A Subtype of α¹Adrenoceptor Mediates Depression of Conduction in Purkinje Fibers Exposed to Halothane

The experimental protocols were approved by the Animal Care Committee of the Medical College of Wisconsin. Purkinje fibers were obtained from the hearts of adult mongrel dogs killed during halothane anesthesia. The preparations were dissected from the free running intracavitary portions of the conduction system connecting the main bundle branches to the papillary muscles. Small branches were divided 2-3 mm away from the main false tendon and segments measuring 6-12 mm in length were pinned to the floor of a 2-ml tissue chamber and superfused at 5-7 ml/min (time constant approximately 20 s) with 37 degrees Celsius Tyrode's solution equilibrated with 97% Oxygen²-3% CO². The superfusate contained the following (millimolar): NaCl 137, KCl 4.0, CaCl²1.8, MgCl²0.5, NaH²PO⁴0.9, NaHCO³16, and dextrose 5.5 and 50 micro Meter Sodium ethylenediamine tetraacetic acid to limit catecholamine oxidation. All preparations were stimulated orthodromically at 150 beats/min with bipolar platinum wire electrodes and twice-threshold square-wave pulses 2 ms in duration. Action potentials were obtained from two fibers 3-8 mm apart and at least 2 mm from the stimulation site by using glass microelectrodes and intracellular amplifiers. Only drug trials in which both cell impalements were maintained for 10-20 min were accepted. Every effort was made to maintain continuous impalements throughout each sequence of experimental interventions without rezeroing the amplifiers by withdrawal into the superfusate. The action potential signals were monitored on oscilloscopes, sampled by analogue-to-digital conversion, stored, and analyzed with a computer by standard methods for this laboratory. [20]The rate of phase 0 depolarization was measured by electronic differentiation and a peak and hold detector. Conduction velocity was calculated from the time between the phase 0 upstrokes of the action potentials measured at rapid sweep with a digital oscilloscope and conduction distance measured at the end of each experiment with calibrated dividers.

The dose-response relation for the actions of epinephrine on conduction velocity of fibers exposed to halothane was determined at two anesthetic concentrations within a group of ten preparations in a balanced randomized sequence of trials. Solutions with and without L-epinephrine HCl, added as fractions of 1 ml stock solution (1 mg/ml, Adrenalin Chloride, Parke-Davis, Morris Plains, NJ) to measured volumes of superfusate, were preequilibrated with halothane from a single vaporizer. The tissues were first exposed to halothane [22]for at least 20 min to obtain equilibrium. The anesthetic concentrations in the tissue bath were sampled before and after the drug trials, measured by gas chromatography and are reported with each group. The drug trials were conducted by rapidly switching between solutions and timing the trial from entry of the solution into the tissue chamber. Our preliminary studies under these conditions [20]indicated that epinephrine transiently decreased velocity, reaching minimum values within 3-5 min and dissipating by 10-12 min, despite continuing exposure to both halothane and epinephrine. Therefore the conduction velocities and action potential characteristics were measured of 1-min intervals during 5-min trials of epinephrine exposure and at least 15 min of epinephrine-free superfusion was used for return to baseline between trials. The preparations were exposed, in randomized order, to trials of 0.2, 1, 2, and 5 micro Meter epinephrine at each of two anesthetic concentrations. These epinephrine concentrations correspond approximately to reported plasma values associated with "just-threshold" epinephrine doses in halothane (0.2 micro Meter, 40-50 ng/ml) and pentobarbital (1.5 micro Meter, about 300 ng/ml)-anesthetized dogs [23-25]and to about three times that which might occur (5 micro Meter, 1 mg/l) at equilibrium in the plasma after intravenous injection of 1 mg epinephrine in the perioperative setting.

Receptor mechanisms modulating Purkinje fiber conduction in the presence of halothane were studied in groups of 6-12 preparations. The changes in conduction velocity in response to 5 micro Meter epinephrine with halothane were first determined before and then after 30 min exposure to 1 micro Meter DL-metoprolol (Sigma, St. Louis, MO) or 1 micro Meter prazosin HCl (Pfizer, New York, NY). The responses to 5 micro Meter L-phenylephrine HCl (Sigma) and 5 micro Meter HCl (Sigma) were evaluated in the absence and presence of halothane in two groups to investigate the potentiation of alpha¹-adrenergic effects by halothane and to exclude a possible effect of alpha²-adrenergic activation. [26]The roles of the two pharmacologically distinguishable alpha¹-adrenoceptor subtypes in modulating conduction with halothane were evaluated by establishing control responses to trials of 0.2, 1.0 and 5 micro Meter epinephrine, in randomized order, in the presence of 0.2 micro Meter propranolol HCl (Sigma). One half of this group was then exposed for 30 min to 0.5 micro Meter WB4101 (Research Biochemicals, Natick, MA), and the others were exposed to 0.5 micro Meter CEC (Research Biochemicals). The epinephrine trials were then repeated in the presence of the antagonists to compare their effects on conduction depression at equimolar concentrations. Additional preparations were pretreated for 30 min with CEC (0.5 micro Meter) to attenuate responses (alpha^1B) sensitive to the alkylating agent and washed for 1 h to reduce the potential [7]for competition by CEC at the alpha^1A-receptor. These preparations were thereafter studied to determine the degree of inhibition of conduction changes by a lower "more selective" concentration (0.1 micro Meter) of the alpha^1Aantagonist WB4101 as used by others in this model. [7-8]The actions of epinephrine, angiotensin II, carbamylcholine, and endothelin 1 (Sigma) on conduction in fibers exposed to halothane were determined in two other groups of false tendons.

The findings are reported as values of the mean plus/minus SEM. All values obtained within an experimental group were evaluated by repeated-measures analysis of variance and means at specific times were compared using Waller-Duncan's least significant difference method. [27]A probability level of 0.05 or less was considered significant.

(Figure 1) illustrates the dose-response relation for the conduction velocity at the time (3 min) of maximum epinephrine effect at two halothane concentrations. Halothane alone decreased conduction velocity (P less or equal to 0.05) from a mean drug free control value of 2.21 plus/minus 0.10 m/s by 5% and 11% at the low (0.46 mM) and high (0.86 mM) concentrations, respectively. 5 micro Meter epinephrine with 0.86 mM halothane decreased velocity by 33% (to 1.31 plus/minus 0.05 m/s) of the value with halothane (1.97 plus/minus 0.08 m/s) alone. Significant depression of conduction (-5% relative to halothane alone, P less or equal to 0.01) was observed even at the lowest (0.2 micro Meter) epinephrine and halothane (0.46 mM) doses. The depression of conduction was larger (P less or equal to 0.05) at high compared to low epinephrine doses at each halothane concentration and larger (P less or equal to 0.05) at high compared to lower halothane concentration at each epinephrinedose. The depression of conduction on exposure to epinephrine was not associated with significant reductions of the rate of phase 0 depolarization or action potential amplitude (data not shown).

(Figure 2) shows the changes of conduction velocity on exposure to 5 micro Meter with 0.45 mM halothane before (top) and after (bottom) treatment with prazosin and metoprolol. The depression of conduction by epinephrine with halothane was completely abolished by 1 micro Meter prazosin but remained in the presence of 1 micro Meter metoprolol. The relative roles of alpha¹and alpha²adrenoceptors in the interaction on conduction was evaluated in two groups with the alpha-agonists phenylephrine and clonidine. As shown in Figure 3(top), phenylephrine (5 micro Meter) in the absence of halothane slightly decreased (-2%) conduction velocity at 4-5 min of exposure (P less or equal to 0.05) relative to the preceding drug free control value (time 0). Halothane alone (0.4 mM) decreased velocity (P less or equal to 0.01) to 2.06 plus/minus 0.07 m/s (at time 0) from a preceding control value of 214 plus/minus 0.08 m/s. The actions of phenylephrine were potentiated in the presence of halothane. The alpha sub 1 agonist produced a larger (P less or equal to 0.01) decrease of velocity at maximum effect with halothane (-0.25 plus/minus 0.04 m/s at 4 min of exposure, about -12% vs. halothane at time 0) than in the absence of halothane (-0.05 plus/minus 0.01 m/s at 4 min, -2% vs. time 0). The mean velocity was significantly less at 4 min of phenylephrine exposure with halothane (1.81 plus/minus 0.07 m/s) than without halothane (2.11 plus/minus 0.08 m/s) largely because of the accentuated transient effect of alpha¹activation in the presence of anesthetic. In contrast, the alpha²agonist clonidine, as shown in Figure 3(bottom), did not depress conduction velocity without or with halothane in a different group of preparations that exhibited depression of conduction by epinephrine with halothane.

(Figure 4) (top) illustrates the control conduction velocities obtained over time in 12 preparations on exposure to three doses of epinephrine with 0.7 mM halothane and 0.2 micro Meter propranolol before treatment with alpha¹-subtype antagonists. Figure 4(bottom) shows the averaged control velocities at the times of maximum epinephrine effect for each dose and those found after treatment with equimolar (0.5 micro Meter) WB4101 or CEC. The four values at 2-5 min of epinephrine exposure were averaged to increase the power of the comparison between the groups of 6 preparations treated with the two drugs. Both antagonists shifted the dose-response curve to higher epinephrine concentrations. Between antagonists, WB4101 attenuated the depression of conduction at each epinephrine dose to a greater degree (P less or equal to 0.01) than CEC, which antagonized the decreases of conduction velocity by only about 40%. Table 1shows the averaged conduction velocities found at maximum epinephrine effect with halothane in the absence and presence of 0.1 micro Meter WB4101 after CEC treatment and washing. Compared to the control group illustrated in Figure 4(n = 12), pretreatment with CEC also attenuated (P less or equal to 0.05) the response to 5 micro Meter epinephrine (Table 1, n = 10) with halothane by about 40%. In these preparations after washout of CEC, WB4101 (0.1 micro Meter) produced 87% inhibition of the remaining velocity decrease resulting from 0.2 micro Meter epinephrine and proportionately less inhibition at higher agonist concentrations in a manner consistent with competitive antagonism of an action on conduction mediated by the alpha^1A-adrenoceptor subtype.

(Figure 5) (top) illustrates the conduction responses to epinephrine, angiotensin and carbamylcholine with halothane in one group of preparations. Both epinephrine and the muscarinic receptor agonist depressed conduction velocity with halothane, whereas angiotensin did not affect conduction. Figure 5(bottom) shows the responses to epinephrine and endothelin in another group of preparations. Endothelin with halothane depressed conduction with a slower onset than epinephrine with halothane, and there was a marked delay in washout (data not shown) of the endothelin effect such that return to the value with halothane alone required 30-40 min.

The results of this study indicate that epinephrine, at submicromolar concentrations (0.2 micro Meter) comparable to just-threshold dysrhythmogenic plasma epinephrine concentrations with halothane in vivo, [23-25]transiently slows conduction in canine Purkinje fibers exposed to halothane. These actions are mediated largely by the WB4101-sensitive subtype of alpha¹adrenoceptor, which has been reported to lead to phosphoinositide hydrolysis by PLC in cardiac tissues [7]and similar depression of conduction was found on activation of other G-protein-linked receptors (endothelin and muscarinic) known to stimulate PLC. [21]The results demonstrate a direct prodysrhythmic alpha¹-mediated interaction between catecholamines and halothane and support the hypothesis [19]that the mechanism generating halothane-epinephrine dysrhythmias may involve abnormal conduction leading to reentry.

Reynolds and Chiz [19]originally reported that epinephrine markedly potentiated the conduction slowing by halothane. We previously found that 5 micro Meter epinephrine alone produces no or minimal decreases of conduction velocity, but significantly greater transient depression in Purkinje fibers exposed to halothane. [20,22]In the present study we observed marked potentiation by halothane of a small transient decrease of velocity produced by 5 micro Meter phenylephrine alone (Figure 3) and suspect that larger pharmacologic concentrations of alpha¹agonists transiently depress conduction in the absence of potentiating anesthetic agent. The actions of epinephrine with halothane on conduction (Figure 1) varied from slight transient depression (-5%) with just-threshold concentrations (0.2 micro Meter) and 0.46 mM (1.5 vol%) halothane to substantial depression (-33%) at high epinephrine (5 micro Meter) and halothane (0.86 mM, 2.8%) doses. The interaction was more dependent on the halothane than epinephrine concentrations because doubling the halothane concentration produced approximately the same transient depression of velocity as a five- to tenfold increase of epinephrine dose. The relation between the degree of conduction delay observed in isolated Purkinje fiber models and the more complicated phenomena of unidirectional block and slow conduction leading to ventricular reentry in situ is not known. The latter may involve propagation of premature or postmature responses at lower membrane potentials and intrinsic spatial inhomogeneities in refractory periods, membrane excitability and fiber geometry not present in isolated false tendons. [28]However the degree of transient conduction impairment (-10 to -20%) produced by epinephrine with halothane in this study is similar in magnitude to that in Purkinje fibers cross-superfused with blood from dogs exhibiting intraventricular conduction delay and wide QRS complexes as a result of administration of toxic doses of quinidine. [29]Proof that the observed slowing of conduction by epinephrine with halothane is sufficient to facilitate reentry would require mapping activation in the reentrant circuit during onset of the dysrhythmia, localization of the site of unidirectional block and demonstration of sufficient conduction delay to permit reexcitation proximal to the site of block on recovery of excitability.

The findings that the depression of conduction by epinephrine with halothane is not altered by beta¹-adrenergic blockade, that it is antagonized by prazosin and reproduced by the alpha¹agonist phenylephrine but not by the alpha²agonist clonidine, clearly indicate that this response is mediated by alpha¹adrenoceptors. Two pharmacologic subtypes of alpha¹adrenoceptors, designated alpha^1Aand alpha^1B, were described by Han et al. [30]and Minneman [31]based in part on receptor binding studies and differential inhibition of alpha¹-mediated functional effects by WB4101 and CEC in various tissues. A total of four alpha¹-adrenoceptor subtypes have been identified in molecular cloning studies, [32,33]one of which (alpha^1C), when expressed in an artificial cell line, also stimulates phosphoinositol hydrolysis but cannot be distinguished by these two antagonists. [32]However the relation between the pharmacologically distinct receptors and the cloned alpha¹subtypes is not yet clear and they have not been shown to mediate any specific effect on cardiac tissues. [9].

Del Balzo et al. [7]have shown that WB4101-sensitive alpha¹adrenoceptors mediate an increase of automaticity in canine Purkinje fibers by a mechanism involving linkage by a pertussis toxin-insensitive G-protein to stimulation of PLC. PLC applied extracellularly also enhances Purkinje automaticity. [11]The increased automaticity and action potential prolongation [8]resulting from alpha¹-adrenoceptor activation in Purkinje fibers were antagonized by WB4101 at a concentration (0.1 micro Meter) that Del Balzo et al. [7]reported to effectively (> 95%) inhibit norepinephrine stimulation of inositol trisphosphate (IP³) production by PLC in isolated rat myocytes. The action potential changes probably result from inhibition of outward Potassium sup + currents [8,10]and have an important influence on abnormal automaticity in ischemic fibers. [34]On the other hand, CEC-sensitive alpha¹adrenoceptors were reported to decrease Purkinje fiber automaticity by a mechanism sensitive to pertussis toxin, [7]which leads to activation of the Sodium sup + -Potassium sup + electrogenic pump. [12]Del Balzo et al. [7]reported that WB4101, but not CEC, antagonized norepinephrine stimulated IP³generation in isolated rat myocytes and suggested that only the alpha^1A-receptor subtype, linked by a pertussis toxin-insensitive G-protein, stimulated PLC. However, recent contradictory studies in fresh adult rat myocytes indicate that both the alpha^1Aand alpha^1Bsubtypes lead to stimulation of PLC [18]and that CEC (100 micro Meter) typically inhibits epinephrine stimulated IP³generation by about 33%. Our findings of greater antagonism of conduction slowing by equimolar (0.5 micro Meter) WB4101 than CEC demonstrate that the reduced responses in the presence of the subtype antagonists are not caused by deterioration of the preparation. The 40% inhibition of velocity decreases resulting from epinephrine with halothane in the two groups pretreated and washed (Table 1) or studied in the continued presence of CEC (Figure 4) are consistent with antagonism of a moderate degree of conduction slowing mediated by the alpha^1Bsubtype. On the other hand, the substantial (87%) inhibition of velocity changes by a low (0.1 micro Meter) concentration of WB4101 after pretreatment with CEC suggest that the conduction slowing is probably largely mediated by the same WB4101-sensitive alpha sub 1A -adrenoceptor subtype as that reported to increase automaticity and action potential duration in Purkinje fibers. [7,8]These findings are limited in that we can not exclude actions at the other cloned subtypes of alpha¹adrenoceptors, which may or may not exist as distinct subtypes in the heart. [18,35]Further studies of the relative densities of alpha¹-receptor subtypes in canine Purkinje fibers, the pertussis toxin sensitivity of their coupling G-proteins and their relation to specific isoforms of PLC [36]will be required to better define the alpha¹adrenoceptor-effector mechanisms modulating conduction in this tissue.

To evaluate separately the potential role of activation of PLC in modulation of conduction in Purkinje fibers exposed to halothane, we investigated the effects of stimulation of other G-protein-linked receptors that produce phosphoinositide hydrolysis in cardiac tissues. Hilal-Dandan et al. [21]compared the effects of several agonists on production of IP³in isolated adult rat myocytes and reported that endothelin 1 was most efficacious, followed by activation of alpha¹and muscarinic receptors, whereas angiotensin II was the least effective. Our findings that endothelin and carbamylcholine, but not angiotensin, also transiently decrease conduction velocity in Purkinje fibers exposed to halothane imply that the alpha¹-adrenoceptor mechanism mediating depression of conduction by epinephrine probably involves stimulation of PLC regardless of which alpha¹-adrenoceptor subtype may mediate this action. The findings suggest a previously unsuspected potential role of hormonal modulation of conduction in Purkinje fibers related to receptor-mediated stimulation of PLC and as yet unknown actions of the intracellular second messengers IP³and diacylglycerol resulting from hydrolysis of sarcolemmal phosphoinositides.

The electrophysiologic mechanism underlying alpha¹adrenoceptor-mediated depression of conduction in Purkinje fibers exposed to halothane is not known. Although action potentials were recorded simultaneously with conduction times, the requirement to maintain impalements in two fibers over long periods of time precludes accuracy in measurements of maximum diastolic potential. In addition, we used a rapid pacing rate (150 beats/min) similar to that at the onset of halothane-epinephrine dysrhythmias in vivo, which tends to minimize drug effects on action potential duration and phase 4 diastolic depolarization. Qualitatively we did not observe any decrease of membrane potential just before the action potential upstroke (the "take-off" potential), as might occur with marked delay of repolarization or spontaneous diastolic depolarization (automaticity), that might explain the velocity decreases during the trials with epinephrine, carbamylcholine or endothelin. As previously reported with epinephrine, [20]the transient decreases of conduction velocity in this study were not accompanied by simultaneous decreases (data not shown) in action potential amplitude or rate of phase 0 depolarization, major determinants of conduction velocity. This relative lack of change in rate of phase 0 depolarization associated with marked depression of conduction by epinephrine with halothane suggests that the mechanism modulating conduction probably involves an action on cell-to-cell coupling, [20,37]rather than or in addition to possible reduction of the peak inward Sodium sup + current. Both high concentrations (2-3%) of halothane [38]and phenylephrine (10 micro Meter) have been reported to uncouple cardiac cells, although the latter preliminary report [26]has not been confirmed. The finding that the conduction changes were not sensitive to beta¹-adrenergic blockade would appear to exclude mechanisms of conduction slowing related to cyclic adenosine monophosphate-dependent depression of Sodium sup + [39,40]channel current or increased Calcium²+ influx through L-type Calcium²+ channels. However, this finding does not exclude potential mechanisms of cellular uncoupling resulting from increases in intracellular Calcium sup 2+ [41]caused by enhanced Calcium²+ release from the sarcoplasmic reticulum or effects on Sodium sup + -Calcium²+ exchange. [9,42,43]Finally, there is increasing evidence [44,45]that the conductance of cardiac gap junctions is related to the phosphorylation state of the connexin proteins forming the channels and may be physiologically modulated by cyclic adenosine monophosphate-dependent protein kinases and protein kinase C. An interaction between halothane and alpha¹adrenoceptor-mediated effects on processes regulating junctional resistance between Purkinje fibers could slow conduction by increasing the current required for propagation or altering threshold without substantial changes of membrane potential [37]as observed in this model.

In conclusion, our studies demonstrate a dose-related negative dromotropic interaction between epinephrine and halothane that transiently slows conduction in canine Purkinje fibers at epinephrine concentrations similar to dysrhythmogenic plasma concentrations of epinephrine in halothane-anesthetized dogs in vivo. The depression of conduction is largely but not exclusively mediated by the WB4101-sensitive alpha^1A-adrenoceptor subtype reportedly coupled to activation of PLC in this model and may also be produced by activation of other hormone receptors (endothelin and muscarinic) known to activate PLC. The modulation of Purkinje fiber conduction velocity appears to involve an adverse potentiation by anesthetics of an alpha sub 1 adrenoceptor-mediated effect on cell-to-cell coupling. The results may explain progressive increase of the epinephrine dysrhythmia threshold dose by alpha¹-antagonists in vivo [46]and support the hypothesis that the mechanism underlying generation of ventricular dysrhythmias by epinephrine during halothane anesthesia may involve abnormal conduction and reentry.