- Email: [email protected]
Positive and negative chronotropic effects of caffeine in spontaneously beating rabbit sino-atrial node cells
Gen. Pharmac. Vol. 24, No. 5, pp. 1223-1230, 1993 Printed in Great Britain. All rights reserved
0306-3623/93 $6.00 + 0.00 Copyright © 1993 Pergamon Press Ltd
POSITIVE A N D NEGATIVE CHRONOTROPIC EFFECTS OF CAFFEINE IN SPONTANEOUSLY BEATING RABBIT SINO-ATRIAL NODE CELLS HIROYASU SATOH Department of Pharmacology, Nara Medical University, Kashihara, Nara 634, Japan [Tel. (07442) 2-3051', Fax (07442) 5-7657] (Received 28 January 1993) Abstraet--l. Effects of caffeine on the automaticity in spontaneously beating rabbit sino-atrial node cells were examined. 2. Caffeine (0.5 2 mM) caused only a positive chronotropic effect. At over 5 mM, caffeine caused an initial positive and subsequently a negative chronotropic effect. Both were not modified by pindolol (1 # M) and atropine (1 #M). 3. At 10 mM, a dysrhythmia occurred in all 9 preparations. The effects were reversible, and dysrhythmia also disappeared after washout. 4. When extracellular K + was increased from 2.7 to 8 mM, the dysrhythmia induced by caffeine (10 raM) was abolished, and the sinus rate was increased. 5. Addition of tetrodotoxin (TTX) (10 -7 M) in the presence of caffeine (10 mM) also abolished the dysrhythmia and increased the sinus rate. 6. In addition, a decrease in extracellular Ca2+([Ca]o) to 0.5mM abolished the dysrhythmia and increased the sinus rate, whereas increasing [Ca]o to 7.4 mM potentiated the negative chronotropic effect and failed to inhibit the dysrhythmia. 7. These results indicate that the positive chronotropic effect is due to the stimulatory effect of caffeine, and the negative effect is directly and indirectly due to development of cellular calcium overload.
INTRODUCTION
Caffeine possesses multiple actions; (a) release of Ca 2+ from sarcoplasmic reticulum (SR) (Weber and Herz, 1968; Fuchs, 1969), (b) inhibition of re-uptake to SR (Thorpe, 1973; Fabiato and Fabiato, 1975) and (c) caffeine is a phosphodiesterase inhibitor, and thereby accumulates the cellular cyclic A M P level, resulting in enhancement of Ca 2+ current (lc~) and positive inotropic effect (Satoh and Vassaile, 1985; Satoh, 1993). But in canine Purkinje fibers, caffeine causes not only a positive inotropic effect, but also a subsequent negative inotropic effect due to excessive elevation of cellular Ca 2+ concentration ([Ca]i) (Satoh and Vassalle, 1985, 1989). Many electrophysiological effects of caffeine on ionic currents are still controversial. Caffeine enhances/ca in frog atria (Goto et al., 1979) and in rat cardiomyocytes (Yatani et al., 1984), whereas caffeine inhibits /ca in Purkinje fibers (Eisner and Lederer, 1979; Hasegawa and Vassalle, 1986). Furthermore, caffeine increases the time-dependent outward K + current (I K) (Scholtz and Reuter, 1976), but does not produce any appreciable changes in I K (Yatani et al., 1984). Thus, it is not clear about the caffeine effects on Ic~ and IK in cardiac muscle cells. The difference might be dependent on [Ca]~ level before caffeine application (Bourinet et al., 1992; Satoh and Hashimoto, 1991; Satoh and Tsuchida, oP 24~-~
1993; Satoh et al., 1989). If caffeine induces the cellular calcium overload in rabbit sino-atrial (SA) node cells, like in Purkinje fibers (Satoh and Vassalle, 1985, 1989), the ionic currents of the pacemaker cells may also be modulated by caffeine-induced elevation of [Ca]i. Satoh (1993) has recently shown that caffeine enhanced/ca and inhibited I K in rabbit SA node cells. The aim of the present experiments is to investigate the modulation of pacemaker activity by application of caffeine. Especially, the mechanism underlying the subsequent negative chronotropic effect induced by caffeine is discussed. METHODS
Preparations and recording of action potentials Rabbits, weighing 1.5-2.0 kg, were killed by a blow to the neck and exsanguinated. After removing the right atrium, strips of the SA node tissue were made by dissecting the tissue in a direction perpendicular to the crista terminalis. The specimens were made smaller by dissections to a final dimension of about 0.25 x 0.25 mm for comparison with the changes in the ionic currents by caffeine (Satoh, 1993). The specimens were superfused with a bathing solution oxygenated by 100% O: at 36°C and left spontaneously beating. Conventional glass microelectrodes filled with 3 M KCI were used and the resistances were 20-30 MQ. Solution The composition of the bating solution were (in mM): NaC1137, KC12.7, MgCI.,0.5, CaCI_, 1.8, and HEPES 5.0. The pH was adjusted to 7.4 with NaOH. The drugs used were caffeine (Sigma Chemical Co., St Louis, MO, U.S.A.),
1223
1224
HIROYASUSATOH
CAFFEINE /rnin
220 F _
2 min
0.5S /rain J 200F~_ 160 L
1 mM
I
"
a
[
2 mM
I i~, ~
- V V J V v' v ,
I by
•
IT
I i/ ," I!
0 -60 VJ
b
10 /min
I
2O0
i
I
r
i
J
_
Fig. 1. Positive chronotropic effect induced by caffeine in rabbit SA node cells. Upper panel: sinus rate is shown at 0.5 and 1 mM caffeine. Lower panel: at 2 mM caffeine, the spontaneous action potentials, the maximum rate of depolarization (l/ma0, and the sinus rate are represented. Triangles (a and b) on the action potential recording indicate the action potentials and the Vmax at slow speed on the right panel. pindolol (Sandoz, Basel, Switzerland), atropine sulfate (Wako Pure Chemical Ind. Osaka, Japan), and tetrodotoxin (TTX) (Sankyo Pharmaceutical Co. Tokyo, Japan).
Statistical analysis Results are presented as mean + SEM. The significance of differences was assessed with Student's t-test for paired data and a level of probability P < 0.05 was considered significant.
RESULTS
Concentration-dependent effect of caffeine The
chronotropic
taneously
beating
effects rabbit
of
caffeine
SA
node
in
spon-
cells
were
examined. Caffeine at 0.5 to 2mM caused only a positive chronotropic effect (Fig. 1). At 5 and 10mM, caffeine caused a transient positive chronotropic effect and then a profound negative chronotropic effect (Figs 2 and 3). The percentage changes in the chronotropic effects are summarized in Table 1. At 5 mM, caffeine caused a positive chronotropic effect, and suddenly a dysrhythmia occurred during the subsequent negative effects, as shown in Fig. 2. The incidence of dysrhythmia is also given in Table 1. Similarly, caffeine (10mM) initially caused a positive effect, and secondly a negative effect (Fig. 3). During the negative effect,
Table 1. Positive and negative chronotropic effects in the presence of caffeine, and incidence of dysrhythmia Chronotropic effect Positive
Negative
Caffeine 0.5 mM
6
6.3 + 2.9
1 mM
6
10.3 + 2 . 2 * *
6 8 9
20.7 + 3.8** 25.3 -t- 3.4** 17.5 + 2.8**
7.3 + 3.1 17.3 + 2.8** 44.6 + 4.0***
5
18.8 + 3.4
49.0_+ 1.2
4
33.5 + 3.5
48.5 + 2.8
2 mM 5 mM I0 mM Pindolol 10 7M + caffeine 10mM Atropine 10 6 M + caffeine 10 mM
Incidence of dysrhythmia
---
Values (%) represent mean _+SEM, as compared with control value. *P < 0.05, **P < 0.01, ***P < 0.001, with respect to control values.
Effect of caffeine on SA node cells
CAFFEINE
1225
5 mM 0.5s
2 rain !
!
!
i
13'1
0
Vt
.2
10
/min _._./----
250 f
e----
150I
oll Fig. 2. Occurrence of dysrhythmia by caffeine in spontaneously beating rabbit SA node cells. Spontaneous action potentials, maximum rate of depolarization (Vmax),and sinus rate are shown at slow and fast speeds. dysrhythmia was elicited. The positive effect at concentrations ranging from 0.5 to 5 mM was concentration-dependent. At 10 mM caffeine enhanced the sinus rate (17.5 + 2.8%, n = 9 , P < 0.01), but the value decreased as compared to the value at 5 mM caffeine. The positive and negative effects were reversible, and the dysrhythmia disappeared after washout. During the washout of caffeine (10mM), the sinus rate had a transient enhancement (rebound) by 7.9 + 0.8% (n = 9, P > 0.05) and then, was gradually recovered to the control value (Fig. 3). In the presence of pindolol (1/~M), caffeine (10mM) caused positive (by 18.8_3.4%, n = 5 ,
I
P > 0 . 0 5 ) and negative (by 49.0+ 1.2%, n = 5 , P < 0 . 0 5 ) chronotropic effects (Table 1). Also, caffeine (10 mM) in the presence of atropine (1 pM) produced the positive and negative chronotropic effects by 33.5+3.5% (n = 4 , P <0.05) and by 48.5 + 2.8% (n =4, P >0.05), respectively. Both positive and negative chronotropic effects were not mediated through fl-adrenoceptor and muscarinic receptor stimulations. Effect o f high K +
The dysrhythmia occurred during the negative chronotropic effect following the positive effect in
CAFFEINE 10mM
l
2mip
tt , !,
mY
OF~,' 1
~
-6o L / ~ Ws 10[ /rain 250 I
100
Om5 s
._---
i'
Fig. 3. Dysrhythmia induced by high concentrations of caffeine. Application of caffeine (10 mM) caused a transient positive chronotropic effect and then, elicited dysrhythmia. After washout, regular rhythm was resumed, and the sinus rate was recovered to control value followed by a small rebound.
,,
,
i
1226
HIROYASU SATOH ,
CAFFEINE 10 mM b
a
m
I
KSmM
c
d
a
,,
,
!,, ii
¸
0
-60
b
W:
S!!,!
:] l,,
10 /min
200~ ! 2 min 100
":
:. . . . . ,
0
,r,r . . . . . . . . . ~.,
I
~'4,
d
Fig. 4. Modulation by high K + solution of negative chronotropic effect and dysrhythmia induced by caffeine. Caffeine (10 raM) induced dysrhythmia followed by a transient increase in the sinus rate. Addition of high K + abolished the dysrhythmia, but did not return the sinus rate to the control value. Triangles (a~d) on the top of the action potential recording indicate the action potentials at slow speed on the right panel. the presence of caffeine (10 m M ) (Fig. 4). Increasing extracellular K + concentration from 2.7 to 8 m M during the negative chronotropic effect depressed the action potential amplitude and the maximum rate of depolarization (I?max), and abolished the dysrhythmia in all 6 preparations. High K + solution reduces the [Ca]i level (Ellis, 1977). The depressed sinus rate was increased by 65.3 _+ 6.4% (n = 6, P < 0.001) by taking the value of the negative chronotropic effect at 1 0 m M as control (but not recovered to the control value) (Table 2). The rebound of recovery was 20.1__+ 3.5% (n = 5, P < 0.05). Effect of TTX
Addition of T T X (10-TM) also abolished the dysrhythmia and regular rhythm was resumed, as shown in Fig. 5. The sinus rate was increased by Table 2. Modulation of negative chronotropic effect induced by caffeine and incidence of sinus arrest Chronotropic Incidenceof n effect sinus arrest K + 8mM 6 +65.3 _+6.4*** 0 TTX 0.1 ,uM 5 +28.0_+5.3** 0 Ca2+ 0.5 mM 5 + 15.3 _+5.5* 1 Ca2+ 7.4mM 6 -52.1 _+1.2"** 3 Values (%) represent mean +_ SEM. Caffeine (10 mM) caused the negative chronotropic effect by 43.7 _+ 3.8% (n = 22, P < 0.001) following the positive chronotropic effect (by 20.5_+3.9%, n = 2 2 , P <0.05). *P <0.05, **P <0.01, ***P <0.001, with respect to the values of negative chronotropic effect at l0 mM caffeine.
28.0 _+ 5.3% (n = 5, P < 0.05) (Table 2). When the bath solution was changed to normal solution, the rebound of sinus rate was produced (by 25.3 +_ 2.9%, n = 5, P < 0.01). And then, the sinus rate was gradually recovered to control. E f f e c t s o f e x t r a c e l l u l a r l o w a n d high Ca 2+ solutions
Decrease in extracellular Ca 2+ concentration to 0.5 m M (from 1.8 m M in normal solution) abolished the dysrhythmia, and increased the sinus rate (Fig. 6). The increase in sinus rate was 15.3 _+ 5.5% (n = 5, P < 0.05), but subsequently a sinus arrest occurred in I of 5 preparations (Table 2). The rebound during washout was 3 1 . 8 + 2 . 0 % ( n = 5 , P<0.01). In contrast, increasing Ca 2+ concentration to 7.4 m M potentiated the negative chronotropic effect induced by caffeine (10 raM) by 52.1 _+ 1.2% (n = 6, P < 0.001) (Fig. 7). The high Ca 2+ not only failed to inhibit the dysrhythmia, but also increased further the incidence of sinus arrest (in 3 of 6 preparations) (Table 2). Similarly, the rebound of sinus rate during washout was induced by 22.3 _+ 4.2% (n = 4, P < 0.05), accompanied with dysrhythmia. DISCUSSION
It has already been shown that caffeine caused biphasic inotropic actions in canine Purkinje fibers (Satoh and Vassalle, 1985, 1989). The positive response was due to Ic~ enhancement and Ca 2+
Effect of caffeine on SA node cells
I TTX 10 -7 M I CAFFEINE 10 mM b
a
2 min
c
m
mV
o
a
-60
1227
II/l !l
,:i i /'d :/d d 'd "d ,,,
0.SS i
il
10
\
!,
t ~T--~ V T 'T 'T --~ ,/'.'
/-C
Fig. 5. TTX inhibition in the dysrhythmia induced by caffeine. Caffeine (10 mM) elicited dysrhythmia followed by positive chronotropic effect. Addition of TTX (10 -7 M) abolished the dysrhythmia, but the sinus rate was not recovered to the control value.
release from the stores. The subsequent negative response was presumably due to [Ca]i elevation (cellular calcium overload) by Ca 2÷ re-uptake inhibition into SR (Thorpe, 1973; Fabiato and Fabiato, 1975), as well as by /Ca enhancement and Ca 2÷ release. The findings in the present study are as follows: (a) caffeine (0.5 to 1 mM) caused only a positive
I
CAFFEINE 10 mM
I
chronotropic effect, and (b) at over 5 mM, caffeine produced an initial positive and subsequent negative chronotropic effect. (c) The positive chronotropic effect was not modified by pindolol. (d) The negative chronotropic effect (unaffected by atropine) was exaggerated by high Ca :+ but was inhibited by high K +, T T X and low Ca 2÷.
Ca 0.5 mM
2 rain
mV
V/ 10 /rain 20O
S
100
r
Fig. 6. Effect of low Ca 2+ on the dysrhythmia-occurring SA node cells. Dysrhythmia induced by caffeine (10 mM) was abolished by addition of low Ca 2÷ (0.5 mM). Decrease in Ca 2÷ concentration caused a positive chronotropic effect under the calcium overload condition.
1228
HIROYASUSATOH
I
CAFFEINE
I Ca 7.4mM I
?mip
10 m M
0 -60
10 /min
50
...... . .
.
~
"
Fig. 7. Effect of high Ca -~+ on the dysrhythmia-occurring SA node cells. Increase in Ca 2+ concentration (to 7.4 mM) decreased further the depressed sinus rate by caffeine (10 mM).
Positive chronotropic effect The positive chronotropic effect was produced at low concentrations of caffeine (0.5 to 1 mM), and at higher concentrations, as initial response after application of caffeine. The positive effect was concentration-dependent (at 0.5 to 5 mM), but caffeine at 10 mM produced the positive effect less than at 5 mM (although the value was still enhanced as compared with control value). The positive response was not modified by pindolol, but was rather potentiated by atropine (probably due to loss of muscarinic receptor stimulation). Therefore, the positive chronotropic effect induced by caffeine is not mediated through fl-adrenoceptor stimulation by endogenous catecholamines, but would be due to /ca enhancement and Ca 2+ release from SR (Weber and Herz, 1968; Fuchs, 1969). There are three hypotheses for the pacemaker current in SA node cells (Noble, 1984); (a) the /ca hypothesis, (b) the I K decay hypothesis, and (c) the hyperpolarization-activated inward current (Ih) hypothesis. In voltage-clamped SA node cells, caffeine enhanced /Ca and lh, and inhibited IK (Satoh, 1993). Thus, the initial positive chronotropic effect would be due to the enhancements of/ca and I h. Surprisingly, these enhancements were smaller in the presence of 10 rather than 5 mM caffeine, consistent with the positive chronotropic effect of caffeine at high concentration in the present experiments. This inhibition would be secondly produced by depressant effect due to cellular calcium overload. Because Ic, amplitude is
strongly inhibited under the excessive [Ca]~ condition (Satoh et al., 1989). Recently, it has been reported that the T-type Ca 2+ current makes a major contribution to the spontaneous activity of SA node cells (Hagiwara et al., 1988). If so, caffeine might directly stimulate the T-type Ca 2+ channels. To elucidate this, the extended experiments require using single SA nodal cells. The secondary positive chronotropic effect was the rebound of recovery during washout. A similar recovery rebound has already been shown in the inotropic effect in canine Purkinje fibers (Satoh and Vassalle, 1985). Once caffeine was discontinued, Ca 2+ re-uptake into SR would be stimulated, resulting in the re-start of exaggerated and compensatory Ca 2+ release. In addition, the rebound recovery of [Ca]~ levels has actually been observed in fura-2 loaded single SA node cells (Satoh, 1993).
Negative chronotropic effect The negative chronotropic effect was subsequently produced by application of caffeine at 2 m M or higher in a concentration-dependent manner, which was unaffected by atropine. In the presence of both atropine and pindolol, caffeine (10 mM) potentiated the decrease in sinus rate, but not significantly. The potentiation by pindolol would be due to loss of fl-adrenoceptor stimulation, but the potentiation by atropine is not clear. If muscarinic receptor stimulation is blocked by atropine, the decline in sinus rate by caffeine should be reduced in the presence of
Effect of caffeine on SA node cells atropine. These results suggest that caffeine may have a direct negative chronotropic action. Noble (1984) suggested that Ca 2÷ current plays an important role only during the last 30% of diastolic potential, generating spontaneous action potentials in SA nodal cells. Our previous studies have also demonstrated similar results (Satoh and Hashimoto, 1991; Satoh and Tsuchida, 1993). In voltage-clamped rabbit SA node cells, caffeine enhanced/ca and/h, and inhibited I K (Satoh, 1993). Thus, the reduction in I K may cause the negative chronotropic effect in this study. To prolong the cycle length in the presence of caffeine, since caffeine enhanced /ca, the maximum diastolic potential (MDP) must be hyperpolarized (although if the threshold potential is not shifted). However, the MDP was not significantly hyperpolarized by caffeine (Satoh, 1993). Thus, it seems unlikely that the pacemaker activity is caused by only one ionic current, which is also a result obtained from our studies on the modulation of pacemaker activity (Satoh et al., 1989; Satoh and Hashimoto, 1991; Satoh, 1991; Satoh and Tsuchida, 1993). The mechanism of negative effect was examined by procedures to decline [Ca]i (TTX, high K ÷ and low Ca 2÷) and to elevate [Ca]i(high Ca 2÷) (Satoh and Vassalle, 1985). High K ÷ decreases intracellular Na ÷ activity (aiNa) in quiescent and active Purkinje fibers (Ellis, 1977; Lee and Vassalle, 1983), and therefore presumably decreases [Ca]i through an enhanced Na42a exchange (Mullins, 1979). In addition, high K ÷ depolarizes the resting potential, which could indirectly inhibit the slow inward current and thus contribute to the decrease in [Ca]i. Similarly, TTX decreases aiNa in quiescent and active Purkinje fibers (Vassalle and Lee, 1984; Deitmer and Ellis, 1980) and decreased contractile force markedly (Satoh and Vassalle, 1985). Low extracellular Ca 2+ also decreases [Ca]i and /ca decreases (Reuter, 1967). In the present experiments, therefore, high K ÷, TTX and low Ca 2+ share a similar action on [Ca]i and on the chronotropic effect. On the other hand, high Ca 2÷ enhances /ca and elevates [Ca]i level. In this study, the further negative chronotropic effect of caffeine at high Ca 2÷ was produced. These results suggest that caffeine initially caused the stimulatory action, and then, the negative chronotropic effect might be induced by development of excessive Ca 2÷ load in cytosol. The caffeine-induced cellular calcium overload has already been demonstrated in canine and sheep Purkinje fibers (Satoh and Vassalle, 1985, 1989; Hasegawa et al., 1987) and in rabbit SA node cells (Satoh, 1993). Caffeine elicited a transient inward current (/Ti) and an inward tail current (lex), resulting in induction of arrhythmias. During calcium over-
1229
load (the negative chronotropic effect), caffeine (5 to 10 mM) still enhanced/ca and lb. Actually, the [Ca]i level in rabbit SA node cells was enhanced using fura-2 Ca 2+ fluorescent dye. Also, the rebound recovery of chronotropic effect during washout reflected the excessive Ca 2+ load induced by caffeine. Therefore, these results indicate that the negative chronotropic effect is due to the development of cellular calcium overload, accompanied by the depressions in ionic currents.
REFERENCES
Bourinet E., Fournier F., Lory P., Charnet P. and Nargeot J. (1992) Protein kinase C regulation of cardiac calcium channels expressed in Xenopus ooeytes. Pfliigers Archs 421, 247-255. Deitmer J. W. and Ellis D. (1980) The intracellular sodium activity of sheep heart Purkinje fibers: Effects of local anaesthetics and tetrodotoxin. J. PhysioL (Lond.) 300, 269-282. Eisner D. A. and Lederer W. J. (1979) Inotropic and arrhythmogenic effects of potassium-depleted solutions on mammalian cardiac muscle. J. Physiol. (Lond.) 294, 255-277. Ellis D. (1977) The effects of external cations and ouabain on the intracellular sodium activity of sheep heart Purkinje fibers. J. Physiol. (Lond.) 273, 211-240. Fabiato A. and Fabiato F. (1975) Contraction induced by a calcium-triggered release of calcium from the sarcoplasmic reticulum of single skinned cardiac cells. J. Physiol. 249, 469-495. Fuchs F. (1969) Inhibition of sarcotubular calcium transport by caffeine: Species and temperature dependence. Biochim. Biophys. Acta. 172, 566-570. Goto M., Yatani A. and Ehara E. (1979) Interaction between caffeine and adenosine on the membrane current and tension component in the bull frog atrial muscle. Jap. J. Physiol. 28, 393-409. Hagiwara N., Irisawa H. and Kameyama M. (1988) Contribution of two types of calcium currents to the pacemaker potentials of rabbit sino-atrial node cells. J. Physiol. (Lond.) 395, 233-253. Hasegawa J. and Vassalle M. (1986) The effect of caffeine on the inward current in sheep cardiac Purkinje fibers. Fed: Proc. 45, 288. Hasegawa J., Satoh H. and Vassalle M. (1987) Induction of the oscillatory current by low concentration of caffeine in sheep cardiac Purkinje fibers. Naunyn-Schmiedeberg's Archs Pharmac. 335, 310-320. Lee K. S. and Vassalle M. (1983) Role of calcium in the inotropic effects of caffeine in cardiac Purkinje fibers. Int. J. Cardiol. 3, 421-434. Mullins L. J. (1979) The generation of electrical currents in cardiac fibers by Na/Ca exchange. Am. J. Physiol. 236, CI03--CI10. Noble D. (1984) The surprising heart: A review of recent progress in cardiac electrophysiology.J. Physiol. (Lond.) 353, 1-50. Reuter H. (1967) The dependence of slow inward current in Purkinje fibers on the extracellular calciumconcentration. J. Physiol. (Lond.) 192, 479-492. Satoh H. (1991) Pharmacology and therapeutic effects of mepirodipine. Cardiovasc. Drug Rev. 9, 340-356. Satoh H. (1993) Caffeinedepression in spontaneous activity in rabbit sino-atrial node cells. Gen. Pharmac. 24, 555-563. Satoh H. and Hashimoto K. (1991) Comparative effects of a new calcium channel antagonist, mepirodipine,
1230
H1ROYASUSATOH
on rabbit spontaneously beating sino-atrial node cells. Eur. J. Pharmac. 193, 9 13. Satoh H. and Tsuchida K. (1993) Comparison of a calcium antagonist, CD-349, with nifedipine and verapamil in rabbit spontaneously beating sino-atrial node cells. J. Cardiovasc. Pharmac. 21,685-692. Satoh H. and Vassalle M. (1985) Reversal of caffeineinduced calcium overload in cardiac Purkinje fibers. J. Pharmac. Exp. Ther. 234, 172-179. Satoh H. and Vassalle M. (1989) Role of calcium in caffeine-norepinephrine interactions in cardiac Purkinje fibers. Am. J. Physiol. 257, H226-H237. Satoh H., Tsuchida K. and Hashimoto K. (1989) Electrophysiological actions of A23187 and X-537A in rabbit sino-atrial node cells. Naunyn-Schmiedeberg's Archs Pharmac. 339, 320-326.
Scholtz H. and Reuter H. (1976) Effect of theophylline on membrane currents in mammalian cardiac muscle. Naunyn-Schmiedeberg's Archs Pharmac. 293, RI9. Thorpe W. R. (1973) Some effects of caffeine and quinidine on sarcoplasmic reticulum of skeletal and cardiac muscle. Can. J. Physiol. Pharmac. 51,499-503. Vassalle M. and Lee C. O. (1984) The relationship among intracellular sodium activity, calcium, and strophanthidin inotropy in canine cardiac Purkinje fibers. J. Gen. Physiol. 83, 287-307. Weber A. and Herz R. (1968) The relationship between caffeine contracture of intact muscle and the effect of caffeine on reticulum. J. Gen. Physiol. 52, 750-759. Yatani A. Imoto Y. and Goto M. (1984) The effects of caffeine on the electrical properties of isolated, single rat ventricular cells. Jap. J. Physiol. 34, 337-349.
0306-3623/93 $6.00 + 0.00 Copyright © 1993 Pergamon Press Ltd
POSITIVE A N D NEGATIVE CHRONOTROPIC EFFECTS OF CAFFEINE IN SPONTANEOUSLY BEATING RABBIT SINO-ATRIAL NODE CELLS HIROYASU SATOH Department of Pharmacology, Nara Medical University, Kashihara, Nara 634, Japan [Tel. (07442) 2-3051', Fax (07442) 5-7657] (Received 28 January 1993) Abstraet--l. Effects of caffeine on the automaticity in spontaneously beating rabbit sino-atrial node cells were examined. 2. Caffeine (0.5 2 mM) caused only a positive chronotropic effect. At over 5 mM, caffeine caused an initial positive and subsequently a negative chronotropic effect. Both were not modified by pindolol (1 # M) and atropine (1 #M). 3. At 10 mM, a dysrhythmia occurred in all 9 preparations. The effects were reversible, and dysrhythmia also disappeared after washout. 4. When extracellular K + was increased from 2.7 to 8 mM, the dysrhythmia induced by caffeine (10 raM) was abolished, and the sinus rate was increased. 5. Addition of tetrodotoxin (TTX) (10 -7 M) in the presence of caffeine (10 mM) also abolished the dysrhythmia and increased the sinus rate. 6. In addition, a decrease in extracellular Ca2+([Ca]o) to 0.5mM abolished the dysrhythmia and increased the sinus rate, whereas increasing [Ca]o to 7.4 mM potentiated the negative chronotropic effect and failed to inhibit the dysrhythmia. 7. These results indicate that the positive chronotropic effect is due to the stimulatory effect of caffeine, and the negative effect is directly and indirectly due to development of cellular calcium overload.
INTRODUCTION
Caffeine possesses multiple actions; (a) release of Ca 2+ from sarcoplasmic reticulum (SR) (Weber and Herz, 1968; Fuchs, 1969), (b) inhibition of re-uptake to SR (Thorpe, 1973; Fabiato and Fabiato, 1975) and (c) caffeine is a phosphodiesterase inhibitor, and thereby accumulates the cellular cyclic A M P level, resulting in enhancement of Ca 2+ current (lc~) and positive inotropic effect (Satoh and Vassaile, 1985; Satoh, 1993). But in canine Purkinje fibers, caffeine causes not only a positive inotropic effect, but also a subsequent negative inotropic effect due to excessive elevation of cellular Ca 2+ concentration ([Ca]i) (Satoh and Vassalle, 1985, 1989). Many electrophysiological effects of caffeine on ionic currents are still controversial. Caffeine enhances/ca in frog atria (Goto et al., 1979) and in rat cardiomyocytes (Yatani et al., 1984), whereas caffeine inhibits /ca in Purkinje fibers (Eisner and Lederer, 1979; Hasegawa and Vassalle, 1986). Furthermore, caffeine increases the time-dependent outward K + current (I K) (Scholtz and Reuter, 1976), but does not produce any appreciable changes in I K (Yatani et al., 1984). Thus, it is not clear about the caffeine effects on Ic~ and IK in cardiac muscle cells. The difference might be dependent on [Ca]~ level before caffeine application (Bourinet et al., 1992; Satoh and Hashimoto, 1991; Satoh and Tsuchida, oP 24~-~
1993; Satoh et al., 1989). If caffeine induces the cellular calcium overload in rabbit sino-atrial (SA) node cells, like in Purkinje fibers (Satoh and Vassalle, 1985, 1989), the ionic currents of the pacemaker cells may also be modulated by caffeine-induced elevation of [Ca]i. Satoh (1993) has recently shown that caffeine enhanced/ca and inhibited I K in rabbit SA node cells. The aim of the present experiments is to investigate the modulation of pacemaker activity by application of caffeine. Especially, the mechanism underlying the subsequent negative chronotropic effect induced by caffeine is discussed. METHODS
Preparations and recording of action potentials Rabbits, weighing 1.5-2.0 kg, were killed by a blow to the neck and exsanguinated. After removing the right atrium, strips of the SA node tissue were made by dissecting the tissue in a direction perpendicular to the crista terminalis. The specimens were made smaller by dissections to a final dimension of about 0.25 x 0.25 mm for comparison with the changes in the ionic currents by caffeine (Satoh, 1993). The specimens were superfused with a bathing solution oxygenated by 100% O: at 36°C and left spontaneously beating. Conventional glass microelectrodes filled with 3 M KCI were used and the resistances were 20-30 MQ. Solution The composition of the bating solution were (in mM): NaC1137, KC12.7, MgCI.,0.5, CaCI_, 1.8, and HEPES 5.0. The pH was adjusted to 7.4 with NaOH. The drugs used were caffeine (Sigma Chemical Co., St Louis, MO, U.S.A.),
1223
1224
HIROYASUSATOH
CAFFEINE /rnin
220 F _
2 min
0.5S /rain J 200F~_ 160 L
1 mM
I
"
a
[
2 mM
I i~, ~
- V V J V v' v ,
I by
•
IT
I i/ ," I!
0 -60 VJ
b
10 /min
I
2O0
i
I
r
i
J
_
Fig. 1. Positive chronotropic effect induced by caffeine in rabbit SA node cells. Upper panel: sinus rate is shown at 0.5 and 1 mM caffeine. Lower panel: at 2 mM caffeine, the spontaneous action potentials, the maximum rate of depolarization (l/ma0, and the sinus rate are represented. Triangles (a and b) on the action potential recording indicate the action potentials and the Vmax at slow speed on the right panel. pindolol (Sandoz, Basel, Switzerland), atropine sulfate (Wako Pure Chemical Ind. Osaka, Japan), and tetrodotoxin (TTX) (Sankyo Pharmaceutical Co. Tokyo, Japan).
Statistical analysis Results are presented as mean + SEM. The significance of differences was assessed with Student's t-test for paired data and a level of probability P < 0.05 was considered significant.
RESULTS
Concentration-dependent effect of caffeine The
chronotropic
taneously
beating
effects rabbit
of
caffeine
SA
node
in
spon-
cells
were
examined. Caffeine at 0.5 to 2mM caused only a positive chronotropic effect (Fig. 1). At 5 and 10mM, caffeine caused a transient positive chronotropic effect and then a profound negative chronotropic effect (Figs 2 and 3). The percentage changes in the chronotropic effects are summarized in Table 1. At 5 mM, caffeine caused a positive chronotropic effect, and suddenly a dysrhythmia occurred during the subsequent negative effects, as shown in Fig. 2. The incidence of dysrhythmia is also given in Table 1. Similarly, caffeine (10mM) initially caused a positive effect, and secondly a negative effect (Fig. 3). During the negative effect,
Table 1. Positive and negative chronotropic effects in the presence of caffeine, and incidence of dysrhythmia Chronotropic effect Positive
Negative
Caffeine 0.5 mM
6
6.3 + 2.9
1 mM
6
10.3 + 2 . 2 * *
6 8 9
20.7 + 3.8** 25.3 -t- 3.4** 17.5 + 2.8**
7.3 + 3.1 17.3 + 2.8** 44.6 + 4.0***
5
18.8 + 3.4
49.0_+ 1.2
4
33.5 + 3.5
48.5 + 2.8
2 mM 5 mM I0 mM Pindolol 10 7M + caffeine 10mM Atropine 10 6 M + caffeine 10 mM
Incidence of dysrhythmia
---
Values (%) represent mean _+SEM, as compared with control value. *P < 0.05, **P < 0.01, ***P < 0.001, with respect to control values.
Effect of caffeine on SA node cells
CAFFEINE
1225
5 mM 0.5s
2 rain !
!
!
i
13'1
0
Vt
.2
10
/min _._./----
250 f
e----
150I
oll Fig. 2. Occurrence of dysrhythmia by caffeine in spontaneously beating rabbit SA node cells. Spontaneous action potentials, maximum rate of depolarization (Vmax),and sinus rate are shown at slow and fast speeds. dysrhythmia was elicited. The positive effect at concentrations ranging from 0.5 to 5 mM was concentration-dependent. At 10 mM caffeine enhanced the sinus rate (17.5 + 2.8%, n = 9 , P < 0.01), but the value decreased as compared to the value at 5 mM caffeine. The positive and negative effects were reversible, and the dysrhythmia disappeared after washout. During the washout of caffeine (10mM), the sinus rate had a transient enhancement (rebound) by 7.9 + 0.8% (n = 9, P > 0.05) and then, was gradually recovered to the control value (Fig. 3). In the presence of pindolol (1/~M), caffeine (10mM) caused positive (by 18.8_3.4%, n = 5 ,
I
P > 0 . 0 5 ) and negative (by 49.0+ 1.2%, n = 5 , P < 0 . 0 5 ) chronotropic effects (Table 1). Also, caffeine (10 mM) in the presence of atropine (1 pM) produced the positive and negative chronotropic effects by 33.5+3.5% (n = 4 , P <0.05) and by 48.5 + 2.8% (n =4, P >0.05), respectively. Both positive and negative chronotropic effects were not mediated through fl-adrenoceptor and muscarinic receptor stimulations. Effect o f high K +
The dysrhythmia occurred during the negative chronotropic effect following the positive effect in
CAFFEINE 10mM
l
2mip
tt , !,
mY
OF~,' 1
~
-6o L / ~ Ws 10[ /rain 250 I
100
Om5 s
._---
i'
Fig. 3. Dysrhythmia induced by high concentrations of caffeine. Application of caffeine (10 mM) caused a transient positive chronotropic effect and then, elicited dysrhythmia. After washout, regular rhythm was resumed, and the sinus rate was recovered to control value followed by a small rebound.
,,
,
i
1226
HIROYASU SATOH ,
CAFFEINE 10 mM b
a
m
I
KSmM
c
d
a
,,
,
!,, ii
¸
0
-60
b
W:
S!!,!
:] l,,
10 /min
200~ ! 2 min 100
":
:. . . . . ,
0
,r,r . . . . . . . . . ~.,
I
~'4,
d
Fig. 4. Modulation by high K + solution of negative chronotropic effect and dysrhythmia induced by caffeine. Caffeine (10 raM) induced dysrhythmia followed by a transient increase in the sinus rate. Addition of high K + abolished the dysrhythmia, but did not return the sinus rate to the control value. Triangles (a~d) on the top of the action potential recording indicate the action potentials at slow speed on the right panel. the presence of caffeine (10 m M ) (Fig. 4). Increasing extracellular K + concentration from 2.7 to 8 m M during the negative chronotropic effect depressed the action potential amplitude and the maximum rate of depolarization (I?max), and abolished the dysrhythmia in all 6 preparations. High K + solution reduces the [Ca]i level (Ellis, 1977). The depressed sinus rate was increased by 65.3 _+ 6.4% (n = 6, P < 0.001) by taking the value of the negative chronotropic effect at 1 0 m M as control (but not recovered to the control value) (Table 2). The rebound of recovery was 20.1__+ 3.5% (n = 5, P < 0.05). Effect of TTX
Addition of T T X (10-TM) also abolished the dysrhythmia and regular rhythm was resumed, as shown in Fig. 5. The sinus rate was increased by Table 2. Modulation of negative chronotropic effect induced by caffeine and incidence of sinus arrest Chronotropic Incidenceof n effect sinus arrest K + 8mM 6 +65.3 _+6.4*** 0 TTX 0.1 ,uM 5 +28.0_+5.3** 0 Ca2+ 0.5 mM 5 + 15.3 _+5.5* 1 Ca2+ 7.4mM 6 -52.1 _+1.2"** 3 Values (%) represent mean +_ SEM. Caffeine (10 mM) caused the negative chronotropic effect by 43.7 _+ 3.8% (n = 22, P < 0.001) following the positive chronotropic effect (by 20.5_+3.9%, n = 2 2 , P <0.05). *P <0.05, **P <0.01, ***P <0.001, with respect to the values of negative chronotropic effect at l0 mM caffeine.
28.0 _+ 5.3% (n = 5, P < 0.05) (Table 2). When the bath solution was changed to normal solution, the rebound of sinus rate was produced (by 25.3 +_ 2.9%, n = 5, P < 0.01). And then, the sinus rate was gradually recovered to control. E f f e c t s o f e x t r a c e l l u l a r l o w a n d high Ca 2+ solutions
Decrease in extracellular Ca 2+ concentration to 0.5 m M (from 1.8 m M in normal solution) abolished the dysrhythmia, and increased the sinus rate (Fig. 6). The increase in sinus rate was 15.3 _+ 5.5% (n = 5, P < 0.05), but subsequently a sinus arrest occurred in I of 5 preparations (Table 2). The rebound during washout was 3 1 . 8 + 2 . 0 % ( n = 5 , P<0.01). In contrast, increasing Ca 2+ concentration to 7.4 m M potentiated the negative chronotropic effect induced by caffeine (10 raM) by 52.1 _+ 1.2% (n = 6, P < 0.001) (Fig. 7). The high Ca 2+ not only failed to inhibit the dysrhythmia, but also increased further the incidence of sinus arrest (in 3 of 6 preparations) (Table 2). Similarly, the rebound of sinus rate during washout was induced by 22.3 _+ 4.2% (n = 4, P < 0.05), accompanied with dysrhythmia. DISCUSSION
It has already been shown that caffeine caused biphasic inotropic actions in canine Purkinje fibers (Satoh and Vassalle, 1985, 1989). The positive response was due to Ic~ enhancement and Ca 2+
Effect of caffeine on SA node cells
I TTX 10 -7 M I CAFFEINE 10 mM b
a
2 min
c
m
mV
o
a
-60
1227
II/l !l
,:i i /'d :/d d 'd "d ,,,
0.SS i
il
10
\
!,
t ~T--~ V T 'T 'T --~ ,/'.'
/-C
Fig. 5. TTX inhibition in the dysrhythmia induced by caffeine. Caffeine (10 mM) elicited dysrhythmia followed by positive chronotropic effect. Addition of TTX (10 -7 M) abolished the dysrhythmia, but the sinus rate was not recovered to the control value.
release from the stores. The subsequent negative response was presumably due to [Ca]i elevation (cellular calcium overload) by Ca 2÷ re-uptake inhibition into SR (Thorpe, 1973; Fabiato and Fabiato, 1975), as well as by /Ca enhancement and Ca 2÷ release. The findings in the present study are as follows: (a) caffeine (0.5 to 1 mM) caused only a positive
I
CAFFEINE 10 mM
I
chronotropic effect, and (b) at over 5 mM, caffeine produced an initial positive and subsequent negative chronotropic effect. (c) The positive chronotropic effect was not modified by pindolol. (d) The negative chronotropic effect (unaffected by atropine) was exaggerated by high Ca :+ but was inhibited by high K +, T T X and low Ca 2÷.
Ca 0.5 mM
2 rain
mV
V/ 10 /rain 20O
S
100
r
Fig. 6. Effect of low Ca 2+ on the dysrhythmia-occurring SA node cells. Dysrhythmia induced by caffeine (10 mM) was abolished by addition of low Ca 2÷ (0.5 mM). Decrease in Ca 2÷ concentration caused a positive chronotropic effect under the calcium overload condition.
1228
HIROYASUSATOH
I
CAFFEINE
I Ca 7.4mM I
?mip
10 m M
0 -60
10 /min
50
...... . .
.
~
"
Fig. 7. Effect of high Ca -~+ on the dysrhythmia-occurring SA node cells. Increase in Ca 2+ concentration (to 7.4 mM) decreased further the depressed sinus rate by caffeine (10 mM).
Positive chronotropic effect The positive chronotropic effect was produced at low concentrations of caffeine (0.5 to 1 mM), and at higher concentrations, as initial response after application of caffeine. The positive effect was concentration-dependent (at 0.5 to 5 mM), but caffeine at 10 mM produced the positive effect less than at 5 mM (although the value was still enhanced as compared with control value). The positive response was not modified by pindolol, but was rather potentiated by atropine (probably due to loss of muscarinic receptor stimulation). Therefore, the positive chronotropic effect induced by caffeine is not mediated through fl-adrenoceptor stimulation by endogenous catecholamines, but would be due to /ca enhancement and Ca 2+ release from SR (Weber and Herz, 1968; Fuchs, 1969). There are three hypotheses for the pacemaker current in SA node cells (Noble, 1984); (a) the /ca hypothesis, (b) the I K decay hypothesis, and (c) the hyperpolarization-activated inward current (Ih) hypothesis. In voltage-clamped SA node cells, caffeine enhanced /Ca and lh, and inhibited IK (Satoh, 1993). Thus, the initial positive chronotropic effect would be due to the enhancements of/ca and I h. Surprisingly, these enhancements were smaller in the presence of 10 rather than 5 mM caffeine, consistent with the positive chronotropic effect of caffeine at high concentration in the present experiments. This inhibition would be secondly produced by depressant effect due to cellular calcium overload. Because Ic, amplitude is
strongly inhibited under the excessive [Ca]~ condition (Satoh et al., 1989). Recently, it has been reported that the T-type Ca 2+ current makes a major contribution to the spontaneous activity of SA node cells (Hagiwara et al., 1988). If so, caffeine might directly stimulate the T-type Ca 2+ channels. To elucidate this, the extended experiments require using single SA nodal cells. The secondary positive chronotropic effect was the rebound of recovery during washout. A similar recovery rebound has already been shown in the inotropic effect in canine Purkinje fibers (Satoh and Vassalle, 1985). Once caffeine was discontinued, Ca 2+ re-uptake into SR would be stimulated, resulting in the re-start of exaggerated and compensatory Ca 2+ release. In addition, the rebound recovery of [Ca]~ levels has actually been observed in fura-2 loaded single SA node cells (Satoh, 1993).
Negative chronotropic effect The negative chronotropic effect was subsequently produced by application of caffeine at 2 m M or higher in a concentration-dependent manner, which was unaffected by atropine. In the presence of both atropine and pindolol, caffeine (10 mM) potentiated the decrease in sinus rate, but not significantly. The potentiation by pindolol would be due to loss of fl-adrenoceptor stimulation, but the potentiation by atropine is not clear. If muscarinic receptor stimulation is blocked by atropine, the decline in sinus rate by caffeine should be reduced in the presence of
Effect of caffeine on SA node cells atropine. These results suggest that caffeine may have a direct negative chronotropic action. Noble (1984) suggested that Ca 2÷ current plays an important role only during the last 30% of diastolic potential, generating spontaneous action potentials in SA nodal cells. Our previous studies have also demonstrated similar results (Satoh and Hashimoto, 1991; Satoh and Tsuchida, 1993). In voltage-clamped rabbit SA node cells, caffeine enhanced/ca and/h, and inhibited I K (Satoh, 1993). Thus, the reduction in I K may cause the negative chronotropic effect in this study. To prolong the cycle length in the presence of caffeine, since caffeine enhanced /ca, the maximum diastolic potential (MDP) must be hyperpolarized (although if the threshold potential is not shifted). However, the MDP was not significantly hyperpolarized by caffeine (Satoh, 1993). Thus, it seems unlikely that the pacemaker activity is caused by only one ionic current, which is also a result obtained from our studies on the modulation of pacemaker activity (Satoh et al., 1989; Satoh and Hashimoto, 1991; Satoh, 1991; Satoh and Tsuchida, 1993). The mechanism of negative effect was examined by procedures to decline [Ca]i (TTX, high K ÷ and low Ca 2÷) and to elevate [Ca]i(high Ca 2÷) (Satoh and Vassalle, 1985). High K ÷ decreases intracellular Na ÷ activity (aiNa) in quiescent and active Purkinje fibers (Ellis, 1977; Lee and Vassalle, 1983), and therefore presumably decreases [Ca]i through an enhanced Na42a exchange (Mullins, 1979). In addition, high K ÷ depolarizes the resting potential, which could indirectly inhibit the slow inward current and thus contribute to the decrease in [Ca]i. Similarly, TTX decreases aiNa in quiescent and active Purkinje fibers (Vassalle and Lee, 1984; Deitmer and Ellis, 1980) and decreased contractile force markedly (Satoh and Vassalle, 1985). Low extracellular Ca 2+ also decreases [Ca]i and /ca decreases (Reuter, 1967). In the present experiments, therefore, high K ÷, TTX and low Ca 2+ share a similar action on [Ca]i and on the chronotropic effect. On the other hand, high Ca 2÷ enhances /ca and elevates [Ca]i level. In this study, the further negative chronotropic effect of caffeine at high Ca 2÷ was produced. These results suggest that caffeine initially caused the stimulatory action, and then, the negative chronotropic effect might be induced by development of excessive Ca 2÷ load in cytosol. The caffeine-induced cellular calcium overload has already been demonstrated in canine and sheep Purkinje fibers (Satoh and Vassalle, 1985, 1989; Hasegawa et al., 1987) and in rabbit SA node cells (Satoh, 1993). Caffeine elicited a transient inward current (/Ti) and an inward tail current (lex), resulting in induction of arrhythmias. During calcium over-
1229
load (the negative chronotropic effect), caffeine (5 to 10 mM) still enhanced/ca and lb. Actually, the [Ca]i level in rabbit SA node cells was enhanced using fura-2 Ca 2+ fluorescent dye. Also, the rebound recovery of chronotropic effect during washout reflected the excessive Ca 2+ load induced by caffeine. Therefore, these results indicate that the negative chronotropic effect is due to the development of cellular calcium overload, accompanied by the depressions in ionic currents.
REFERENCES
Bourinet E., Fournier F., Lory P., Charnet P. and Nargeot J. (1992) Protein kinase C regulation of cardiac calcium channels expressed in Xenopus ooeytes. Pfliigers Archs 421, 247-255. Deitmer J. W. and Ellis D. (1980) The intracellular sodium activity of sheep heart Purkinje fibers: Effects of local anaesthetics and tetrodotoxin. J. PhysioL (Lond.) 300, 269-282. Eisner D. A. and Lederer W. J. (1979) Inotropic and arrhythmogenic effects of potassium-depleted solutions on mammalian cardiac muscle. J. Physiol. (Lond.) 294, 255-277. Ellis D. (1977) The effects of external cations and ouabain on the intracellular sodium activity of sheep heart Purkinje fibers. J. Physiol. (Lond.) 273, 211-240. Fabiato A. and Fabiato F. (1975) Contraction induced by a calcium-triggered release of calcium from the sarcoplasmic reticulum of single skinned cardiac cells. J. Physiol. 249, 469-495. Fuchs F. (1969) Inhibition of sarcotubular calcium transport by caffeine: Species and temperature dependence. Biochim. Biophys. Acta. 172, 566-570. Goto M., Yatani A. and Ehara E. (1979) Interaction between caffeine and adenosine on the membrane current and tension component in the bull frog atrial muscle. Jap. J. Physiol. 28, 393-409. Hagiwara N., Irisawa H. and Kameyama M. (1988) Contribution of two types of calcium currents to the pacemaker potentials of rabbit sino-atrial node cells. J. Physiol. (Lond.) 395, 233-253. Hasegawa J. and Vassalle M. (1986) The effect of caffeine on the inward current in sheep cardiac Purkinje fibers. Fed: Proc. 45, 288. Hasegawa J., Satoh H. and Vassalle M. (1987) Induction of the oscillatory current by low concentration of caffeine in sheep cardiac Purkinje fibers. Naunyn-Schmiedeberg's Archs Pharmac. 335, 310-320. Lee K. S. and Vassalle M. (1983) Role of calcium in the inotropic effects of caffeine in cardiac Purkinje fibers. Int. J. Cardiol. 3, 421-434. Mullins L. J. (1979) The generation of electrical currents in cardiac fibers by Na/Ca exchange. Am. J. Physiol. 236, CI03--CI10. Noble D. (1984) The surprising heart: A review of recent progress in cardiac electrophysiology.J. Physiol. (Lond.) 353, 1-50. Reuter H. (1967) The dependence of slow inward current in Purkinje fibers on the extracellular calciumconcentration. J. Physiol. (Lond.) 192, 479-492. Satoh H. (1991) Pharmacology and therapeutic effects of mepirodipine. Cardiovasc. Drug Rev. 9, 340-356. Satoh H. (1993) Caffeinedepression in spontaneous activity in rabbit sino-atrial node cells. Gen. Pharmac. 24, 555-563. Satoh H. and Hashimoto K. (1991) Comparative effects of a new calcium channel antagonist, mepirodipine,
1230
H1ROYASUSATOH
on rabbit spontaneously beating sino-atrial node cells. Eur. J. Pharmac. 193, 9 13. Satoh H. and Tsuchida K. (1993) Comparison of a calcium antagonist, CD-349, with nifedipine and verapamil in rabbit spontaneously beating sino-atrial node cells. J. Cardiovasc. Pharmac. 21,685-692. Satoh H. and Vassalle M. (1985) Reversal of caffeineinduced calcium overload in cardiac Purkinje fibers. J. Pharmac. Exp. Ther. 234, 172-179. Satoh H. and Vassalle M. (1989) Role of calcium in caffeine-norepinephrine interactions in cardiac Purkinje fibers. Am. J. Physiol. 257, H226-H237. Satoh H., Tsuchida K. and Hashimoto K. (1989) Electrophysiological actions of A23187 and X-537A in rabbit sino-atrial node cells. Naunyn-Schmiedeberg's Archs Pharmac. 339, 320-326.
Scholtz H. and Reuter H. (1976) Effect of theophylline on membrane currents in mammalian cardiac muscle. Naunyn-Schmiedeberg's Archs Pharmac. 293, RI9. Thorpe W. R. (1973) Some effects of caffeine and quinidine on sarcoplasmic reticulum of skeletal and cardiac muscle. Can. J. Physiol. Pharmac. 51,499-503. Vassalle M. and Lee C. O. (1984) The relationship among intracellular sodium activity, calcium, and strophanthidin inotropy in canine cardiac Purkinje fibers. J. Gen. Physiol. 83, 287-307. Weber A. and Herz R. (1968) The relationship between caffeine contracture of intact muscle and the effect of caffeine on reticulum. J. Gen. Physiol. 52, 750-759. Yatani A. Imoto Y. and Goto M. (1984) The effects of caffeine on the electrical properties of isolated, single rat ventricular cells. Jap. J. Physiol. 34, 337-349.