- Email: [email protected]
Effects of α-adrenergic agents on transient inward current in rabbit Purkinje fibers
J Mol
Cell
Cardio122,
191-200
Effects
( 1990)
of a-Adrenergic Agents on Transient Current in Rabbit Purkinje Fibers Gregory
R. Ferrier*
and Edward
Inward
Carmeliet
Laboratory of Physiology, University of Leuven, Campus Gasthuisberg, B-3000 Leuven, Belgium (Received 5 January 1989, accepted in revisedform 11 October 1989) G. FERRIER AND E. CARMELIET. Effects ofa-Adrenergic Agents on Transient Inward Current in Rabbit Purkinje Fibers. Journal ofMolecular and Cellular Cardiolo~ (1990) 22, 191-200. Reported effects of a-adrenergic agents on
oscillatory afterpotentials (OAP) are conflicting. Therefore, we used standard two microelectrode voltage clamp techniques to determine the effects of phenylephrine and prazosin on the transient inward current (Zri) responsible for OAP. The Zti was induced in isolated rabbit Purkinje fibers either by acetylstrophanthidin or 8 rnr,r Ca. The magnitude of the Zti was determined at various membrane potentials after activation by steps to - 10 mV from a holding potential of -80 mV. When the Zti was induced by acetylstrophanthidin, phenylephrine (10e7 to 10e5 M) caused inhibition of Zti at all potentials tested. Phenylephrine also caused a significant decrease in net outward current at plateau voltages. Both effects were blocked by prazosin (5 x IO-’ M) but not by propranolol (5 x lo-’ M). Prazosin also strongly inhibited the Z,t in the absence ofphenylephrine. At 5 x 10v7 M, prazosin did not affect sodium current activated by voltage steps from -80 to -45 mV or maximum upstroke velocity during interruptions of the voltage clamp. When the Ztt was induced by 8 rnM Ca, the effect of phenylephrine, but not prazosin, reversed so that phenylephrine increased the amplitude of the Ztt. Thus, a-adrenergic agonists may exert either inhibitory or stimulatory effects on the Ztt depending on the mechanism by which the current is induced. Additional effects on OAP amplitude may be induced by changes in action potential duration mediated through actions on net outward current. Prazosin may suppress OAP by an action on the Ztt which is independent of a-adrenergic or local anaesthetic actions. KEY
a-Adrenergic agonists; fiber; Afterpotentials.
WORDS:
Purkinje
a-Adrenergic
antagonists;
Introduction Several studies suggest that a-adrenergic agonists promote cardiac arrhythmias in various settings including reperfusion of previously ischemic myocardium (Sheridan et al., 1980; Penny et al., 1985). In studies utilizing isolated tissues, Ferrier et al. (1985) demonstrated that exposure of Purkinje tissue to ischemic conditions followed by return to nonischemic conditions induced a series of potential arrhythmic mechanisms, including oscillatory afterpotentials (OAP, also referred to as delayed afterdepolarizations). The possibility that OAP constitute an important mechanism of reperfusion arrhythmias has also been suggested by Pogwizd and Corr (1987) who used computer-assisted mapping to document the *Please address all correspondence Medical Building, Dalhousie University, 0022-2828/90/020191
+ 10 $03.0010
Cardiac
arrhythmias;
Transient
inward
current;
initiation of reperfirsion arrhythmias in open chest cats subjected to coronary artery occlusion and reflow. They observed that 75% of nonsustained ventricular tachycardias induced by reperfusion in cats, were initiated by a non-reentrant mechanism in the subendocardium. They suggested that their observations were compatible with OAP as the mechanism responsible for initiation of these arrhythmias, and further suggested that Madrenergic activation might be responsible for promotion of OAP during reperfusion. Direct evidence for potentiation of OAP by cl-adrenergic agonists has not been consistent. Kimura et al. (1984,1987) showed that phenylephrine or methoxamine would promote the development of OAP in response to 8 mM Ca
to: Gregory R. Ferrier, Department of Pharmacology, Halifax, Nova Scotia, Canada B3H 4H7.
Sir Charles
0 1990 Academic
Tupper
Press Limited
192
G. R. Ferrier
in isolated cat Purkinje fibers. On the other hand, Hewett and Rosen (1984) found no effect of these same agonists on OAP induced by ouabain in canine Purkinje fibers. These inconsistent observations might be explained if a-agonists have multiple actions with opposing effects on OAP so that in different settings either no effect or exacerbation of OAPinduced arrhythmias is seen. Indeed, it is well known that a-adrenergic agonists prolong action potential duration (Giotti et al., 1973). This effect would be expected to promote the appearance and increase the amplitude of OAP (Kass et al., 1978a). In contrast, the effects of a-agonistp directly on the transient inward current (Iti) are unknown. Therefore, the present study was conducted under voltage clamp conditions in order to determine the effects of a-adrenergic agonists on the Iti under conditions in which there can be no influences from changes in action potential duration. This information is necessary to clarify the basis for the conflicting observations from various laboratories concerning the effects of a-adrenergic agents on OAP.
and
E. Carmeliet
ing steps was 3 per min in most experiments. Recordings Were made on a Tektronix 5lllA storage oscilloscope and a Gould Brush (2400) rectilinear pen recorder. All drugs except propranolol were prepared freshly as aqueous stock solutions ( lop3 or 10e4 M). Aliquots of stock solutions were diluted directly in the Tyrode’s superfusate to achieve final concentrations. Phenylephrine, and prazosin were obtained from Sigma. Propranolol (Inderal) was obtained as injectable solution from ICI, and phentolamine (regitine), from CIBA. All except some preliminary experiments were conducted in the presence of 5 X lo-’ M propranolol to eliminate p-adrenergic effects. Results Effects of phmylephrine on Iti induced by acetylstrophanthidin The effect of phenylephrine on the magnitude of 1,i induced by acetylstrophanthidin in a representative experiment is illustrated in Fi-
Methods The effects of phenylephrine were studied on Iti induced either by acetylstrophanthidin (AS) or elevated calcium concentration (8 mM). All studies were conducted on rabbit Purkinje fibers to eliminate differences due to species. Rabbits were sacrificed by cervical dislocation. Short segments (0.8 to 1 .O mm) of Purkinje fibers dissected from both ventricles were obtained by crushing the isolated fibers with a wire grid (Aronson et al., 1973). The composition of modified Tyrode’s solutions used for dissection and superfusion have been reported by Mubagwa and Carmeliet (1983). Experiments were performed at 37°C. The two microelectrode voltage clamp technique was used. Bevelled microelectrodes filled with 2.0 M KC1 and having resistances between 5 and 12 MLI were used for both current passage and voltage recording. Voltage steps were generated by a digital timing circuit which controlled the command voltage of the voltage clamp feed-back circuit. The Zti was elicited by 2- or 3-s long depolarizing voltage steps usually from a holding potential of -80 mV. The repetition rate of depolariz-
2s Control
PE (IO+M)
Wash
FIGURE 1. Effect of phenylephrine on Zti induced by acetylstrophanthidin. Top traces in each panel are records of transmembrane current; bottom traces show membrane potential. Panel B was recorded 5 min after addition of phenylephrine; Panel C was recorded 5 min after switching to phenylephrine-free solution. Z,i, Transient inward current; PE, phenylephrine.
Alpha
Effects on TI Current
gure 1. The voltage steps imposed upon the tissue are shown at the bottom of each panel. The holding potential was -80 mV. Corresponding transmembrane currents are indicated by the top traces in each panel. Figure 1 (a) shows that a prominent Itr was observed following return to the holding potential after a 2-s voltage step to - 10 mV. Exposure of the tissue to phenylephrine ( IOF M) for 5 min resulted in a marked decrease in the peak current induced by the voltage step [Fig. 1 (b)]. The magnitude of the Iti returned to control levels within 5 min of washing phenylephrine from the tissue bath [Fig. 1 (c)]. This observation indicates that the direct effect of phenylephrine on the transmembrane current underlying OAP induced by acetlystrophanthidin is inhibitory, rather than stimulatory. Dose-response curves for phenylephrine were determined in three preliminary experiments. Inhibitory effects first appeared at approximately 10e8 M and maximal effects were induced by 10v6 or IO-’ M phenylephrine. The latter concentrations were used in subsequent experiments. Effects of phenylephrine in 12 experiments are summarized in Figure 2 (a). The magnitude of peak 1ti during control conditions and in the presence of phenylephrine are shown by the bars on the left. The inhibitory effect of phenylephrine was moderate (approximately 20%) but statistically significant at the P < 0.01 level (paired t-test). Phenylephrine also decreased net outward current at the end of the activating step to
193
- 10 mV. This is shown by the bars on the right in Figure 2(a) and also was statistically significant (P < 0.0 1, paired t-test). This effect represents a decrease in net tendency to repolarize and is compatible with prolongation of action potential duration reported by others (Giotti et al., 1973), and would be expected to promote induction of oscillatory afterpotentials indirectly. The inhibitory effect of phenylephrine on both Zti and Zstepwere completely abolished by a-adrenergic blockade in four experiments (two with prazosin, 5 x lop7 M; two with phentolamine, 1 or 5 x 10d6 M). It would therefore seem likely that both effects are mediated by al-adrenergic receptor activation. If the two effects also share a common pathway subsequent to al-receptor activation, one might predict that there would be a strong correlation between the magnitudes of these two effects. However, as shown by Figure 2(b), no correlation could be shown between the magnitudes of decrease in Zti and Istep. The decrease in Zti caused by phenylephrine might represent a decrease in the maximum current that can be activated or alternatively a shift in the current-voltage rela.tionship or activation properties of the 1ti. To differentiate between these possibilities, we determined the effects of phenylephrine on the current-voltage relationship and activation curves for the Iti in three experiments. Figure 3 shows the current-voltage relationship for a representative experiment. As indicated by
16,
50
(b)
’ r =-0.122 P < 0.45
L
40 -
.
a ,jz :30P 2 8 zo8
.
. . . .
IO.
%
. .
PE
C
0 I
step
I IO
I 20 % Decrease
I 30
I 40
!
Iti
FIGURE 2. (a) Effects ofphenylephrine on the peak magnitude ofZti and the net outward current (Istep) at the end of the activating voltage step. Both effects were significant at P < 0.01 level (**; paired t-test). (b) Lack of correlation between phenylephrine induced decreases in Z,,,, and Zti. r, Regression coefficient.
194
G. R. Ferrier
and
the schematic, the holding potential was again -80 mV and the Iti was activated by a voltage step to - 10 mV. However, the magnitude of the Iti was determined upon return to different membrane potentials ranging from -100 to -30 mV. The control current-voltage relationship showed peak current at membrane potentials near -65 mV. Phenylephrine caused a marked decrease in the magnitude of the Iti observed at all voltages with minimal shift in the current voltage relationship toward more positive potentials. The effects of phenylephrine completely reversed within 15 min of washing the drug from the tissue bath. A representative determination of “activation curves” is shown in Figure 4(a). Although the Iti is not activated through classic gating mechanisms, the magnitude of Iti varies with the voltage of the activating step (Kass et al., 1978a,b). As indicated by the schematic inset, the magnitude of the 1ti activated by 2-s voltage steps to various membrane potentials was measured at the holding potential. The control relationship is indicated by the curve in Figure 4(a) (open circles). Phenylephrine caused a decrease in the magnitude of the Zti as shown by the curve with filled circles [Fig. 4(a)]. This effect reversed when phenylephrine was washed from the tissue bath [Fig. 4(a) triangles]. Convergence
Membrane -100
-80
-60
potential
E. Carmeliet
of the control and phenylephrine treatment relationships at more negative activation voltages indicates that phenylephrine had little or no effect on the activation properties of the Iti. These experiments indicate that the main effect of phenylephrine is to decrease the magnitude of the Iti rather than to shift the current-voltage relationship or activation characteristics of the current. E#ect of prazosin on I,i induced by acetlystrophanthidin If the decrease in the magnitude of the Iti caused by phenylephrine is mediated by activation of a-adrenergic receptors, this effect should be blocked by a-adrenergic antagonists. The effects of prazosin were determined in the same experiment as illustrated in Figure 4(a) following washout of phenylephrine. The new control relationship is shown in Figure 4(b) (open circles). The preparation was then exposed to prazosin (5 x 10 -7 M) for 10 min and the “activation curve” was redetermined [Fig. 4(b), filled circles]. Prazosin alone caused a marked decrease in the magnitude of the 1ti especially at more positive activation steps. In the continued presence of prazosin, phenylephrine caused no further decrease in the magnitude of the 1u. Blockade of the inhibitory effects of phenylephrine by prazosin
(mV) -40
-20
0
-5
A IO-‘hn o Wash
FIGURE voltage step voltage step
3. Effects of phenylephrine sequence used to determine to - 10 mV.
on the current-voltage relationship the magnitude of the Zti activated
PE (5 min) (15
min)
of the Z,i. The schematic inset at various membrane potentials
indicates following
the a
Alpha IO 9 8
Effects
on TI
Current
195
prising. However, sodium channel blockade is not considered to be a prominent action of n Wash PE (IO min) although slight depression (4 to prazosin, 7 10%) of rabbit Purkinje fiber action potential 6 upstroke has been reported at concentrations 5 four or eight times higher than those used1 in 4 the present study (Dukes and Vaughan Wil3 liams, 1984). Several possible actions milght underlie the inhibitory effects of prazosin. The drug might: (a) have local anaesthetic activity under the conditions of this study; (b) displace acetylstrophanthidin from its receptor; (c) exert partial a-agonist activity in this experimental system; (d) directly interfere with lthe mechanism underlying induction of the Iri, either by direct inhibition of the Itl or by 5inhibition at a step in Ca handling directly related to induction of the current. Two approaches were used to establish whether inhibition of the Ie by prazosin wa.s accompanied by inhibition ofsodium influx at I I I I I I the concentrations used and under the con-40 -20 -10 0 -30 ditions of our experiments. In the first, the l/stepCm”) maximum rate of rise of the action potential was taken as a measure of the rapid sodium FIGURE 4. (a) Effect of phenylephrine on the relationship between peak Z,i and the voltage ofthe activation step inward current. The upstroke velocity was by the schematic (v,,,,)~ Vstep was varied as indicated determined at various membrane potentials inset. The magnitude of the Zti was measured at a holding by systematically varying the holding potentpotential of -80 mV. (b) Blockade of the effects of ial and interrupting the voltage clamp for a phenylephrine on the relationship between peak Zti and Vstep.PE, Phenylephrine; PZ, prazosin. IO-ms period every 5 s. A stimulus was delivered through the current passing electrode 1 ms after turning off the clamp and thus, or phentolamine (1 x low7 M) was confirmed before the membrane potential showed a in four experiments. Inhibition of the 1tr by ~1- measurable change from the holding potentadrenergic antagonist alone was observed in ial. The resulting upstroke was electronically three experiments with prazosin (5 x 10e7 M) differentiated. Plots of the maximum rate of and three experiments with phentolamine (1 rise as a function of membrane potential for or 5 x 10e6~). pre-treatment control, prazosin treatment, and 10 min washout of prazosin for a representative experiment are shown in Figure Assessment of possible mechanisms of inhibitory 5(a). Prazosin failed to change maximum rate action of praeosin on Iti of rise at any membrane potential. Absence lof Agents that block the fast inward sodium effect of prazosin on upstroke velocity was current, such as tetrodotoxin, quinidine, or confirmed in four experiments. lidocaine inhibit the 1ti and decrease the We also assessed possible effects of prazosin amplitudes of oscillatory afterpotentials (Kass on slowly inactivating sodium current. Caret al., 1978b; Wasserstrom and Ferrier, 1982; meliet (1987) has shown that a component of Rosen and Danilo, 1980). It is established that sodium current in rabbit Purkinje fibws phentolamine also has sodium channel blockinactivates very slowly and can contribute to ing efficacy and that this effect is first seen at net currents observed during the plateau of concentrations close to those used in the prethe action potential and during diastole. The sent study (Northover, 1983). The inhibitory voltage clamp protocol used to activate the effect of phentolamine therefore was not surslowly inactivating sodium current is shown
196
G. R. Ferrier
and
schematically in Figure 5(b). The sodium current was measured during 3-s long depolarizing steps to -45 mV from a holding potential of - 80 mV at 60-s intervals. Essentially lOOo/o of the inward current activated during the depolarizing steps is carried by a sodium conductance that is tetrodotoxin sensitive (Carmeliet, 1987). The current activated during the control period in a representative experiment is shown in the top panel of Figure 5(b). Exposure ofthe tissue to 1O-6 M prazosin for 10 min had no effect on the magnitude or time course of this current [Fig. 5(b), middle panel]. The bottom panel of Figure 5(b) shows that the current traces recorded during the control and prazosin treatment periods were superimposable. Identical observations were made in three experiments. These observations indicate that the inhibition of the 1,: by prazosin is probably not mediated by local anaesthetic action. To determine whether the inhibitory effects of prazosin on the Iu were mediated by displacement of acetylstrophanthidin, we examined the effects of prazosin in three preparations in which the current was induced by elevated calcium concentration in the absence of a digitalis agent. Figure 6 shows records from one of those experiments. The top left panel shows that following exposure of rabbit Purkinje fibers to 8 mM Ca and 2 mM K, activation steps to 0, - 10, or -20 mV re-
E. Ckrmeliet
sulted in prominent 1ti currents. Following exposure of the tissue to prazosin, the magnitudes of the Iri induced by activation steps to the same potentials clearly decreased as seen in the top right panel. Prazosin caused similar inhibition of the calcium-induced Zti in all three experiments. Thus the inhibitory effect of prazosin was not mediated by displacement of acetylstrophanthidin. The experiment illustrated in Figure 6 also addressed the possibility that prazosin might exert partial agonist activity under our experimental conditions. A partial agonist would be expected to exert a-adrenergically mediated inhibitory effects of its own, but attenuate the effects of a full agonist like phenylephrine. If prazosin has partial agonist activity it should be possible to block that effect with another aadrenergic antagonist like phentolamine. Therefore, following washout of prazosin we exposed the Purkinje fiber to phentolamine at a concentration lo-fold that of prazosin (Fig. 6, bottom left panel). Phentolamine caused a decrease in the magnitude of the 1ti as expected. We therefore adjusted the activating voltage steps 10 mV positive to again elicit currents of substantial magnitudes. The records in Figure 6 (bottom left) were recorded after 15 min exposure to phentolamine at which time the amplitudes of the 1ti were stable. The preparation was then exposed to the same concentration of prazosin as before
(b)
Control
12OOr -T ;
800
3 -$
600
56
Prazosin (lO-‘nn) 10 min
400
i
Control +
Prozosin 01
’ -85
I
I
-80
-75
“Take-off”
potential
FIGURE 5. (a) Lack of effect of brane potentials during interruption transmembrane potential at initiation prazosin. The voltage clamp protocol during control and prazosin treatment
,
-70
/ I
-65
(mV)
prazosin on maximum upstroke
5nAk
G
mEmv
velocity of Purkinje fibers at different transmemof the voltage clamp. Vmax, maximum upstroke velocity; “take-or potential, of the upstroke. (b) Lack of inhibition of slowly inactivating sodium current by is indicated schematically by the inset. The top and middle traces show currents periods. The bottom trace shows the control and prazosin traces superimposed.
Alpha
Effects on TI Current 8mM
Ca.
2mMK
Control V step
O1 ‘\:, i
(mV)
Prazosin -20
-10 1
o-
(5 x 16’~) -20
Oi
j \
197
I’! / / j
-10
;
I\
‘+---+ 6
‘1 Phentoiamine
(5 x 10v6 M)
Phentolamine
r
‘T + Prazosin
FIGURE 6. Effects of prazosin alone or with phentolamine on Z,i induced by 8 mM Ca and 2 rnM K. Numbers beside outward (upward) current excursions indicate the voltage of the respective activation steps ( VSteP). VH, hoiding potential. Top panels shdw that inhibition ofZ,i by prazosin occurs in the absence of acetylstrophanthdin. Bottom panels show that phentolamine does not prevent inhibition of Z,i by prazosin.
but in the continued presence of phentolamine. Prazosin again exerted its inhibitory effect (Fig. 6, bottom Gght) thereby indicating that it is unlikely that prazosin’s inhibitory effect on the 1,i in the absence of agonist is mediated by partial a-adrenergic agonist activity. This observation was confirmed in three experiments in which the I,i was induced by calcium and three experiments in which the 1ti was induced by acetylstrophanthidin.
Effects of phenylephrine
on
I,i induced by high Ca
Observations presented so far indicate that both phenylephrine and prazosin exert inhibitory effects on the 1ti. However, observations that a-adrenergic agonists promote induction of oscillatory afterpotentials were made in experiments in which afterpotentials were induced by elevated calcium concentration (Kimura et al., 1984, 1987). We therefore also determined the effects of phenylephrine in preparations in which the Iti was induced by 8 mM Ca and 2 mM K. Figure 7 shows records of transmembrane current from a representative experiment. The voltage steps used to elicit the 1i, were identical to those in
experiments with acetlystrophanthidin. The control record shows that high calcium also elicited a large Iti. However, in six of six preparations in which the I,i was induced by elevated Ca, phenylephrine caused a marked increase in the peak amplitudes of the 1ti (control: 16.7 + 2.4 nA + S.E.; phenylephrine: 29.8 + 2.6 nA + s.E.; P < 0.001, paired ttest). This stimulatory effect of phenylephrine dissipated when the drug was washed from the tissue bath. Effects of phenylephrine on the net inward current at the end of the activation step (Istep) were weak or absent. To determine whether the stimulatory effect of phenylephrine on Zti induced by high calcium concentration was mediated through activation of al-adrenoceptors, the effect of phenylephrine on peak Iti was assessed in five additional experiments in which tissues were pre-treated with prazosin (5 x 10m7 M). Under these conditions, phenylephrine no longer caused a significant change in peak 1ti (control: 13.3 f 2.4 nA & S.E.; phenylephrine: 15.1 + 3.4 nA & S.E.). Thus, both the stimulatory effect of phenylephrine on Iti induced by high calcium and the inhibit09 effect on Zti induced by acetylstrophanthidin were abolished by specific al-adrenergic blockade.
198
G. R. Ferrier _-
8mt.! Control
and Ca.
E. Carmeliet 2mM
PE (IO+hn)
K Wash
-I
5 nA
5s
FIGURE is indicated
7. Stimulatory effect ofphenylephrine by the schematic inset
on Zti induced
Discussion Dzyerential
effects of phenylephrine on Iti induced by acetlystrophanthidin and Ca
The present study suggests that earlier contradictory observations by others concerning the effect of a-adrenergic agonists on OAP, in fact, might both be correct. Alpha-adrenergic agonists may exert either inhibitory or stimulatory effects on the 1ri depending on the mechanism by which the current is induced. Thus, the stimulatory effect of phenylephrine on OAP induced by high calcium concentration reported by Kimura et al. (1984, 1987) is compatible with the stimulatory effect of phenylephrine on the Irr seen in the present study. Similarly, the lack of effect of phenylephrine on OAP induced by ouabain reported by Hewett and Rosen ( 1984) may represent the sum of the inhibitory effect of phenylephrine on the Iu plus an opposing stimulatory effect mediated by prolongation of the action potential duration. Thus, it becomes important to recognize that triggered activity secondary to the 1ti can have important and diverse characteristics depending on the conditions which induce the Iti. Although the present observations provide a possible explanation for earlier conflicting studies, they do not clearly predict the direction of the effect that might be exerted on OAP-induced arrhythmias in other settings
by 8 rnM Ca and 2 rnM K. The voltage
clamp
protocol
such as reperfusion. This is so because the chain of events involved in induction of OAP in reperfusion is not fully known and may not closely mimic either of the present test systems (Ferrier et al., 1985). Also, ischemia and reperfusion are known to modify the number of aadrenergic receptors at the cell surface (Corr et al., 1981; Maisel et al., 1987; Dillon et al., 1988), and to convert the response of idioventricular rate to stellate nerve stimulation from one that is propranolol sensitive to one that is phentolamine sensitive (Sheridan et al., 1980). This alteration of response to nerve stimulation could represent either a change in the mechanism of the idioventricular activity or modification of the coupling of aadrenoceptors to cellular or membrane functions. The present study underscores the importance of pre-treating preparations with blocking agents in order to identify the specific receptor invoIved in drug or transmitter actions. In experiments on Irr or OAP in which the agonist is introduced first and the blocker is added later, the inhibitory effect of prazosin on the 1tr could easily be misinterpreted as reversal of a stimulatory effect of phenylephrine mediated by oll-adrenoceptor blockade. This error could occur even in settings in which the agonist might have no effect mediated by a-adrenoceptors. At the con-
Alpha
Effects
centration used in the present study, prazosin reduced the amplitude but did not abolish the Zti. Although the inhibitory effect of prazosin appears to be mediated by an action at a site other than the a-adrenoceptor, this agent would still occupy al-adrenoceptors and block stimulation by phenylephrine. Thus, it was still possible to test the effect of agonist after pre-treatment with prazosin. This permitted demonstration that both the inhibitory and stimulatory effects of phenylephrine on the 1ti were mediated by specific al-adrenoceptor activation. The presence of propranolol throughout, eliminated effects mediated by fiadrenoceptor activation. The mechanism by which opposite effects of phenylephrine on the magnitude of the Iti are elicited depending on the method of induction of the current is not clear. Phenylephrine is known to have several effects in cardiac tissue that would be expected to affect generation of this current. This agent can induce calcium dependent slow action potentials in rabbit papillary muscle depolarized by elevated levels of potassium (Miura et al., 1987). Further, Bruckner and Scholz (1984) reported that phenylephrine increased the magnitude of the slow inward calcium current in bovine trabeculae in the presence of fi-adrenergic blockade with propranolol. An increase in calcium influx would be expected to promote induction of the 1ti and OAP. On the other hand, Shah et al. (1988) have reported that aadrenergic agonists decrease the intracellular sodium concentration possibly by stimulating the sodium pump. One would predict that this effect would cause a decrease in intracellular calcium concentration and thereby inhibit the induction of the 1tr or OAP. The relative importance of these two actions might differ with the conditions that induce the Iti. In the presence of an elevated calcium gradient, an increase in the open time or number of slow channels would be expected to greatly accentuate calcium influx by this route. The effect of altering the intracellular concentration of sodium might be of lesser importance in this setting, and a net stimulatory effect may result. However, when the sodium pump is partially inhibited by digitalis, the effects of phenylephrine on intracellular sodium concentration may predominate and result in a net inhibitory effect on the 1rr.
on TI Current
199
An alternate explanation for the opposing effects of phenylephrine on the Iti could be proposed on the basis of evidence suggesting that the 1rr may be derived from two different current carrying components: a Na/Ca exchange current and a non-selective cation current (for a discussion see Ehara et al., 1988). The relative importance of these two components would depend on the experimental condition. In high extracellular Ca, the contribution of the inward Na/Ca exchange current may be greater (due to low intracellular Na) than that of the nonspecific cation current. In digitalis intoxication on the other hand the nonspecific cation current may be favoured. If one hypothesizes that PE stimulates Na/Ca exchange but decreases the contribution of a non-selective cation current, the Iti and OAP would be enhanced in high Ca, but decreased in digitalis intoxication.
Inhibitory
efects of prazosin on I,i induced by acetylstrophanthidin or Ca
The present study also demonstrates that prazosin has an inhibitory effect on Zu which is independent of its a-adrenergic antagonist activity. Clearly studies identifying a-adrenergic mechanisms in the generation of arrhythmia, on the basis of effects of u-antagonists added after pretreatment with agonist, must be interpreted with great caution because inhibitory effects unrelated to a-adrenergic blockade may have been mistaken for specific aadrenergic antagonism. The mechanism of the inhibitory action of prazosin on the 1ti is not known. This effect was not abolished by a lo-fold higher con centration of phentolamine. Also, the effect of prazosin did not become stimulatory, like that of phenylephrine, when the Irr was induced by high Ca instead of acetylstrophanthidin. Therefore, it is unlikely that the effect is mediated via partial agonist activity at the CIadrenoceptor. We could find no evidence of sodium channel blocking action at a concentration that strongly inhibited the Zri, and the inhibitory effect was exerted on the Iti when it was induced by either acetylstrophanthidin or high calcium concentration. Thus, neither local anaesthetic action nor displacement of digitalis-like agents are tenable ex-
200
G. R. Ferrier
planations. Prazosin therefore must inhibit the Iri charge carrying mechanism or act at some link between calcium overload and activation of the current. These observations also suggest that prazosin might have antiarrhythmic efficacy even in settings in which sympathetic activity is not a prominent factor in induction of the arrhythmia.
and E. Carmeliet Acknowledgements The authors thank J. Prenen for excellent technical assistance and X. Han for participation in some of the experiments. This study was supported in part by the Medical Research Council of Canada, and the Visiting Scientist Program of the Canadian Heart Foundation.
References ARONSON RS, GELLES JM,
HOFFMAN BF (1973) A new method for producing short cardiac Purkinje fibers suitable for clamp. J Appl Physiol34: 527-530. BRUCKNER R, SCHOLZ H (1984) Effects ofa-adrenoceptor stimulation with phenylephrine in the presence of propranolol on force of contraction, slow inward current and cyclic AMP content in the bovine heart. Br J Pharmacol82: 223-232. CARMELIET C (1987) Slow inactivation of the sodium current in rabbit cardiac Purkinje fibres. Pfliigers Arch 408: 18-26. CORR PB, SHAYMAN JA, KRAMER JB (1981) I ncreased a-adrenergic receptors in ischemic cat myocardium. A potential mediator for electrophysiological derangements. J Clin Invest 67: 1232-1236. DILLON JS, Gu XH, NAYLER WG (1988) Alpha, adrenoceptors in the ischaemic and reperfused myocardium. J Mol Cell Cardiol 20: 725-735. DUKES ID, VAUGHAN WILLIAMS EM (1984) Electrophysiological effects of cl-adrenoceptor antagonists in rabbit sinoatria1 node, cardiac Purkinje cells and papillary muscles. Br J Pharmacol83: 419-426. EHARA T, NOMA A, ONO K (1988) Calcium-activated non-selective cation channel in ventricular cells isolated from adult guinea pig heart. J Physiol, Lond 403: 117-133. FERRIER GR, MOFFAT MP, LUKAS A (1985) P ossible mechanisms of ventricular arrhythmias elicited by &hernia followed by reperfusion. Studies on isolated canine ventricular tissues. Circ Res 56: 184-194. GIOTTI A, LEDDA F, MANNAIONI PF (1973) Effects of noradrenaline and isoprenaline, in combination with a- and breceptor blocking substances, on the action potential of cardiac purkinje fibres. J Physiol, Lond 229: 99-l 13. HEWETT KW, ROSEN MR (1984) Alpha and beta adrenergic interactions with ouabain-induced delayed afterdepolarizations. J Pharmacol Exp Ther 229: 188-192. KASS RS, LEDERER WJ, TSIEN W, WEINGART R (1978a) Role of calcium ions in transient inward currents and aftercontractions induced by strophanthidin in cardiac Purkinje fibres. J Physiol, Lond 281: 187-208. KA~.s RS, TSIEN W, WEINGART, R (1978b) I onic basis of transient inward current induced by strophanthidin in cardiac Purkinje fibres. J Physiol, Lond 281: 209-226. KIMURA S, CAMERON, JS, KOZLOVSKIS PL, BASSETT, AL, MYERBURG, RJ (1984) Delayed afterdepolarizations and triggered activity induced in feline Purkinje fibers by cc-adrenergic stimulation in the presence ofelevated calcium levels. Circulation 70: 1074-1082. KIMURA S, BASSETT AL, KOHYA T, KOZLOVS~IS PL, MYERBURG RJ (1987) Automaticity, triggered activity, and responses to adrenergic stimulation in cat subendocardial Purkinje fibers after healing of myocardial infarction. Circulation 75: 651-660. MAISEL AS, MOTULSKY HJ, ZIEGLER G, INSEL PA (1987) Ischemiaand agonist-induced changes in cc- and /l-adrenergic traffic in guinea pig hearts. Am J Physiol 253: H 1159-H 1166. MIURA Y, INUI J, IMAMURA H (1978) Alpha-adrenoceptor-mediated restoration of calcium-dependent potential in the partially depolarized rabbit papillary muscle. Naunyn-Schmiedeberg’s Arch Pharmacol302: 201-205. MUBAGWA K, CARMELIET E (1983) Effects of acetylcholine on electrophysiological properties of rabbit cardiac Purkinje fibers. Circ Res 53: 740-75 1. NORTHOVER BJ (1983) A comparison of the electrophysiological actions of phentolamine with those of some other antiarrhythmic drugs on tissues isolated from the rat heart. Br J Pharmacol 80: 85-93. PENNY WJ, CULLING W, LEWIS MJ, SHERIDAN DJ (1985) Antiarrhythmic and electrophysiological effects of aIpha adrenoceptor blockade during myocardial ischaemia and reperfusion in isolated guinea-pig heart. J Mel Cell Cardiol voltage
17: 399409. POGWIZD SM,
CORR PB (1987) Electrophysiologic mechanisms underlying arrhythmias due to reperfusion of ischemic myocardium. Circulation 76: 404426. ROSEN MR, DANILO P (1980) Effects of tetrodotoxin, lidocaine, verapamil and AHR-2666 on ouabain-induced delayed afterdepolarizations in canine Purkinje fibers. Circ Res 46: 117-124. SHAH A, COHEN IS, ROSEN MR (1988) Stimulation of cardiac alpha receptors increases Na/K pump current and decreases gs via a pertussis toxin-sensitive pathway. Biophys J 54: 219-225. SHERIDAN DJ, PENKOSKE PA, SOBEL BE, CORR PB (1980) Alpha adrenergic contributions to dysrhythmia during myocardial ischemia and reperfusion in cats. J Clin Invest 65: 16 1-l 7 1. WASSERSTROM JA, FERRIER GR (1982) Effects of phenytoin and quinidine on digitalis-induced oscillatory afterpotentials, aftercontractions, and inotropy in canine ventricular tissues. J Mol Cell Cardiol 14: 725-736.
Cell
Cardio122,
191-200
Effects
( 1990)
of a-Adrenergic Agents on Transient Current in Rabbit Purkinje Fibers Gregory
R. Ferrier*
and Edward
Inward
Carmeliet
Laboratory of Physiology, University of Leuven, Campus Gasthuisberg, B-3000 Leuven, Belgium (Received 5 January 1989, accepted in revisedform 11 October 1989) G. FERRIER AND E. CARMELIET. Effects ofa-Adrenergic Agents on Transient Inward Current in Rabbit Purkinje Fibers. Journal ofMolecular and Cellular Cardiolo~ (1990) 22, 191-200. Reported effects of a-adrenergic agents on
oscillatory afterpotentials (OAP) are conflicting. Therefore, we used standard two microelectrode voltage clamp techniques to determine the effects of phenylephrine and prazosin on the transient inward current (Zri) responsible for OAP. The Zti was induced in isolated rabbit Purkinje fibers either by acetylstrophanthidin or 8 rnr,r Ca. The magnitude of the Zti was determined at various membrane potentials after activation by steps to - 10 mV from a holding potential of -80 mV. When the Zti was induced by acetylstrophanthidin, phenylephrine (10e7 to 10e5 M) caused inhibition of Zti at all potentials tested. Phenylephrine also caused a significant decrease in net outward current at plateau voltages. Both effects were blocked by prazosin (5 x IO-’ M) but not by propranolol (5 x lo-’ M). Prazosin also strongly inhibited the Z,t in the absence ofphenylephrine. At 5 x 10v7 M, prazosin did not affect sodium current activated by voltage steps from -80 to -45 mV or maximum upstroke velocity during interruptions of the voltage clamp. When the Ztt was induced by 8 rnM Ca, the effect of phenylephrine, but not prazosin, reversed so that phenylephrine increased the amplitude of the Ztt. Thus, a-adrenergic agonists may exert either inhibitory or stimulatory effects on the Ztt depending on the mechanism by which the current is induced. Additional effects on OAP amplitude may be induced by changes in action potential duration mediated through actions on net outward current. Prazosin may suppress OAP by an action on the Ztt which is independent of a-adrenergic or local anaesthetic actions. KEY
a-Adrenergic agonists; fiber; Afterpotentials.
WORDS:
Purkinje
a-Adrenergic
antagonists;
Introduction Several studies suggest that a-adrenergic agonists promote cardiac arrhythmias in various settings including reperfusion of previously ischemic myocardium (Sheridan et al., 1980; Penny et al., 1985). In studies utilizing isolated tissues, Ferrier et al. (1985) demonstrated that exposure of Purkinje tissue to ischemic conditions followed by return to nonischemic conditions induced a series of potential arrhythmic mechanisms, including oscillatory afterpotentials (OAP, also referred to as delayed afterdepolarizations). The possibility that OAP constitute an important mechanism of reperfusion arrhythmias has also been suggested by Pogwizd and Corr (1987) who used computer-assisted mapping to document the *Please address all correspondence Medical Building, Dalhousie University, 0022-2828/90/020191
+ 10 $03.0010
Cardiac
arrhythmias;
Transient
inward
current;
initiation of reperfirsion arrhythmias in open chest cats subjected to coronary artery occlusion and reflow. They observed that 75% of nonsustained ventricular tachycardias induced by reperfusion in cats, were initiated by a non-reentrant mechanism in the subendocardium. They suggested that their observations were compatible with OAP as the mechanism responsible for initiation of these arrhythmias, and further suggested that Madrenergic activation might be responsible for promotion of OAP during reperfusion. Direct evidence for potentiation of OAP by cl-adrenergic agonists has not been consistent. Kimura et al. (1984,1987) showed that phenylephrine or methoxamine would promote the development of OAP in response to 8 mM Ca
to: Gregory R. Ferrier, Department of Pharmacology, Halifax, Nova Scotia, Canada B3H 4H7.
Sir Charles
0 1990 Academic
Tupper
Press Limited
192
G. R. Ferrier
in isolated cat Purkinje fibers. On the other hand, Hewett and Rosen (1984) found no effect of these same agonists on OAP induced by ouabain in canine Purkinje fibers. These inconsistent observations might be explained if a-agonists have multiple actions with opposing effects on OAP so that in different settings either no effect or exacerbation of OAPinduced arrhythmias is seen. Indeed, it is well known that a-adrenergic agonists prolong action potential duration (Giotti et al., 1973). This effect would be expected to promote the appearance and increase the amplitude of OAP (Kass et al., 1978a). In contrast, the effects of a-agonistp directly on the transient inward current (Iti) are unknown. Therefore, the present study was conducted under voltage clamp conditions in order to determine the effects of a-adrenergic agonists on the Iti under conditions in which there can be no influences from changes in action potential duration. This information is necessary to clarify the basis for the conflicting observations from various laboratories concerning the effects of a-adrenergic agents on OAP.
and
E. Carmeliet
ing steps was 3 per min in most experiments. Recordings Were made on a Tektronix 5lllA storage oscilloscope and a Gould Brush (2400) rectilinear pen recorder. All drugs except propranolol were prepared freshly as aqueous stock solutions ( lop3 or 10e4 M). Aliquots of stock solutions were diluted directly in the Tyrode’s superfusate to achieve final concentrations. Phenylephrine, and prazosin were obtained from Sigma. Propranolol (Inderal) was obtained as injectable solution from ICI, and phentolamine (regitine), from CIBA. All except some preliminary experiments were conducted in the presence of 5 X lo-’ M propranolol to eliminate p-adrenergic effects. Results Effects of phmylephrine on Iti induced by acetylstrophanthidin The effect of phenylephrine on the magnitude of 1,i induced by acetylstrophanthidin in a representative experiment is illustrated in Fi-
Methods The effects of phenylephrine were studied on Iti induced either by acetylstrophanthidin (AS) or elevated calcium concentration (8 mM). All studies were conducted on rabbit Purkinje fibers to eliminate differences due to species. Rabbits were sacrificed by cervical dislocation. Short segments (0.8 to 1 .O mm) of Purkinje fibers dissected from both ventricles were obtained by crushing the isolated fibers with a wire grid (Aronson et al., 1973). The composition of modified Tyrode’s solutions used for dissection and superfusion have been reported by Mubagwa and Carmeliet (1983). Experiments were performed at 37°C. The two microelectrode voltage clamp technique was used. Bevelled microelectrodes filled with 2.0 M KC1 and having resistances between 5 and 12 MLI were used for both current passage and voltage recording. Voltage steps were generated by a digital timing circuit which controlled the command voltage of the voltage clamp feed-back circuit. The Zti was elicited by 2- or 3-s long depolarizing voltage steps usually from a holding potential of -80 mV. The repetition rate of depolariz-
2s Control
PE (IO+M)
Wash
FIGURE 1. Effect of phenylephrine on Zti induced by acetylstrophanthidin. Top traces in each panel are records of transmembrane current; bottom traces show membrane potential. Panel B was recorded 5 min after addition of phenylephrine; Panel C was recorded 5 min after switching to phenylephrine-free solution. Z,i, Transient inward current; PE, phenylephrine.
Alpha
Effects on TI Current
gure 1. The voltage steps imposed upon the tissue are shown at the bottom of each panel. The holding potential was -80 mV. Corresponding transmembrane currents are indicated by the top traces in each panel. Figure 1 (a) shows that a prominent Itr was observed following return to the holding potential after a 2-s voltage step to - 10 mV. Exposure of the tissue to phenylephrine ( IOF M) for 5 min resulted in a marked decrease in the peak current induced by the voltage step [Fig. 1 (b)]. The magnitude of the Iti returned to control levels within 5 min of washing phenylephrine from the tissue bath [Fig. 1 (c)]. This observation indicates that the direct effect of phenylephrine on the transmembrane current underlying OAP induced by acetlystrophanthidin is inhibitory, rather than stimulatory. Dose-response curves for phenylephrine were determined in three preliminary experiments. Inhibitory effects first appeared at approximately 10e8 M and maximal effects were induced by 10v6 or IO-’ M phenylephrine. The latter concentrations were used in subsequent experiments. Effects of phenylephrine in 12 experiments are summarized in Figure 2 (a). The magnitude of peak 1ti during control conditions and in the presence of phenylephrine are shown by the bars on the left. The inhibitory effect of phenylephrine was moderate (approximately 20%) but statistically significant at the P < 0.01 level (paired t-test). Phenylephrine also decreased net outward current at the end of the activating step to
193
- 10 mV. This is shown by the bars on the right in Figure 2(a) and also was statistically significant (P < 0.0 1, paired t-test). This effect represents a decrease in net tendency to repolarize and is compatible with prolongation of action potential duration reported by others (Giotti et al., 1973), and would be expected to promote induction of oscillatory afterpotentials indirectly. The inhibitory effect of phenylephrine on both Zti and Zstepwere completely abolished by a-adrenergic blockade in four experiments (two with prazosin, 5 x lop7 M; two with phentolamine, 1 or 5 x 10d6 M). It would therefore seem likely that both effects are mediated by al-adrenergic receptor activation. If the two effects also share a common pathway subsequent to al-receptor activation, one might predict that there would be a strong correlation between the magnitudes of these two effects. However, as shown by Figure 2(b), no correlation could be shown between the magnitudes of decrease in Zti and Istep. The decrease in Zti caused by phenylephrine might represent a decrease in the maximum current that can be activated or alternatively a shift in the current-voltage rela.tionship or activation properties of the 1ti. To differentiate between these possibilities, we determined the effects of phenylephrine on the current-voltage relationship and activation curves for the Iti in three experiments. Figure 3 shows the current-voltage relationship for a representative experiment. As indicated by
16,
50
(b)
’ r =-0.122 P < 0.45
L
40 -
.
a ,jz :30P 2 8 zo8
.
. . . .
IO.
%
. .
PE
C
0 I
step
I IO
I 20 % Decrease
I 30
I 40
!
Iti
FIGURE 2. (a) Effects ofphenylephrine on the peak magnitude ofZti and the net outward current (Istep) at the end of the activating voltage step. Both effects were significant at P < 0.01 level (**; paired t-test). (b) Lack of correlation between phenylephrine induced decreases in Z,,,, and Zti. r, Regression coefficient.
194
G. R. Ferrier
and
the schematic, the holding potential was again -80 mV and the Iti was activated by a voltage step to - 10 mV. However, the magnitude of the Iti was determined upon return to different membrane potentials ranging from -100 to -30 mV. The control current-voltage relationship showed peak current at membrane potentials near -65 mV. Phenylephrine caused a marked decrease in the magnitude of the Iti observed at all voltages with minimal shift in the current voltage relationship toward more positive potentials. The effects of phenylephrine completely reversed within 15 min of washing the drug from the tissue bath. A representative determination of “activation curves” is shown in Figure 4(a). Although the Iti is not activated through classic gating mechanisms, the magnitude of Iti varies with the voltage of the activating step (Kass et al., 1978a,b). As indicated by the schematic inset, the magnitude of the 1ti activated by 2-s voltage steps to various membrane potentials was measured at the holding potential. The control relationship is indicated by the curve in Figure 4(a) (open circles). Phenylephrine caused a decrease in the magnitude of the Zti as shown by the curve with filled circles [Fig. 4(a)]. This effect reversed when phenylephrine was washed from the tissue bath [Fig. 4(a) triangles]. Convergence
Membrane -100
-80
-60
potential
E. Carmeliet
of the control and phenylephrine treatment relationships at more negative activation voltages indicates that phenylephrine had little or no effect on the activation properties of the Iti. These experiments indicate that the main effect of phenylephrine is to decrease the magnitude of the Iti rather than to shift the current-voltage relationship or activation characteristics of the current. E#ect of prazosin on I,i induced by acetlystrophanthidin If the decrease in the magnitude of the Iti caused by phenylephrine is mediated by activation of a-adrenergic receptors, this effect should be blocked by a-adrenergic antagonists. The effects of prazosin were determined in the same experiment as illustrated in Figure 4(a) following washout of phenylephrine. The new control relationship is shown in Figure 4(b) (open circles). The preparation was then exposed to prazosin (5 x 10 -7 M) for 10 min and the “activation curve” was redetermined [Fig. 4(b), filled circles]. Prazosin alone caused a marked decrease in the magnitude of the 1ti especially at more positive activation steps. In the continued presence of prazosin, phenylephrine caused no further decrease in the magnitude of the 1u. Blockade of the inhibitory effects of phenylephrine by prazosin
(mV) -40
-20
0
-5
A IO-‘hn o Wash
FIGURE voltage step voltage step
3. Effects of phenylephrine sequence used to determine to - 10 mV.
on the current-voltage relationship the magnitude of the Zti activated
PE (5 min) (15
min)
of the Z,i. The schematic inset at various membrane potentials
indicates following
the a
Alpha IO 9 8
Effects
on TI
Current
195
prising. However, sodium channel blockade is not considered to be a prominent action of n Wash PE (IO min) although slight depression (4 to prazosin, 7 10%) of rabbit Purkinje fiber action potential 6 upstroke has been reported at concentrations 5 four or eight times higher than those used1 in 4 the present study (Dukes and Vaughan Wil3 liams, 1984). Several possible actions milght underlie the inhibitory effects of prazosin. The drug might: (a) have local anaesthetic activity under the conditions of this study; (b) displace acetylstrophanthidin from its receptor; (c) exert partial a-agonist activity in this experimental system; (d) directly interfere with lthe mechanism underlying induction of the Iri, either by direct inhibition of the Itl or by 5inhibition at a step in Ca handling directly related to induction of the current. Two approaches were used to establish whether inhibition of the Ie by prazosin wa.s accompanied by inhibition ofsodium influx at I I I I I I the concentrations used and under the con-40 -20 -10 0 -30 ditions of our experiments. In the first, the l/stepCm”) maximum rate of rise of the action potential was taken as a measure of the rapid sodium FIGURE 4. (a) Effect of phenylephrine on the relationship between peak Z,i and the voltage ofthe activation step inward current. The upstroke velocity was by the schematic (v,,,,)~ Vstep was varied as indicated determined at various membrane potentials inset. The magnitude of the Zti was measured at a holding by systematically varying the holding potentpotential of -80 mV. (b) Blockade of the effects of ial and interrupting the voltage clamp for a phenylephrine on the relationship between peak Zti and Vstep.PE, Phenylephrine; PZ, prazosin. IO-ms period every 5 s. A stimulus was delivered through the current passing electrode 1 ms after turning off the clamp and thus, or phentolamine (1 x low7 M) was confirmed before the membrane potential showed a in four experiments. Inhibition of the 1tr by ~1- measurable change from the holding potentadrenergic antagonist alone was observed in ial. The resulting upstroke was electronically three experiments with prazosin (5 x 10e7 M) differentiated. Plots of the maximum rate of and three experiments with phentolamine (1 rise as a function of membrane potential for or 5 x 10e6~). pre-treatment control, prazosin treatment, and 10 min washout of prazosin for a representative experiment are shown in Figure Assessment of possible mechanisms of inhibitory 5(a). Prazosin failed to change maximum rate action of praeosin on Iti of rise at any membrane potential. Absence lof Agents that block the fast inward sodium effect of prazosin on upstroke velocity was current, such as tetrodotoxin, quinidine, or confirmed in four experiments. lidocaine inhibit the 1ti and decrease the We also assessed possible effects of prazosin amplitudes of oscillatory afterpotentials (Kass on slowly inactivating sodium current. Caret al., 1978b; Wasserstrom and Ferrier, 1982; meliet (1987) has shown that a component of Rosen and Danilo, 1980). It is established that sodium current in rabbit Purkinje fibws phentolamine also has sodium channel blockinactivates very slowly and can contribute to ing efficacy and that this effect is first seen at net currents observed during the plateau of concentrations close to those used in the prethe action potential and during diastole. The sent study (Northover, 1983). The inhibitory voltage clamp protocol used to activate the effect of phentolamine therefore was not surslowly inactivating sodium current is shown
196
G. R. Ferrier
and
schematically in Figure 5(b). The sodium current was measured during 3-s long depolarizing steps to -45 mV from a holding potential of - 80 mV at 60-s intervals. Essentially lOOo/o of the inward current activated during the depolarizing steps is carried by a sodium conductance that is tetrodotoxin sensitive (Carmeliet, 1987). The current activated during the control period in a representative experiment is shown in the top panel of Figure 5(b). Exposure ofthe tissue to 1O-6 M prazosin for 10 min had no effect on the magnitude or time course of this current [Fig. 5(b), middle panel]. The bottom panel of Figure 5(b) shows that the current traces recorded during the control and prazosin treatment periods were superimposable. Identical observations were made in three experiments. These observations indicate that the inhibition of the 1,: by prazosin is probably not mediated by local anaesthetic action. To determine whether the inhibitory effects of prazosin on the Iu were mediated by displacement of acetylstrophanthidin, we examined the effects of prazosin in three preparations in which the current was induced by elevated calcium concentration in the absence of a digitalis agent. Figure 6 shows records from one of those experiments. The top left panel shows that following exposure of rabbit Purkinje fibers to 8 mM Ca and 2 mM K, activation steps to 0, - 10, or -20 mV re-
E. Ckrmeliet
sulted in prominent 1ti currents. Following exposure of the tissue to prazosin, the magnitudes of the Iri induced by activation steps to the same potentials clearly decreased as seen in the top right panel. Prazosin caused similar inhibition of the calcium-induced Zti in all three experiments. Thus the inhibitory effect of prazosin was not mediated by displacement of acetylstrophanthidin. The experiment illustrated in Figure 6 also addressed the possibility that prazosin might exert partial agonist activity under our experimental conditions. A partial agonist would be expected to exert a-adrenergically mediated inhibitory effects of its own, but attenuate the effects of a full agonist like phenylephrine. If prazosin has partial agonist activity it should be possible to block that effect with another aadrenergic antagonist like phentolamine. Therefore, following washout of prazosin we exposed the Purkinje fiber to phentolamine at a concentration lo-fold that of prazosin (Fig. 6, bottom left panel). Phentolamine caused a decrease in the magnitude of the 1ti as expected. We therefore adjusted the activating voltage steps 10 mV positive to again elicit currents of substantial magnitudes. The records in Figure 6 (bottom left) were recorded after 15 min exposure to phentolamine at which time the amplitudes of the 1ti were stable. The preparation was then exposed to the same concentration of prazosin as before
(b)
Control
12OOr -T ;
800
3 -$
600
56
Prazosin (lO-‘nn) 10 min
400
i
Control +
Prozosin 01
’ -85
I
I
-80
-75
“Take-off”
potential
FIGURE 5. (a) Lack of effect of brane potentials during interruption transmembrane potential at initiation prazosin. The voltage clamp protocol during control and prazosin treatment
,
-70
/ I
-65
(mV)
prazosin on maximum upstroke
5nAk
G
mEmv
velocity of Purkinje fibers at different transmemof the voltage clamp. Vmax, maximum upstroke velocity; “take-or potential, of the upstroke. (b) Lack of inhibition of slowly inactivating sodium current by is indicated schematically by the inset. The top and middle traces show currents periods. The bottom trace shows the control and prazosin traces superimposed.
Alpha
Effects on TI Current 8mM
Ca.
2mMK
Control V step
O1 ‘\:, i
(mV)
Prazosin -20
-10 1
o-
(5 x 16’~) -20
Oi
j \
197
I’! / / j
-10
;
I\
‘+---+ 6
‘1 Phentoiamine
(5 x 10v6 M)
Phentolamine
r
‘T + Prazosin
FIGURE 6. Effects of prazosin alone or with phentolamine on Z,i induced by 8 mM Ca and 2 rnM K. Numbers beside outward (upward) current excursions indicate the voltage of the respective activation steps ( VSteP). VH, hoiding potential. Top panels shdw that inhibition ofZ,i by prazosin occurs in the absence of acetylstrophanthdin. Bottom panels show that phentolamine does not prevent inhibition of Z,i by prazosin.
but in the continued presence of phentolamine. Prazosin again exerted its inhibitory effect (Fig. 6, bottom Gght) thereby indicating that it is unlikely that prazosin’s inhibitory effect on the 1,i in the absence of agonist is mediated by partial a-adrenergic agonist activity. This observation was confirmed in three experiments in which the I,i was induced by calcium and three experiments in which the 1ti was induced by acetylstrophanthidin.
Effects of phenylephrine
on
I,i induced by high Ca
Observations presented so far indicate that both phenylephrine and prazosin exert inhibitory effects on the 1ti. However, observations that a-adrenergic agonists promote induction of oscillatory afterpotentials were made in experiments in which afterpotentials were induced by elevated calcium concentration (Kimura et al., 1984, 1987). We therefore also determined the effects of phenylephrine in preparations in which the Iti was induced by 8 mM Ca and 2 mM K. Figure 7 shows records of transmembrane current from a representative experiment. The voltage steps used to elicit the 1i, were identical to those in
experiments with acetlystrophanthidin. The control record shows that high calcium also elicited a large Iti. However, in six of six preparations in which the I,i was induced by elevated Ca, phenylephrine caused a marked increase in the peak amplitudes of the 1ti (control: 16.7 + 2.4 nA + S.E.; phenylephrine: 29.8 + 2.6 nA + s.E.; P < 0.001, paired ttest). This stimulatory effect of phenylephrine dissipated when the drug was washed from the tissue bath. Effects of phenylephrine on the net inward current at the end of the activation step (Istep) were weak or absent. To determine whether the stimulatory effect of phenylephrine on Zti induced by high calcium concentration was mediated through activation of al-adrenoceptors, the effect of phenylephrine on peak Iti was assessed in five additional experiments in which tissues were pre-treated with prazosin (5 x 10m7 M). Under these conditions, phenylephrine no longer caused a significant change in peak 1ti (control: 13.3 f 2.4 nA & S.E.; phenylephrine: 15.1 + 3.4 nA & S.E.). Thus, both the stimulatory effect of phenylephrine on Iti induced by high calcium and the inhibit09 effect on Zti induced by acetylstrophanthidin were abolished by specific al-adrenergic blockade.
198
G. R. Ferrier _-
8mt.! Control
and Ca.
E. Carmeliet 2mM
PE (IO+hn)
K Wash
-I
5 nA
5s
FIGURE is indicated
7. Stimulatory effect ofphenylephrine by the schematic inset
on Zti induced
Discussion Dzyerential
effects of phenylephrine on Iti induced by acetlystrophanthidin and Ca
The present study suggests that earlier contradictory observations by others concerning the effect of a-adrenergic agonists on OAP, in fact, might both be correct. Alpha-adrenergic agonists may exert either inhibitory or stimulatory effects on the 1ri depending on the mechanism by which the current is induced. Thus, the stimulatory effect of phenylephrine on OAP induced by high calcium concentration reported by Kimura et al. (1984, 1987) is compatible with the stimulatory effect of phenylephrine on the Irr seen in the present study. Similarly, the lack of effect of phenylephrine on OAP induced by ouabain reported by Hewett and Rosen ( 1984) may represent the sum of the inhibitory effect of phenylephrine on the Iu plus an opposing stimulatory effect mediated by prolongation of the action potential duration. Thus, it becomes important to recognize that triggered activity secondary to the 1ti can have important and diverse characteristics depending on the conditions which induce the Iti. Although the present observations provide a possible explanation for earlier conflicting studies, they do not clearly predict the direction of the effect that might be exerted on OAP-induced arrhythmias in other settings
by 8 rnM Ca and 2 rnM K. The voltage
clamp
protocol
such as reperfusion. This is so because the chain of events involved in induction of OAP in reperfusion is not fully known and may not closely mimic either of the present test systems (Ferrier et al., 1985). Also, ischemia and reperfusion are known to modify the number of aadrenergic receptors at the cell surface (Corr et al., 1981; Maisel et al., 1987; Dillon et al., 1988), and to convert the response of idioventricular rate to stellate nerve stimulation from one that is propranolol sensitive to one that is phentolamine sensitive (Sheridan et al., 1980). This alteration of response to nerve stimulation could represent either a change in the mechanism of the idioventricular activity or modification of the coupling of aadrenoceptors to cellular or membrane functions. The present study underscores the importance of pre-treating preparations with blocking agents in order to identify the specific receptor invoIved in drug or transmitter actions. In experiments on Irr or OAP in which the agonist is introduced first and the blocker is added later, the inhibitory effect of prazosin on the 1tr could easily be misinterpreted as reversal of a stimulatory effect of phenylephrine mediated by oll-adrenoceptor blockade. This error could occur even in settings in which the agonist might have no effect mediated by a-adrenoceptors. At the con-
Alpha
Effects
centration used in the present study, prazosin reduced the amplitude but did not abolish the Zti. Although the inhibitory effect of prazosin appears to be mediated by an action at a site other than the a-adrenoceptor, this agent would still occupy al-adrenoceptors and block stimulation by phenylephrine. Thus, it was still possible to test the effect of agonist after pre-treatment with prazosin. This permitted demonstration that both the inhibitory and stimulatory effects of phenylephrine on the 1ti were mediated by specific al-adrenoceptor activation. The presence of propranolol throughout, eliminated effects mediated by fiadrenoceptor activation. The mechanism by which opposite effects of phenylephrine on the magnitude of the Iti are elicited depending on the method of induction of the current is not clear. Phenylephrine is known to have several effects in cardiac tissue that would be expected to affect generation of this current. This agent can induce calcium dependent slow action potentials in rabbit papillary muscle depolarized by elevated levels of potassium (Miura et al., 1987). Further, Bruckner and Scholz (1984) reported that phenylephrine increased the magnitude of the slow inward calcium current in bovine trabeculae in the presence of fi-adrenergic blockade with propranolol. An increase in calcium influx would be expected to promote induction of the 1ti and OAP. On the other hand, Shah et al. (1988) have reported that aadrenergic agonists decrease the intracellular sodium concentration possibly by stimulating the sodium pump. One would predict that this effect would cause a decrease in intracellular calcium concentration and thereby inhibit the induction of the 1tr or OAP. The relative importance of these two actions might differ with the conditions that induce the Iti. In the presence of an elevated calcium gradient, an increase in the open time or number of slow channels would be expected to greatly accentuate calcium influx by this route. The effect of altering the intracellular concentration of sodium might be of lesser importance in this setting, and a net stimulatory effect may result. However, when the sodium pump is partially inhibited by digitalis, the effects of phenylephrine on intracellular sodium concentration may predominate and result in a net inhibitory effect on the 1rr.
on TI Current
199
An alternate explanation for the opposing effects of phenylephrine on the Iti could be proposed on the basis of evidence suggesting that the 1rr may be derived from two different current carrying components: a Na/Ca exchange current and a non-selective cation current (for a discussion see Ehara et al., 1988). The relative importance of these two components would depend on the experimental condition. In high extracellular Ca, the contribution of the inward Na/Ca exchange current may be greater (due to low intracellular Na) than that of the nonspecific cation current. In digitalis intoxication on the other hand the nonspecific cation current may be favoured. If one hypothesizes that PE stimulates Na/Ca exchange but decreases the contribution of a non-selective cation current, the Iti and OAP would be enhanced in high Ca, but decreased in digitalis intoxication.
Inhibitory
efects of prazosin on I,i induced by acetylstrophanthidin or Ca
The present study also demonstrates that prazosin has an inhibitory effect on Zu which is independent of its a-adrenergic antagonist activity. Clearly studies identifying a-adrenergic mechanisms in the generation of arrhythmia, on the basis of effects of u-antagonists added after pretreatment with agonist, must be interpreted with great caution because inhibitory effects unrelated to a-adrenergic blockade may have been mistaken for specific aadrenergic antagonism. The mechanism of the inhibitory action of prazosin on the 1ti is not known. This effect was not abolished by a lo-fold higher con centration of phentolamine. Also, the effect of prazosin did not become stimulatory, like that of phenylephrine, when the Irr was induced by high Ca instead of acetylstrophanthidin. Therefore, it is unlikely that the effect is mediated via partial agonist activity at the CIadrenoceptor. We could find no evidence of sodium channel blocking action at a concentration that strongly inhibited the Zri, and the inhibitory effect was exerted on the Iti when it was induced by either acetylstrophanthidin or high calcium concentration. Thus, neither local anaesthetic action nor displacement of digitalis-like agents are tenable ex-
200
G. R. Ferrier
planations. Prazosin therefore must inhibit the Iri charge carrying mechanism or act at some link between calcium overload and activation of the current. These observations also suggest that prazosin might have antiarrhythmic efficacy even in settings in which sympathetic activity is not a prominent factor in induction of the arrhythmia.
and E. Carmeliet Acknowledgements The authors thank J. Prenen for excellent technical assistance and X. Han for participation in some of the experiments. This study was supported in part by the Medical Research Council of Canada, and the Visiting Scientist Program of the Canadian Heart Foundation.
References ARONSON RS, GELLES JM,
HOFFMAN BF (1973) A new method for producing short cardiac Purkinje fibers suitable for clamp. J Appl Physiol34: 527-530. BRUCKNER R, SCHOLZ H (1984) Effects ofa-adrenoceptor stimulation with phenylephrine in the presence of propranolol on force of contraction, slow inward current and cyclic AMP content in the bovine heart. Br J Pharmacol82: 223-232. CARMELIET C (1987) Slow inactivation of the sodium current in rabbit cardiac Purkinje fibres. Pfliigers Arch 408: 18-26. CORR PB, SHAYMAN JA, KRAMER JB (1981) I ncreased a-adrenergic receptors in ischemic cat myocardium. A potential mediator for electrophysiological derangements. J Clin Invest 67: 1232-1236. DILLON JS, Gu XH, NAYLER WG (1988) Alpha, adrenoceptors in the ischaemic and reperfused myocardium. J Mol Cell Cardiol 20: 725-735. DUKES ID, VAUGHAN WILLIAMS EM (1984) Electrophysiological effects of cl-adrenoceptor antagonists in rabbit sinoatria1 node, cardiac Purkinje cells and papillary muscles. Br J Pharmacol83: 419-426. EHARA T, NOMA A, ONO K (1988) Calcium-activated non-selective cation channel in ventricular cells isolated from adult guinea pig heart. J Physiol, Lond 403: 117-133. FERRIER GR, MOFFAT MP, LUKAS A (1985) P ossible mechanisms of ventricular arrhythmias elicited by &hernia followed by reperfusion. Studies on isolated canine ventricular tissues. Circ Res 56: 184-194. GIOTTI A, LEDDA F, MANNAIONI PF (1973) Effects of noradrenaline and isoprenaline, in combination with a- and breceptor blocking substances, on the action potential of cardiac purkinje fibres. J Physiol, Lond 229: 99-l 13. HEWETT KW, ROSEN MR (1984) Alpha and beta adrenergic interactions with ouabain-induced delayed afterdepolarizations. J Pharmacol Exp Ther 229: 188-192. KASS RS, LEDERER WJ, TSIEN W, WEINGART R (1978a) Role of calcium ions in transient inward currents and aftercontractions induced by strophanthidin in cardiac Purkinje fibres. J Physiol, Lond 281: 187-208. KA~.s RS, TSIEN W, WEINGART, R (1978b) I onic basis of transient inward current induced by strophanthidin in cardiac Purkinje fibres. J Physiol, Lond 281: 209-226. KIMURA S, CAMERON, JS, KOZLOVSKIS PL, BASSETT, AL, MYERBURG, RJ (1984) Delayed afterdepolarizations and triggered activity induced in feline Purkinje fibers by cc-adrenergic stimulation in the presence ofelevated calcium levels. Circulation 70: 1074-1082. KIMURA S, BASSETT AL, KOHYA T, KOZLOVS~IS PL, MYERBURG RJ (1987) Automaticity, triggered activity, and responses to adrenergic stimulation in cat subendocardial Purkinje fibers after healing of myocardial infarction. Circulation 75: 651-660. MAISEL AS, MOTULSKY HJ, ZIEGLER G, INSEL PA (1987) Ischemiaand agonist-induced changes in cc- and /l-adrenergic traffic in guinea pig hearts. Am J Physiol 253: H 1159-H 1166. MIURA Y, INUI J, IMAMURA H (1978) Alpha-adrenoceptor-mediated restoration of calcium-dependent potential in the partially depolarized rabbit papillary muscle. Naunyn-Schmiedeberg’s Arch Pharmacol302: 201-205. MUBAGWA K, CARMELIET E (1983) Effects of acetylcholine on electrophysiological properties of rabbit cardiac Purkinje fibers. Circ Res 53: 740-75 1. NORTHOVER BJ (1983) A comparison of the electrophysiological actions of phentolamine with those of some other antiarrhythmic drugs on tissues isolated from the rat heart. Br J Pharmacol 80: 85-93. PENNY WJ, CULLING W, LEWIS MJ, SHERIDAN DJ (1985) Antiarrhythmic and electrophysiological effects of aIpha adrenoceptor blockade during myocardial ischaemia and reperfusion in isolated guinea-pig heart. J Mel Cell Cardiol voltage
17: 399409. POGWIZD SM,
CORR PB (1987) Electrophysiologic mechanisms underlying arrhythmias due to reperfusion of ischemic myocardium. Circulation 76: 404426. ROSEN MR, DANILO P (1980) Effects of tetrodotoxin, lidocaine, verapamil and AHR-2666 on ouabain-induced delayed afterdepolarizations in canine Purkinje fibers. Circ Res 46: 117-124. SHAH A, COHEN IS, ROSEN MR (1988) Stimulation of cardiac alpha receptors increases Na/K pump current and decreases gs via a pertussis toxin-sensitive pathway. Biophys J 54: 219-225. SHERIDAN DJ, PENKOSKE PA, SOBEL BE, CORR PB (1980) Alpha adrenergic contributions to dysrhythmia during myocardial ischemia and reperfusion in cats. J Clin Invest 65: 16 1-l 7 1. WASSERSTROM JA, FERRIER GR (1982) Effects of phenytoin and quinidine on digitalis-induced oscillatory afterpotentials, aftercontractions, and inotropy in canine ventricular tissues. J Mol Cell Cardiol 14: 725-736.