Introduction

Top Abstract Introduction Methods Results Discussion References

The alpha-1 receptor subtype blockers WB4101 (alpha-1A and alpha-1D) and CEC (alpha-1B) are frequently used to distinguish physiological effects of alpha-1 receptor stimulation (for review, see Minneman, 1988; Terzic et al., 1992). In normal cardiac Purkinje fibers stimulation of a WB4101-sensitive alpha-1 receptor subtype increases ventricular automaticity and prolongs repolarization (Lee et al., 1991). In contrast, stimulation of a chloroethylclonidine-sensitive alpha-1 subtype activates the Na/K pump via signal transduction dependent on a PTX-sensitive G protein (Shah et al., 1988). This pathway decreases intracellular Na activity, slows automatic rate, and accelerates repolarization (Zaza et al., 1990). The balance between these two subtype actions determines the net electrophysiologic response to alpha-1 adrenergic stimulation.

Although little has been done to explore alpha-1 receptor-vagal interactions, some potential importance of these is implied by the vagal and beta adrenergic interactions on the heart. Vagal stimulation has negative chronotropic, dromotropic and inotropic effects, respectively, at sinoatrial, AV nodal and myocardial sites (Loeffelholz and Pappano, 1985). The effects of acetylcholine at atrial and AV nodal levels can occur in the presence or absence of sympathetic innervation and/or beta adrenergic catecholamines (Carrier and Bishop, 1972). Particularly important to the effects of acetylcholine on supraventricular tissues and, perhaps, essential to its actions on ventricle is its antagonism of the receptor-effector coupling pathway of beta adrenergic catecholamines (Levy et al., 1972). This "accentuated antagonism" of beta adrenergic actions is based on acetylcholine's effect to reduce adenylate cyclase activation via a G_i-transduced pathway. This reduces the action of beta adrenergic receptor stimulation to initiate effector responses transduced via the GTP regulatory protein G_s and adenylate cyclase activation (Robishaw and Foster, 1989).

That there is a basis for expecting vagal alpha-1 adrenergic interactions is suggested by the work of Wendt and Martins (1990) in intact canine hearts beta blocked with metoprolol. Phenylephrine increased the Purkinje fiber relative refractory period and vagal stimulation enhanced the increase. No such changes were observed in ventricular endocardial muscle refractory periods. Vagal alpha-1 adrenergic interactions also have been demonstrated in isolated rat atria, in which there is alpha-1 adrenergic inhibition of vagal ACh release (Wetzel et al., 1985). Such inhibition of the vagus would facilitate increases in heart rate during sympathetic stimulation.

Based on this background, our study was designed to assess the effects of alpha-1 receptor subtype stimulation on sinoatrial rate and ventricular repolarization of vagally innervated, isolated guinea pig hearts. Using this model, we studied 1) the effects of phenylephrine perfusion and of vagal stimulation individually on sinoatrial rate and ventricular repolarization and 2) the modulation of alpha-1 adrenergic effects WB4101 and CEC.

	Methods

Top Abstract Introduction Methods Results Discussion References

Male guinea pigs (weight 250-300 g) were anesthetized with i.p. sodium pentobarbital (30 mg/kg). The heart was rapidly excised and via careful dissection the right vagus nerve was maintained intact from the thoracic inlet to the heart. We previously used this technique effectively to stimulate the vagi of Langendorff-perfused hearts in small animals (Shvilkin et al., 1994). The caveat concerning this procedure is that results are referable to the effects of right vagal stimulation only.

The ascending aorta was cannulated and each heart was retrogradedly perfused with Krebs-Henseleit solution containing (in mM:); NaCl, 100; KCl, 2.8; NaEDTA, 0.5; KH₂PO₄, 1.2; CaCl₂, 3.0; glucose, 11.0; HEPES acid, 25; HEPES Na salt, 25, achieving a pH = 7.4. Temperature was kept at 37°C by a glass heat exchanger and the perfusates were bubbled with 100% O₂. The perfusion pressure was fixed at 80 mmHg by adjusting the height of the perfusion bath. Coronary flow was calculated by timed collections of the effluent fluid. All perfusates contained propranolol, 5 × 10⁷ M to induce beta adrenergic blockade, and to format a focus on alpha adrenergic actions.

To record cardiac electrical activity, pairs of silver wires with Teflon coating to the tip were applied to the epicardial surface of the right atrium and the right ventricle. Bipolar electrograms were displayed on a Gould chart recorder at a speed of 100 mm/sec. The QT_c was calculated according to Bazett's formula (Bazett, 1920). Although controversy exists concerning the applicability of Bazett's formula, especially at rapid heart rates where there is a loss of linearity, it is considered to reasonably approximate the QT interval (Ahnve, 1985). We have previously used the formula in studies of small animals (Malfatto et al., 1990).

The hearts initially were allowed to stabilize for 45 min. Those that showed more than 10% variability in sinus rate over the last 20 min were discarded. The right vagus was suspended over platinum-iridium bipolar electrodes connected to an isolation unit driven by a Grass stimulator. Stimulation was accomplished using square wave pulses of 2 msec duration and 10 mA intensity. Stimuli were delivered for 10 sec. In addition, in some protocols, right ventricular pacing was performed using bipolar electrodes to deliver a stimulus strength of twice diastolic threshold and 2 msec duration.

Dose-response curves for phenylephrine and for the effect of vagal stimulation on heart rate and repolarization were obtained by administering cumulative concentrations of phenylephrine from 10⁸ to 10⁶ M and by performing vagal stimulation at 1, 5, 10 and 20 Hz. Other groups of experiments involved the use of WB4101 (10⁷ M) and/or CEC (10⁷ M). Data were stored continuously and measured 20 min after each change in drug concentration by which time a steady-state effect had been achieved. In vagal stimulation experiments, measurements of vagal effect were made at the same 20-min time point.

Statistical analysis. Data were analyzed from preparations having coronary flow >8 ml/min and manifesting at least a 40% decrease in sinus rate on maximal vagal stimulation. Changes in heart rates of <10% were considered to represent normal variability. Heart rate, QT and QT_c were compared at each concentration of agonist with and without antagonist. Statistical analysis was done using one-way analysis of variance with Bonferroni's test when the f value so indicated (Snedecor and Cochran, 1967). P < .05 was considered significant. Analysis of the frequency of events was performed using Fisher's Exact Test. Data are expressed as mean ± S.E.M.

Materials. Phenylephrine and propranolol were purchased from Sigma Chemical Co. (St. Louis, MO) and CEC and WB4101 from Research Biochemicals Inc. (Natick, MA); WB4101 was diluted in 95% ethanol and CEC, phenylephrine and propranolol in distilled water (volume = 100 µl).

	Results

Top Abstract Introduction Methods Results Discussion References

Effects of phenylephrine on sinus rate. As we have demonstrated in earlier studies of ventricular specialized fibers (Rosen et al., 1977; Danilo, 1985), there were two populations of sinus node response to  agonist. Of 13 preparations studied, 8 showed a phenylephrine-induced decrease in sinus rate, and 5 showed no change (fig. 1). When phenylephrine was superfused in the presence of WB4101 a uniform response of sinus rate was seen: all eight preparations manifested decreased automaticity. In contrast, the phenylephrine-induced decrease in sinus rate was attenuated by chloroethylclonidine: of eight preparations, six showed no change in automaticity and two still showed a decrease (P < .05 cf WB: Fisher's exact test). Finally, in the presence of both blockers, a response not unlike control was seen, although the magnitude of the decrease in sinus rate was reduced (data not shown).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Effects of phenylephrine, alone (n = 13) and in the presence of the alpha-1 adrenergic subtype blockers, CEC (n = 8) or WB4101 (n = 8), both 107 M. Phenylephrine significantly reduced sinus rate in eight preparations () and had no effect on five (). In the presence of WB4101 eight preparations showed decreased automaticity (). In the presence of CEC six preparations () showed no change in automaticity and two (data not shown) still decreased. *P < .05 cf control.

6

Effects of phenylephrine during vagal stimulation. Vagal stimulation at 1 to 20 Hz reduced sinus rate in a frequency-dependent fashion (fig. 2A). The reduction in pacemaker responsiveness to vagal stimulation was enhanced by phenylephrine, with this enhancement best seen at 5 Hz (fig. 2A). To further explore the interaction between  agonist and vagal effect on sinus rate another set of experiments was performed in which vagal stimulation at 20 Hz was delivered and phenylephrine 10⁸ to 10⁶ M was then perfused (fig. 2B). Seen more clearly here than with the protocol in figure 2A is the effect of phenylephrine to concentration-dependently reduce sinus rate over the effect seen with vagal stimulation alone. The response persisted in the presence of WB4101 and was attenuated by CEC.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   A, Effects of vagal stimulation at 1 to 20 Hz alone and in the presence of phenylephrine, 108 to 106 M on sinus rate. At 5 Hz all preparations showed decreased automaticity (*P < .05 cf control). At 5 Hz, only the decrease in the presence of phenylephrine > that with vagal stimulation, alone (P < .05). B, Effects of phenylephrine, alone (n = 6) and in the presence of CEC (n = 6) or WB4101 (n = 7) both 107 M, on sinus node responsiveness to vagal stimulation at 20 Hz. Neither CEC nor WB4101 alone significantly altered sinus rate. Vagal stimulation (VS) decreased sinus rate from a control value of approximately 205 b/min (C) to the values indicated as (C + VS). Phenylephrine further reduced the sinus rate beyond that seen with vagal stimulation alone. This decrease was comparable in the presence of WB4101, and was attenuated by CEC. +P < .05 cf control. *P < .05 cf C + VS.

Effects of phenylephrine on ventricular repolarization. In these experiments, two protocols were used. In the first, hearts were in sinus rhythm; in the second, they were paced from the right ventricle to maintain a constant rate 10% > sinus rate. Results of the first protocol are presented in figure 3. As demonstrated in A, in the absence of vagal stimulation, phenylephrine shortened the QT_c interval concentration dependently. The response persisted in the presence of chloroethylclonidine and was blocked by WB4101. As shown in B, vagal stimulation (20 Hz) significantly decreased the QT_c interval. Here, addition of phenylephrine had no further effect, either alone or in the presence of either antagonist.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   A, Effects of phenylephrine, alone (n = 5) and in the presence of CEC (n = 6) or WB4101 (n = 6), both 107 M on the QTc interval. Phenylephrine alone significantly reduced QTc. This effect was also seen in the presence of CEC and was blocked by WB4101. *P < .05 cf control. B, Effects of phenylephrine, alone (n = 6) and in the presence of CEC (n = 6) or WB4101 (n = 7), both 107 M on QTc during vagal stimulation. Vagal stimulation (C + VS) shortened QTc in all three groups (P < .05). Addition of phenylephrine, alone, or in the presence of CEC or WB4101 had no further effect.

7

5

5

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effects of phenylephrine alone and in the presence of CEC or WB4101 on the QT interval during vagal stimulation of hearts paced from the ventricle. Phenylephrine significantly reduced the QT, an action blocked by WB4101, but not CEC. *indicates P < .05 cf control.

	Discussion

Top Abstract Introduction Methods Results Discussion References

There are conflicting data on alpha-1 adrenergic modulation of sinus rate. A negative chronotropic effect induced by methoxamine and phenylephrine has been reported in rabbit sinoatrial node cells (Satoh and Hashimoto, 1988), whereas data from isolated rat right atria indicate that phenylephrine induces a positive chronotropic effect antagonized by WB4101, but not CEC (Williamson et al., 1994). In other studies, direct sinus node artery infusion of phenylephrine at high concentrations accelerated canine sinus rate (James et al., 1968) whereas intravenous infusion of epinephrine to propranolol-treated dogs had no effect on sinus rate (Hordof et al., 1982). In part these studies probably reflect species differences in effects of  agonist on sinus rate, as well as differences in experimental design.

In contrast, prior studies of the ventricular specialized conducting system in dog and rat (Danilo, 1985; Drugge et al., 1985; Rosen et al., 1977) demonstrate two different effects of  agonist. Some preparations show a decrease in rate, others an increase, with the former being blocked by CEC and the latter by WB4101. Studies of the receptor-effector pathways have suggested the one blocked by CEC involves binding of agonist to an alpha-1B receptor subtype and results in stimulation of the Na/K pump via a pertussis toxin (PTX) sensitive G protein (Rosen et al., 1988; Shah et al., 1988; Zaza et al., 1990). In contrast, the WB4101-blocked pathway involves agonist binding to an alpha-1A or alpha-1D subtype and is associated with activation of a phosphatidylinositol-based second messenger system (del Balzo et al., 1990). Results of our study are concordant with these earlier findings: i.e., in guinea pig sinus node, phenylephrine induces two populations of response; either a decrease in automaticity or no change. That the two groups are justifiably considered as separate is based not only on extensive previous experience in the canine heart (Danilo, 1985) but on the response to antagonist: i.e., in the presence of WB4101, 100% of hearts showed decreased automaticity in response to phenylephrine, implying blockade of a receptor subtype that antagonizes the increase in rate, and in the presence of WB4101, there was a significant reduction of the negative chronotropic effect, implying blockade of a receptor pathway that, in fact, decreases rate. Finally, when both blockers were used a distribution of responses (increased or decreased rate) comparable to that with phenylephrine, alone, was seen, albeit with a different magnitude of response, suggesting occupancy of both receptor subtypes.

The interactions of alpha adrenergic and parasympathetic effects are complex. Studies of isolated rat hearts have shown that parasympathetic actions can be regulated through an alpha-1 adrenergic receptor having different effects at pre- and postganglionic levels (Pardini et al., 1991). An alpha-1 mediated facilitation of acetylcholine release in the perfused rat heart has also been reported (Bognar et al., 1990) and, in contrast, alpha-1 mediated inhibition of acetylcholine release has been reported in isolated rat atria (Wetzel et al., 1985). In contrast, no preganglionic alpha adrenergic effect on vagally induced bradycardia has been described in guinea pig hearts (Lew and Angus, 1983).

Postjunctional M₂ muscarinic receptors are linked via a PTX-sensitive G protein to the pacemaker current I_f, which is decreased by acetylcholine (DiFrancesco et al., 1989) and to I_K,ACh which is increased by acetylcholine (Coumi and Wasserstrom, 1994). Both actions decrease automaticity and both are complemented by the G_i-transduced action to oppose beta adrenergic agonist effects on the adenylate cyclase-cAMP second messenger pathway. The alpha adrenergic action to decrease automaticity is the result of a pathway transduced by a member of the same PTX sensitive family of G proteins as G_i (Rosen et al., 1988), but one that stimulates the Na/K pump (Shah et al., 1988; Zaza et al., 1990). This would generate a net outward current counteracting the pacemaker current. The pathway is complementary to rather than redundant with the muscarinic.

In our studies of repolarization, phenylephrine, alone, shortened the QT and the QT_c intervals, actions blocked by WB4101 but not by CEC. There are two important aspects to this observation: first, that both the QT and the QT_c were decreased by phenylephrine is indicative of a direct alpha-1 adrenergic action on the myocardium. In contrast, that the QT_c was decreased as a result of vagal stimulation whereas the absolute QT interval during ventricular pacing was not, suggests that the vagal effect on ventricular repolarization in these experiments was not direct; i.e., had there been a direct effect of right vagal stimulation on ventricular repolarization, then the paced Q-T would have shortened as well. That only the QT_c decreased on vagal stimulation indicates that the slowing of sinus rate, which provides the denominator in the Bazett formula, is responsible for the decrease in the interval measured.

The effect of phenylephrine to shorten repolarization is based on an alpha adrenergic receptor subtype different from that involved in the slowing of sinus rate (blocked by CEC, not WB4101). Previous work on canine Purkinje fiber (Lee et al., 1991) and rat ventricle (Apkon and Nerbonne, 1988) has shown an alpha-1 adrenergic receptor subtype of the same pharmacological profile (blocked by WB4101, not CEC) determines the effect of alpha adrenergic agonist on repolarization. A major difference, however, is that in dog and rat heart, alpha adrenergic agonists prolong repolarization. This prolongation is accompanied by block of I_to1 and/or I_K (Apkon and Nerbonne, 1988). The shortening of repolarization by  agonist in guinea pig ventricle, as recorded on ECG, is associated with a decrease in duration of the guinea pig ventricle action potentialresults opposite to those in dog and rat. The effect is exerted via a PKC-dependent pathway to increase I_K (Dirksen and Sheu, 1990). Hence, the effects of alpha adrenergic agonist on repolarization are clearly species dependent, involving the same alpha-1 adrenergic subtype in guinea pig, rat and dog, but different effector-coupling pathways.

That all experiments were performed in the presence of propranolol increases the likelihood that all effects were alpha adrenergic. However, it leaves unanswered the question of to what extent alpha adrenergic action and vagal interactions are expressed in the absence of beta adrenergic blockade. Wendt and Martins (1990) demonstrated that in the setting of beta adrenergic blockade, alpha adrenergic agonist increases the canine Purkinje fiber, but not the ventricular muscle effective refractory period, and the effect on Purkinje fiber is increased by acetylcholine. Kolman et al. (1976) studied vagal effects on ventricular excitability in a control setting, during left stellate stimulation and during beta adrenergic blockade. Vagal stimulation shifted ventricular strength interval curves later into diastole, signifying a prolongation of refractoriness. This effect was blocked by propranolol. Left stellate stimulation shifted the strength interval curve earlier into diastole and this action overshadowed the vagal effect when combined vagosympathetic stimulation was used. The Kolman study (1976) stresses the importance of vagosympathetic interactions and the dominant beta adrenergic component of sympathetic effect. However, it does not report any alpha adrenergic components. Moreover, it is difficult to clearly relate the results of both of these studies to our own as we measured QT and QT_c intervals, whereas the earlier studies considered relative refractory periods and strength interval curves. Although we can assume a relationship between the duration of the QT interval and that of the refractory periods, the QT interval depends entirely on the duration of inward and outward currents during repolarization whereas refractory periods bring in the additional variable of the recovery of excitability of the fast inward Na current.

In summary, in guinea pig heart, alpha adrenergic agonist, alone, slows sinus rate and its action is concordant with and additive to that of right vagal stimulation rather than resulting in accentuated antagonism. That the receptor subtype involved is alpha-1B, is suggested by the blocking effect of CEC. Consistent with observations in disaggregated guinea pig ventricular myocytes, alpha adrenergic agonist accelerates ventricular repolarization. The alpha adrenergic subtype here may be the alpha-1A or alpha-1D, as the response is blocked by WB4101 and not CEC. Although right vagal stimulation shortens the QT_c this does not reflect a direct action on ventricular repolarization but rather the slowing of sinus rate. Finally, the clinical implications of this work are most readily seen in the settings of beta adrenergic blocker therapy. It is likely that in this milieu both alpha adrenergic action in their own right and interactions with the vagus will be most marked.