (Inoue and Zipes, 1988), but the mechanism has not been delineated. Therefore, we examined the effects of cocaine on I_K(ACh) in both atrial and left ventricular myocytes. Figure 3 demonstrates that extracellular application of 50 µM cocaine significantly decreased I_K(ACh) of atrial myocytes. Cocaine suppressed both inward and outward portions of I_K(ACh), but altered neither the current-voltage relation nor the E_Rev (n = 10). In atrial cells, cocaine at a concentration as low as 1 µM produced 25 ± 8% (fig. 4, P < .05, n = 6) inhibition of I_K(ACh); 100 µM cocaine completely blocked I_K(ACh). The concentration of cocaine necessary to produce 50% suppression of I_K(ACh) (IC₅₀) is 25 µM in atrial cells (fig. 4).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Suppressant effect of cocaine on IK(ACh) in ferret atrial myocytes with the nystatin-perforated patch-clamp method. Extracellular application of 50 µM cocaine strongly suppressed IK(ACh) activated by 1 µM carbachol at all holding potentials (, control; , cocaine) and altered neither the current-voltage relation nor the reversal potential in atrial cells (n = 10). *P < .05; **P < .01; between carbachol () and carbachol plus cocaine ().

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Dose-response curve for cocaine-induced suppression of IK(ACh) in ferret atrial cells with the nystatin-perforated patch-clamp method. The current was activated at 0 mV by extracellular application of 1 µM carbachol (as the control) or 1 µM carbachol plus various concentrations of cocaine. Mean values ± S.E. from five to eight individual cells are plotted. *P < .05; **P < .01; significantly different between the values of the control and cocaine treatment.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Comparison of cocaine-induced suppression of IK(ACh) in atrial and left ventricular myocytes with the nystatin-perforated patch-clamp method. IK(ACh) was activated by 1 µM carbachol and suppressed by 50 µM cocaine in an atrial (A) and a left ventricular (B) myocyte. Note the different current scale in A and B. The oscillation of the base line in A was caused by spontaneous contraction of the cell when it was held at 0 mV. Bars represent perfusion period of carbachol (open) and plus cocaine (solid). (C) Comparison of inhibition of IK(ACh) at 0 mV in atrial (open bars, n = 7) and left ventricular (hatched bars, n = 7) myocytes exposed to 20 and 50 µM cocaine. The normalized suppression is almost the same between two groups of myocytes.

MEG suppression of I_K(ACh). MEG is the major product of pyrolysis of cocaine base during smoking (Cone et al., 1994). Extracellular application of 20 µM MEG produced 53 ± 5% inhibition of I_K(ACh) (n = 9, P < .01, fig. 6) in ferret atrial myocytes. Figure 6A shows the original current trace of I_K(ACh) recorded from an atrial cell. The current was markedly reduced in the presence of 20 µM MEG. The inhibition of I_K(ACh) is concentration-dependent (fig. 6B). The IC₅₀ is 12 µM for MEG (solid line) and 25 µM for cocaine (dotted line), respectively. Thus, compared with cocaine, MEG had a greater effect on I_K(ACh).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Effects of MEG on IK(ACh) with the nystatin-perforated patch-clamp method. (A) The original traces of Ik(ACh) were recorded from an atrial myocyte at a holding potential of 0 mV. Bars represent a perfusion period for carbachol alone (1 µM, open bar) and carbachol (1 µM) plus MEG (20 µM, solid bar). (B) The dose-response curve for MEG-induced inhibition of IK(ACh) in ferret atrial cells. The current was activated at 0 mV by extracellular application of 1 µM carbachol (as the control) or 1 µM carbachol plus various concentrations of MEG. Mean ± S.E. from four to nine individual cells are plotted. *P < .05; **P < .01. The dotted curve in panel B represents the dose-response of cocaine-induced suppression of IK(ACh) in ferret atrial myocytes (see fig. 4).

Effects of cocaine on adenosine-activated K⁺ currents. Stimulation of A₁-purinergic receptors with adenosine can elicit the same K⁺ current as activation of muscarinic receptors in atrial cells (Kurachi et al., 1986; Kurachi, 1994; Ito et al., 1995). Figure 7 summarizes the effects of cocaine on this adenosine-activated K⁺ current (I_K(Ado)). Extracellular application of 100 µM adenosine activated I_K(Ado), which varied from 71 to 579 pA. The average density of I_K(Ado) at 0 mV holding potential was 2.12 ± 0.46 pA/pF for ferret atrial cells (n = 13). Co-application of 50 µM cocaine suppressed the mean current of I_K(Ado) to 0.90 ± 0.33 pA/pF (P < .001), which is 58 ± 5% inhibition.

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Inhibition of adenosine-activated K+ currents by cocaine. The current was activated by extracellular application of 100 µM adenosine (open bar) and 100 µM adenosine plus 50 µM cocaine (hatched bar) in ferret single atrial myocytes at 0 mV with the nystatin-perforated patch-clamp method. Values represent mean ± S.E. of 13 individual experiments. ***P < .001; vs. the control.

Effects of cocaine on GTP-activated K⁺ currents. Intracellular dialysis with GTP or GTPS activates a K⁺ current which is the same K⁺ current activated by ACh (Yatani et al., 1987; Koumi and Wasserstrom, 1994). Figure 8 depicts an example of I_K(ACh) activated by intracellular dialysis with 100 µM GTPS. The superimposed current traces in figure 8A (a) are the inwardly rectifier K⁺ current (I_K1) recorded immediately after forming the whole-cell configuration and without stimulation of muscarinic receptors. Cocaine had no significant effect on I_K1 (data not shown), which is consistent with an previous study (Kimura et al., 1992). After the initial three applications (each for 5 s) of 1 µM carbachol and 10 min after dialysis with 100 µM GTPS, I_K(ACh) was enhanced and maintained at a high level without carbachol stimulation (fig. 8A, b). The net currents activated by GTPS are shown in fig. 8A (d) which is the result of b minus a. Extracellular application of 50 µM cocaine markedly reduced I_K(ACh) activated by GTPS (fig. 8A, c). After subtraction of c from b, the net suppression of I_K(ACh) by cocaine was shown in e. Although I_K(ACh) was significantly suppressed by cocaine, the current-voltage relationship was not altered (fig. 8B). The average inhibition of GTPS-activated I_K(ACh) by 50 µM cocaine was 42 ± 4% (P < .001, n = 8). These results indicate that because cocaine inhibits GTPS-activated I_K(ACh), which bypasses the muscarinic receptor, the cocaine-induced suppression of I_K(ACh) may be mostly by block of the channel pore or inhibition of G_K.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Cocaine blockade of the K+ current in a GTPS-dialyzed atrial myocyte with the classical whole-cell recording technique. (A) Superimposed original current traces of IK1 (a) were recorded at 0.5 min after rupture of the patched membrane. After three repeated exposures (each for 5 s duration) to 1 µM carbachol IK(ACh) was recorded at 10 min after 100 µM GTPS dialysis in the absence (b) and presence of 50 µM cocaine (c). The net IK(ACh) activated by GTPS was shown in d (b  a) and inhibited by cocaine in e (b  c). The currents were activated by 250-ms pulses from a holding potential of 40 mV down to 130 mV up to 20 mV with 10 mV increments every 5 s. Note GTPS-enhanced L-type Ca++ currents in b and d, and cocaine inhibited the current in c and e. (B) Current-voltage relationships of the K+ currents measured at the 200-ms time point for the superimposed traces in a (), b () and c ().

View larger version (15K): [in this window] [in a new window]   Fig. 9.   Desensitization of IK(ACh) in GTPS-dialyzed atrial cells with the classical whole-cell recording technique. IK(ACh) was activated by extracellular perfusion of 1 µM carbachol (open bars) and 1 µM carbachol plus 50 µM cocaine (solid bars). The recordings were made with 10-s pulses at 0 mV holding potential from one atrial cell dialyzed with 5 mM ATP (A) and another atrial cell dialyzed with 5 mM ATP plus 100 µM GTPS (B). Currents of a, b, c and d were recorded at 0.5, 3, 5 and 10 min after forming the whole-cell recording configuration, respectively. Note the increase of the outward steady current after the time of intracellular dialysis with GTPS (B).

Effects of ACh and cocaine on the action potential. Boyett and co-workers (1988) reported that ACh significantly reduced the action potential duration because of activation of I_K(ACh). Because cocaine blocked I_K(ACh) in the present experiments, we further examined the effects of cocaine on carbachol-induced shortening of the action potential duration. Figure 10 illustrates that extracellular application of 1 µM carbachol profoundly shortened action potential duration (from 406 ms of the control to 138 ms, measured at 75% repolarization), increased the threshold for initiation of an action potential (from 16 pA of the control to 24 pA), and prolonged the cycle length of excitability (from 520 ms of the control to 920 ms) in a ferret atrial cell. Co-application of 40 µM cocaine partially reversed the carbachol-induced effects on evoked action potentials. The duration, threshold and cycle length of excitability of action potentials were 186 ms, 20 pA and 420 ms, respectively, when both carbachol and cocaine were perfused. After washout of carbachol and cocaine all the parameters of the action potential recovered completely. Similar results were observed from a total of five atrial and three left ventricular myocytes. These results further suggest that cocaine antagonizes the effects produced by stimulation of cardiac muscarinic receptors.

View larger version (26K): [in this window] [in a new window]   Fig. 10.   Effects of carbachol and cocaine on the duration, threshold and cycle length of the action potential recorded with the classical whole-cell patch-clamp method. (A) The atrial cell was held at approximately 80 mV by injection of a negative constant current (6 pA) and was stimulated with single 20-ms depolarization currents (28 pA) every 10 s. (B) The atrial cell was held at approximately 80 mV by injection of a negative constant current (6 pA) and was stimulated with a series of depolarization currents (50 ms) in 4-pA increments at 10-s intervals. (C) The cell was held at 80 mV by constant injection of 6 pA current and given a pair of superthreshold electrical stimuli (30 ms and 22 pA for each) at 0.1 Hz with various intervals in 100-ms increments. All the recordings were made from the same myocyte. The duration, threshold and the cycle length of recovery of the action potential is shown in the absence (Control) and presence of 1 µM carbachol (Carb), as well as 1 µM carbachol plus 40 µM cocaine (Carb + cocaine) and washout (Washout).

	Discussion

Top Abstract Introduction Methods Results Discussion References

I_K(ACh) of ferret cardiomyocytes. Some early functional studies demonstrated that both atrial and ventricular myocytes receive cholinergic innervation (Kent et al., 1974; Goldman et al., 1983; Takahashi et al., 1985). Recently, cholinergic innervation in the cardiac ventricles was confirmed in rats with the viral tracing method (Standish et al., 1994). The neurons that innervate the ventricles are numerous, and their distribution within the nucleus ambiguus and dorsal motor nucleus of the vagus is similar to that of neurons innervating other cardiac targets, such as the sinoatrial node. Our present study demonstrates that both atrial and ventricular myocytes of ferrets have muscarinic receptors. These results support the previous findings (Boyett et al., 1988) that ACh has a negative inotropic effect on ferret ventricular myocytes and add to the mounting body of evidence that mammalian ventricular myocytes from different species, including humans, contain muscarinic receptors (Loffelholz and Pappano, 1985; Litovsky and Antzelevitch,1990; McMorn et al., 1993; Koumi and Wasserstrom, 1994; Ito et al., 1995). The density of I_K(ACh) varied greatly between individual cells isolated from the same atria or from the same ventricle. This variability is probably caused by the heterogeneity of cholinergic innervation within the myocardium (Loffelholz and Pappano, 1985; Litovsky and Antzelevitch, 1990). The density of I_K(ACh) in ferret ventricular myocytes was 5 times lower than that in atrial cells, which is consistent with the finding of a lower density of I_K(ACh) in human ventricular myocytes (Koumi and Wasserstrom, 1994).

Cocaine blockade of I_K(ACh). We are the first to report that cocaine strongly suppressed the ACh-activated muscarinic K⁺ current. A concentration as low as 1 µM cocaine significantly reduced I_K(ACh) in ferret single heart cells. This concentration is much lower than the concentration, 10 µM, of cocaine required to produce significant blockade of the Na⁺ channel (Crumb and Clarkson, 1990; Xiao and Morgan, unpublished data). The concentration of cocaine to produce 50% inhibition of I_K(ACh) activated by carbachol is 25 µM in ferret atrial cells, which is only half of the concentration (50 µM) of the drug to produce 50% inhibition of I_Na in rat ventricular myocytes (Renard et al., 1994; Xiao and Morgan, unpublished data). Therefore, cocaine blockade of cardiac I_K(ACh) can be an important mechanism for its cardiotoxicity.

Significance. The present data suggest that cocaine blockade of I_K(ACh) may have clinical relevance to its cardiotoxicity. Cardiovascular morbidity and mortality from use of cocaine, particularly among young and healthy adults, have increased markedly in the past several years. Although cardiac intoxication with cocaine has been linked to a sympathomimetic effect and its local anesthetic effects of blocking cardiac Na⁺ channels, the precise mechanisms responsible for life-threatening cardiovascular events remain undetermined. ACh released on vagal stimulation produces the negative chronotropic, dromotropic and inotropic effects on the heart by activation of I_K(ACh). Activation of muscarinic K⁺ channels enhances the membrane stability of cardiac myocytes by hyperpolarization and has a protective effect against some cardiac arrhythmias (Rardon and Bailey, 1983; Rauch and Noroomand, 1991). In this study we found that stimulation of muscarinic receptors resulted in an increase in the stimulatory threshold and the refractory period, which was attenuated by application of cocaine. Wilkerson's work (1989) found that atropine significantly enhanced the cocaine-induced cardiovascular toxicity. Our present results demonstrate that cocaine has a potent blocking effect on the cardiac I_K(ACh), which may enhance cocaine cardiotoxicity.