J. Physiol. (1981), 311, pp.

585-604 585
With 11 text-figures
Printed in Great Britain

THE INOTROPIC ACTIONS OF ADRENALINE

ON FROG VENTRICULAR MUSCLE: RELAXING VERSUS
POTENTIATING EFFECTS
BY MARTIN MORAD, CHRIS SANDERS AND JAMES WEISSt
From the Department of Physiology, University of Pennsylvania School of Medicine,
Philadelphia, Pennsylvania 19104, U.S.A.

(Received 12 December 1979)

SUMMARY
1. In frog ventricle, adrenaline increases the size ofthe action potential, potentiates
twitch tension, and enhances relaxation. Because tension development is directly
controlled by membrane potential in frog ventricle, experiments were designed to
separate the effects of adrenaline on the action potential from its effects on the
development of tension.
2. Comparison of the tension-voltage relations in the presence and absence of
adrenaline showed that during the initial portion of the voltage clamp step,
adrenaline potentiated tension, but beyond 1 sec into the clamp pulse tension was
depressed.
3. The time and voltage dependence of the positive inotropic effect of adrenaline
during voltage clamp pulses were compatible with the kinetics of the slow inward
current, which is known to be augmented by adrenaline in frog and mammalian
ventricle.
4. Ni2+, which has been shown to block the slow inward current in frog ventricle,
also inhibited the positive inotropic effect of adrenaline.
5. The relaxant effect ofadrenaline was demonstrated to be present at least as early
as 600 msec after the onset of membrane depolarization. However, generally 1 sec
or more of membrane depolarization was required before the relaxant effect of
adrenaline predominated over its positive inotropic effect.
6. In catecholamine depleted strips, the augmentation of the action potential and
twitch tension in the presence of adrenaline was found to occur at a sixty-fold lower
concentration than the relaxant effect as judged by suppression of KCl-induced
contractures.
7. Pure fl-receptor agonists reproduced completely the electromechanical effects
of adrenaline on the frog ventricle. a-receptor agonists or antagonists had no effect
on action potential or development of tension.
8. Cyclic AMP and dibutyryl cyclic AMP were found to augment the frog
ventricular action potential and potentiate twitch tension in reserpinized or f-blocked

t Present Address: Division of Cardiology, UCLA Center for Health Sciences, Los Angeles, CA.
0022-3751/81/3510-0660 S07.50 © 1981 The Physiological Society
586 M. MORAD, C. SANDERS AND J. WEISS
frog ventricular strips. However, none of the relaxant effects of catecholamines could
be reproduced by these agents alone.
9. Theophylline produced changes in the action potential similar to those induced
by adrenaline and mimicked both the positive inotropic and relaxant effects of the
drug.
10. The results suggest that the positive inotropic effects of adrenaline results
mainly from changes induced in the action potential plateau. The changes are both
time and voltage dependent, and if inhibited, leave the relaxant effect of adrenaline
unopposed.
11. The findings are consistent with a cyclic AMP-mediated mechanism of the
positive inotropic effect of adrenaline. However, the role of cyclic AMP in mediating
the relaxant effects of adrenaline is less clear.

INTRODUCTION
Adrenaline is known to increase the force of contraction, enhance relaxation, and
change the configuration of the action potential in heart muscle (Kavaler & Morad,
1966; Graham & Lamb, 1967; Morad, 1969; Niedergerke & Page, 1977; also
Coraboeuf, Distel & Boistel, 1955; Gargouil, Tricoche, Fromenty & Coraboeuf, 1958).
In mammalian ventricle the observed increase in the slow inward current produced
by adrenaline (Reuter, 1966, 1967; Reuter & Scholz, 1977b) may be responsible for
elevation and prolongation of the plateau of the action potential and potentiation
oftwitch tension. It has also been suggested that adrenaline enhances the sequestration
of Ca2+ which not only facilitates relaxation but may also increase twitch tension by
augmenting recirculated fraction of Ca2+ (Morad & Goldman, 1973; Morad, Weiss &
Cleemann, 1978; Bassingthwaighte & Reuter, 1972).
Similarly, in the frog heart the adrenaline-induced enhancement of the slow inward
current most likely produces the prolongation ofthe action potential and potentiation
of contraction (Vassort, Rougier, Garnier, Sauviat, Coraboeuf & Gargouil, 1969;
Niedergerke & Page, 1977; Goldman & Morad, 1977 b). Since in the frog ventricle
membrane depolarization directly controls contraction (Morad & Orkand, 1971), in
the present study we attempted to separate the effects of adrenaline on tension
induced by alteration of the action potential from those independent of membrane
potential. The results suggest that alteration of the action potential by adrenaline
is the major cause of potentiation of tension in frog ventricle. On the other hand,
the relaxant effect of the drug does not seem to depend on changes in membrane
potential. The role of cyclic nucleotides as possible mediators of the adrenergic
response in frog ventricle were also investigated. The results are consistent with the
possibility that cyclic AMP mediates the positive inotropic effect of adrenaline.
Although cyclic AMP and its derivatives could be implicated in the adrenaline-induced
enhancement of the relaxation process, their role was not unequivocally supported
by our experiments.
A preliminary report of this work was presented (Morad et al. 1978).
 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 587
METHODS
Preparation. Quiescent strips were prepared from circular cuts through the ventricle of the frog
(Rana pipiens) maintained at room temperature. In the KCl-induced contracture experiments,
strips averaged 0-5-08 mm in diameter and 3-0-5-0 mm in length. In voltage clamp experiments
thinner strips, 0-3-0-5 mm in diameter, were used. All experiments were conducted at room
temperature (23 'C).
Experimental set-up. In the KCl-induced contracture experiments, frog ventricular strips were
mounted in a 0 3 ml. Plexiglass bath. One end was fixed in place with a clamp, and the other end
attached to a tension transducer to record isometric tension. A Grass stimulator (model 544) whose
output was connected to a stimulus isolator (Grass Model SIU5A) was used to stimulate the
preparation along its entire length using Ag/AgCl electrodes. Transmembrane potential was
measured with standard glass micro-electrodes (tip diameter 0 1 usm) connected to the input of a
source follower. Tension and intracellular potential were displayed on an oscilloscope (Tektronix
Model 511) or chart recorder (Brush Mark 220). Voltage clamp experiments were all performed using
the single sucrose gap technique described by Morad & Orkand (1971) and Goldman & Morad
(1977a).
Solutions. The standard perfusate was normal Ringer solution, consisting of 116 mM-NaCl,
2 mM-NaHCO3, 3 mm-KCl, 1 mM-CaCl2. In experiments involving KCl-induced contractures,
Ringer solution was made hypertonic by the addition of 100 mM-KCl. In some experiments
1-2 mM-NiCl2 was added to Ringer solution. For voltage clamp experiments the CaCl2 concentration
in the Ringer solution was reduced to 0-2 mM.
Drugs. Drugs were used in the following doses: adrenaline (Parke Davis), 5 x 10-'-
5 x 10-6 M; isoproterenol (Elkins-Sinn, Inc.), 5 x 10-' M; methoxamine, 4 x 10-64 x 10-5 M; phenyl-
ephrine (Sigma), 5 x 10-4 M; phentolamine (CIBA), 3.5 x 10--3-5 x 10-4 M; propranolol (Ayerst),
10-6M; theophylline (Sigma), 2-5 mM; dibutyryl cyclic AMP (Sigma), 1-5 mM; cyclic AMP (Sigma),
1-5 mM; tetrodotoxin, 10-6 M. Reserpinized frogs were pretreated with reserpine (CIBA), 1-5 mg/kg
i.p., for 3 days before sacrifice.

RESULTS

Fig. 1 (upper panels) shows the effect of adrenaline on the action potential and
contraction of a frog ventricular strip. As the preparation was exposed to adrenaline,
the plateau of the action potential became progressively elevated and prolonged with
successive beats. The rate of rise and magnitude of the contraction increased as did
the time-to-peak of tension. However, the peak of twitch tension occurred earlier
relative to the rapid repolarization phase of the action potential. This observation
shows that the marked potentiation of twitch tension caused by adrenaline occurs
concomitantly with the alteration of the action potential. Similar effects on the
action potential and contraction were observed when the ventricular strip was
exposed to DB cyclic AMP and theophylline (Fig. 1, lower panels). No significant
differences were observed between the effects of non-permeant cyclic AMP and
permeant DB cyclic AMP on the time course or magnitude ofdeveloped tension (Table
1, Fig. 8).
Inotropic effects of adrenaline on the myocardium
It is well known that KCl-induced contractures in frog and mammalian ventricular
muscle are markedly suppressed in the presence of adrenaline (e.g. see Fig. 2 of Morad
et al. 1978; Graham & Lamb, 1967; Morad & Rolett, 1972). Since adrenaline does
not alter significantly the magnitude of membrane depolarization produced by high
concentrations of KCl (Graham & Lamb, 1967), adrenaline appears to have a direct
relaxant effect on frog ventricular muscle.
588 M. MORAD, C. SANDERS AND J. WEISS
Further evidence that the tension-suppressant effects of adrenaline in high KCl
solutions occur primarily because the membrane potential is prevented from being
altered by adrenaline comes from experiments where the membrane potential was
controlled by the voltage clamp technique. Fig. 2 shows that the steady-state tension
generated during a 3 see clamp step is uniformly smaller in the presence of adrenaline
than in its absence at all membrane potentials positive to -30 mV. This observation

30 - -
0

80 0 ] 50 mg

+15 mV S

-80 mVL I0m]iOOmg

J
1 sec
Fig. 1. The upper left panel shows superimposed action potentials and contractions of a
frog ventricular strip as it is exposed to 5 x 10-6 M-adrenaline (EPI). As the action
potential plateau becomes prolonged and elevated, rate of rise and magnitude of
twitch tension increase concomitantly. The upper right panel shows the steady-state
action potential and contraction in the presence of adrenaline. The lower panels show
superimposed action potential and contractions before and after 1 mM-DB cyclic AMP (left)
and 2 mM-theophylline (THEO). Both compounds increase plateau height and action
potential duration and also increased twitch tension. Temp. 23 'C; stimulus frequency,
12/min; [Ca]0 = 0-2 mm.

is consistent with the idea that when adrenaline is prevented from altering the
membrane potential, its primary effect is to suppress tension.
The suppression of the tension-voltage relation in the presence of adrenaline at first
seems to contradict the well known positive inotropic effect of the drug (see action
potentials and contractions in the presence and absence of adrenaline shown in insets
of Fig. 2). However, more extensive investigation of the generated tension
accompanying a clamp step shows that the suppressant effect of adrenaline on tension
requires at least one second from the onset of depolarization to predominate over the
positive inotropic effect of the drug. Fig. 3 compares the tension-voltage relations
in the presence and absence of adrenaline at 0-5 see intervals during 5 see voltage
clamp pulse. Note that at 0 5 see from the onset of the clamp, tension is potentiated
in the presence of adrenaline at potentials negative to + 30 mV. At 1 0 see into the
clamp pulse, the developed tension is similar in the presence and absence of
adrenaline. At 1-5 see into the clamp pulse, the relaxant effect of adrenaline is well
apparent. Thus, during depolarizations longer than about one second, the dominant
 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 589
effect of adrenaline is to enhance relaxation and suppress tension, while at earlier
times the positive inotropic effect of the drug predominates. This effect of the drug
can be seen, in fact, during the time course of a contraction accompanying a single
action potential. For example, in Fig. 1 (upper panel), tension first increases rapidly
and peaks within the first 700-800 msec of the action potential, and then begins to
fall off before complete relaxation occurs upon repolarization.

TABLE 1. The effects of DB cyclic AMP and cyclic AMP on contraction and peak contracture tension
Iso-
cAMP or proterenol
Control DB cAMP (% change) Control 10- g/cc (% change)
6/26* (mg) (mg) (mg) (mg)
Contraction 209 352 +68 214
Peak contracture 247 200 -19 236
6/28*
Contraction 162 261, 228 +51 142
Peak contracture 176 152, 104 -28 190
7/10*
Contraction 114 261 130 133 256 +92
Peak contracture 226 226 0 256 195 -24
7/11 *
Contraction 162 285, 261 +68 221 332 +50
Peak contracture 226 254, 238 +9 249 162 -35
7/15*
Contraction 133 223,190 +55 138 276 +100
Peak contracture 190 209,190 +5 223 81 -64
7/27**
Contraction 147 228, 204 +47 104 200 +92
Peak contracture 142 109,100 -26 152 48 -68
8/if
Contraction 114 256 +124 142
Peak contracture 190 228 +20 238
2/17ff
Contraction 276 589, 532 +103 256 608 + 137
Peak contracture 456 408, 380 -14 437 114 -74
Mean change in response to 1 mM-DB cyclic AMP was + 74±32% (S.D.) for contraction and
-6-6+16% (s.D.) for contractures. Mean change in response to 10-6 g/cc isoproterenol was
+ 94 ± 31 % (S.D.) for contraction and -53 + 32 % (S.D.) for contractures. The percentage change in
response to DB cyclic AMP or isoproterenol was in comparison to the preceding control. When
two runs were done in DB cyclic AMP the values were averaged. * 1 mM-DB cyclic AMP; **
4-5 mM-DB cyclic AMP; f 2-5 mM-cyclic AMP; ft 4 mM-cyclic AMP.

Some meaturement8 reflecting the relaxant capacity of ventricular muscle. Although

the relaxant effect of adrenaline during voltage clamp pulses does not appear to
predominate over the positive inotropic effect until > 1 sec into the clamp step, the
enhancement of relaxation brought about by adrenaline can be detected much earlier
in the time course of the clamp pulse. A rough index of the ability of frog ventricular
muscle to relax may be obtained by measuring the time delay between the
termination of the clamp pulse and the onset of relaxation of tension. In Fig. 4 this
time delay (AT) is compared in the presence and absence of adrenaline for a series
of voltage clamp pulses. The onset of relaxation was found to be faster in the presence
590 M. MORAD, C. SANDERS AND J. WEISS
of adrenaline at all membrane potentials tested. The rate of relaxation following the
voltage clamp pulses was also compared in the presence and absence of adrenaline.
Fig. 5 (upper panel) shows that after a 0-6 see voltage clamp pulse tension decays
exponentially with a time constant of approximately 120 msec. In the presence of
adrenaline (lower panel) tension decays more rapidly, with a time constant of
approximately 55 msec at all membrane potentials tested.

0 0-2 Ca2+ Ringer

20 5 MA
mV r Tension (mg)
30
-80
20 mg

20 5pA
mV

-80
20 mg

-60 -40 -20 0 20 40

Membrane potential (mV)

Fig. 2. The effects of adrenaline (EPI) on the steady-state voltage-tension relations in

frog ventricle. Using a single sucrose gap technique a frog ventricular strip was voltage
clamped to various membrane potentials for a period of 3-5 sec. The final steady-state
tension at the end of the clamp is plotted against membrane potential. Over the range
of membrane potentials tested (-70 to + 40 mV) the final tension is lower in the presence
of 5 x 10-6 M adrenaline (@) than in its absence (0). The upper panels show two
representative voltage clamps to +40 mV in the absence (upper) and presence (lower) of
adrenaline. Temp. 23 'C, strip diam. 0 3 mm, [Ca]o = 0-2 mM.

The results illustrated thus far suggest that alteration of the action potential is in
part responsible for the positive inotropic action of adrenaline. Experiments such as
those illustrated in Fig. 3 strongly implicate that a time- and voltage-dependent
mechanism, which activates and inactivates within 1 sec after the onset of depolar-
ization, is responsible for this effect. Since adrenaline is known to increase the
membrane conductance to Ca2+ within a narrow voltage and time range (Reuter,
1974; Reuter & Scholz, 1977b; Vassort et al. 1969), it is reasonable to postulate that
the positive inotropic effect of adrenaline and alteration of the action potential may
be mediated by a single mechanism. Since Ni2+ was found previously to be a very
effective blocker of Ca2+ channels (Morad & Klitzner, 1978), we examined the effect
 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 591
of adrenaline on the action potential and contraction in the presence of Ni2+. Fig.
6 (upper panels) shows a simultaneous recording of action potential and contraction
from a frog ventricular strip as it was perfused with Ringer solution containing
2 mM-NiCl2. In the presence of Ni2+, the action potential became prolonged, the
plateau was depressed, and twitch tension decreased. Ni2+ has been described to

Ringer (.) EPI (a) 20

20 5 MA
mV
80
mV

Sp-5,AX 20 mg
0
c 1O

80 20mg -80 -40 0 40 Vm ImV)

sc 30 30'

20 _H1-e _ 20c
0)

C~~~~~~~~~~
lo 10 0

-120 -80 -40 0 40 -120 -80 -40 0 40

Vm (mV) Vm (mV)
Fig. 3. The effects of adrenaline (EPI) on voltage-tension relations in frog ventricle. Using
a single sucrose gap technique, a frog ventricular strip was voltage clamped to various
membrane potentials for a period of 3-5 sec. In the three graphs the tension developed
at 0-5, 1 0, and 1-5 sec, respectively, into the clamp is plotted as a function of membrane
potential in the absence (0) and presence (0) of 5 x 10-6 M-adrenaline. At 0-5 sec into
the clamp, tension developed in the presence of adrenaline is greater for membrane
potentials less than + 30 mV. At 1 0 sec, developed tension is slightly reduced in the
presence of adrenaline. By 1-5 sec, developed tension is markedly suppressed in the
presence of adrenaline. Insets show representative voltage clamps to +40 and +20 mV
(with superimposed action potentials and contractions) in the absence (left) and presence
(right) of adrenaline. Temp. 23 C, strip diam. 0-3 mm, [Ca]o = 02 mM.

increase the Na+ selectivity of the overshoot of the action potential from 18 to
61 mV/decade, and to block the slow Ca2+ sensitive inward current, without any
measurable effect on the inward rectifying or delayed rectifying K+ currents in frog
ventricular muscle (Morad & Klitzner, 1978). Fig. 6 (lower panels) shows that when
the Ni2+-treated preparation was exposed to adrenaline, twitch tension was markedly
reduced without significant changes in action potential duration or size. KCl-induced
592 M. MORAD, C. SANDERS AND J. WEISS
contractures in the presence of Ni2+ were also suppressed by adrenaline. These results
are consistent with the idea that if the ionic mechanism which is responsible for the
alteration of the action potential plateau is blocked, the positive inotropic effect of
adrenaline is also suppressed, leaving the relaxant effect unopposed during the time
course of the action potential.

50 --

+ Ringer 10 pA[
0 mV

40 - /

-
- 30
/-r /~~~ 20
II
mg[
200 msec
E EPI

cl20 - - /o pmV

10 -
I
-80 mJ
4 g 20mg(
_- 200 m sec

--20 0 20 40 60 80 100 120

Membrane potential ImV)
Fig. 4. The effect of adrenaline on relaxation of tension following a voltage clamp step
in frog ventricle. In the graph the time interval (AT) between the termination of the
voltage clamp step and relaxation of developed tension is compared before (@) and after
(0) exposure to 5 x 10-6 M-adrenaline (EPI). The onset of relaxation is more rapid in the
presence of adrenaline over the full range of membrane potentials tested (-10 mV to
+60 mV). Insets show two representative voltage clamp steps (with superimposed
action potentials and contractions) in the absence (upper) and presence (lower) of
adrenaline (see arrows). Temp. 23 TC, [Ca]0 = 0-2 mm, strip diam. 0 3 mm.

Pharmacological experiments
Effect of a and , adrenergic agents on frog ventricular muscle. Frog ventricular strips
iwere exposed to receptor selective agonists and antagonists to define the contribution
of a and , receptor stimulation to the total adrenergic response of the tissue. The
pure a-receptor stimulant methoxamine was found to have no significant effect on
the action potential or contraction of frog ventricular strips in concentrations less
than 4 x 1O-5 M. At higher concentrations twitch tension was suppressed. KCI-induced
contractures were not altered in the presence of methoxamine. Application of the a
receptor blocker, phentolamine (3 5 x 10-5 M), to ventricular strips previously exposed
to adrenaline (5 X 10-6 M) or phenylephrine (5 x 10-4 M) also failed to reverse the
elevation or prolongation of the action potential plateau or affect twitch tension
significantly. These results suggest that in frog ventricle the inotropic effects of
adrenaline are not mediated via a receptor stimulation.
In contrast, the pure /1 receptor stimulant isoproterenol did reproduce the major
 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 593
effects of adrenaline. These observations suggest that adrenaline augments the action
potential, increases twitch tension, and enhances relaxation in frog ventricle primarily
through stimulation of the fi receptor with little if any contribution from the a
adrenergic stimulation.
Ringer
100 +40 mV
50 10 MA

-80 mV 20 mg
E
c
.0 10
I-
5 r -125 msec

I- -I- ---- -I- -I.-

50 100 150 200 250
Time (msec)
100 EPI
- +40 mV M
50
10 1A

E
0
a_
fn 20 mg
10
I-

f~~II I

1 50 100 150 200 250

Time (msec)
Fig. 5. The effect of adrenaline on the rate of relaxation of tension following voltage clamp
steps in frog ventricle. In the upper panel, the fall of tension after the termination of
voltage clamp steps to -10 mV (A), 0 mV (A), + 20 mV (A), and + 60 mV (A) is plotted
semilogarithmically as a function of time. Tension decays exponentially with a time
constant (T) of approximately 125 msec. In the lower panel, the preparation has been
exposed to adrenaline (5 x 10-6 M) and the same measurements replotted. The time
constant for relaxation following voltage clamps to 0 mV (0), + 20 mV (/), + 40 mV (E),
and + 60 mV (A) is faster (approximately 58 msec) in the presence of adrenaline. Insets
show representative voltage clamps to + 40 mV in the absence (upper) and presence of
adrenaline. Temp. 23 0C, [Ca]O = 0-2 mm, strip diam. 0 3 mm.

Do8e dependence of the inotropic effects of isoproterenol. The response of the frog
ventricle to various doses of isoproterenol was studied to examine whether the
positive inotropic and relaxant effects of isoproterenol occur at the same concentration.
Fig. 7 shows that potentiation of twitch tension by 50 % occurred at markedly lower
594 M. MORAD, C. SANDERS AND J. WEISS
concentrations than those necessary to suppress KCl-induced contractures in a
preparation depleted of endogenous catecholamine stores. The potentiation of twitch
tension and suppression of contracture tension are plotted as a function of isoproterenol
concentration. The sigmoid relation thus obtained suggests saturation kinetics with
a Km of 5-45 x 10-9 M (r2 = 0998) for the positive inotropic effect and a Km of
3-29 x 1O-7 M (r2 = 0986) for the suppression of contractures as suggested by
Lineweaver-Burke plots. Fig. 7 also shows that propranolol was more effective in

1 mM-Ca +2 mM-Ni,6 min +2 mM-Ni, 18 min

80mV

250 mg [ ,N

1sec

1 mM-Ca, 2 mM-Ni 10-6 EPI

80 mV

L_
50mg[ .. _-l
'

1 sec
Jho 60sec I sec

Fig. 6. The effect of adrenaline on a frog ventricular strip pre-exposed to Ringer containing
2 mM-NiCI2. The upper tracings show simultaneous action potentials and contractions in
normal Ringer and after 6 min and 18 min of exposure to the Ni-containing Ringer
solution. In the presence of Ni, the plateau of the action potential is depressed and
prolonged and twitch tension is decreased. The addition of adrenaline (5 x 10 M) to the
Ni treated preparation (lower panel) produces little further change in the action potential,
yet markedly suppresses twitch tension. Temp. 23 0C, [Ca]o = 1 0 mm.

suppressing the positive inotropic effect than the relaxant effect of isoproterenol. This
observation coupled with sixty-fold differences in the K. for the two effects of
isoproterenol suggests that different pathways may be involved in the two actions
of the drug.
Positive inotropic and relaxant effects ofcyclic AMPand its derivatives and theophylline.
Since cyclic AMP and its derivatives as well as the phosphodiesterase inhibitor
theophylline potentiate tension and alter the action potential in a manner similar to
adrenaline (Fig. 1), it was of interest to examine whether this agent would also mimic
the relaxant effects of adrenaline. Fig. 8 compares the positive inotropic and the
relaxant effects of cyclic AMP, dibutyryl cyclic AMP and theophylline on three
catecholamine-depleted frog ventricular strips. Theophylline at concentrations of
1 mM or higher always produced marked potentiation of twitch tension and suppressed
 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 595
KCI-induced contractures (Fig. 8C). Cyclic AMP and dibutyryl cyclic AMP, on the
other hand, failed to suppress contracture tension significantly even after prolonged
exposures (greater than 90 min) to concentrations (1-5 mM) which were sufficient to
produce maximal increases of twitch tension (Fig. 8A, B, see also Table 1).

100- A 0, 100 8 0
90 8 - ] ;}0C 90
80
~ ~~ ~ ~~~~~
o
~~80
~~~~~~~~~0
0
90
6- l _ -. 70 0-
0.~~~~~~~~~~~~~~~~~~~~
720 - 0
110
a60- 0 60-n
6 0 0X
EF30 -
50-
.C /
0
0 ~~ ~0
0 ~~~~~0 0
40 -o c 40

E 30~ -

~
~ ~ ~ ~ 30
~~~~~~0 0
-20
0 E 20-
10 0 g10

9 8g7 6.5 9 8 7 6 5
-Log 10 isoproterenol concn. Log,, isoproterenol concn.

control hamlf gm.

rpxt - ---
A A A A
KCI
1 min
5.6 X10-9 500 56 x10" 111.1

g/ml. mg the gutlint. sup s of IiIIJ

56 e tension in thesaereartinsisplttdgans5t6 Xg10 conenraJo
£ £ .AA
Fig. 7. The dose-response curve of twitch potentiation and KCl-induced contracture
suppression to isoproterenol in reserpinized frog ventricular strips. In the left graph, the
percentage of maximal twitch potentiation of eight preparations exposed to graded
concentrations of isoproterenol is plotted as a function of isoproterenol concentration in
the absence (0) and presence (Ul) of 10-6 m-propranolol. The concentration of isoproterenol
producing half maximal positive inotropic response is 5-45 x 10-'1 m in the absence of
propranolol. In the upper right panel, the percentage of suppression of KCl-induced
contracture tension in the same preparations is plotted against isoproterenol concentration
in the absence (0) and presence of 106 m-propranolol (U). Half maximal suppression of
Kwl-induced contracture in the absence of propranolol occurs at a sixty-fold greater
isoproterenol concentration (329 x 10-7 m). Lower traces demonstrate a typical response
of twitch tension and KCl-induced contracture tension to graded increases in the
concentration of isoproterenol. Temp. 23 'C, [Ca]0 = I -0 mm~; 1 g/ml. = 4-7 mm.

In the light of the possibility that cyclic AMP and its derivatives may not
effectively penetrate intracellular compartments, or may be broken down rapidly by
the intracellular phosphodiesterases, the effect of these agents on contracture tension
was also tested in the presence of theophylline. In Fig. 9, a reserpinized preparation
has been pre-treated with theophylline in a concentration (0-8 mm) which by itself
596 M. MORAD, C. SANDERS AND J. WEISS
produced a submaximal suppression of KCl-induced contractures. Exposure of such
a preparation to cyclic AMP resulted in a marked additional suppression of
contracture tension (Table 2). This effect was much greater than would be expected
from the natural tendency of contracture tension to diminish with repeated exposure

A R inger 8 C
Ringer Ringer
f
A
_,

Nt
, ;o PA Al
4, -"
A V
KCI V'-A A 1 min
DBcyclicAMP
500 250[-50
m 500
mg I
mg mgL min
15 min 5 min
cyclicAMP DBcyclicAMP THEO

HKI
I1 - -o
IIILT\<
A v A V

Ringer .--- R inger (-N Ringer g

Hulk'
A V A V A v

EPN
EPI
EPI

I-

A A v

Fig. 8. Comparison of the effects of cyclic AMP, DB cyclic AMP, and theophylline (THEO)
on contraction and KCI-induced contractures in reserpinized frog ventricular strips. In
panel A, upper trace, tension is monitored in a frog ventricular strip bathed in Ringer
solution. At the upward arrow the preparation is exposed to hypertonic Ringer containing
100 mM-KCl and a contracture develops. The high KCI Ringer is then removed (downward
arrow) and the normal contractions resume. In the second trace, the preparation is exposed
to 1 mM-cyclic AMP which produces a marked potentiation of twitch tension. In the third
panel, after 45 min exposure to cyclic AMP, the preparation is again exposed to high KCI
Ringer. Contracture tension is only minimally suppressed. Cyclic AMP is then washed out
and in the fourth tracing a KCl-induced contracture repeated. In the fifth panel the
preparation has been exposed to 5 x 10-7 M-&adrenaline- (EPI). Twitch tension is potentiated
and KCl-induced contracture markedly suppressed. In panels B and C the same sequence
is performed using DB cyclic AMP and theophylline, respectively, in place of cyclic AMP.
After 50 min exposure to DB cyclic AMP (2 mM), KCl-induced contracture is not
significantly altered. 2 mM-theophylline, however, does suppress KCI-induced contracture
tension in a manner similar to adrenaline.

to KCl solution. Similar results were obtained with dibutyryl cyclic AMP (Table 2).
No quantitative or qualitative differences were observed between cyclic AMP and
its derivatives, despite the markedly different cellular permeabilities in rat heart
(Robison, Butcher, 0ye, Morgan & Sutherland, 1965). In Table 2, data in quantitative
 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 597
form has been assembled from all the ventricular strips pre-treated with theophylline
and exposed to cyclic AMP or its derivatives. Regardless of the extent to which
theophylline suppressed contracture tension, addition of cyclic AMP or its derivatives
always produced further suppression of contracture tension, which reversed after
removal of cyclic AMP or its derivatives.

1 sec KCI 1 min

250 [2 iii 31f

0-8 mm -theo 2min 1 sec

4 5

KCI |KCI
1 min 1 min
6 7

5 mM-cyclicAMP + L J A A
08mM-theo 15min 1m__minA
8 ~~~~~~~~~9
cclicAMP
washout

_ _ ~~~~1_ A
KA I min min 45 KCI min

Fig. 9. The effect of cyclic AMP on contraction and KCl-induced contractures in the
presence of low concentration of theophylline. In panel (1) twitch and contracture tensions
are monitored in a reserpinized frog ventricular strip. In panel (2) and (3) the positive
inotropic effect of 0-8 mM-theophylline (theo) is recorded. Panels (4) and (5) show that
in the presence of theo, KCI-induced contracture tension is slightly suppressed. The strip
is next exposed to 5 mM-cyclic AMP in addition to the theophylline; twitch tension is
markedly increased (panel 6) while contracture tension is suppressed (panels (7) and (8)).
Washout of cyclic AMP restores contracture and twitch tension back to control levels
(panel 9). Temp. 23 TC, [Ca]o = 10 mM.

In each of these experiments the effectiveness of cyclic AMP, dibutyryl cyclic AMP
and theophylline in suppressing contracture tension was compared to that of
isoproterenol following washout of these agents (Table 2, see also Table 1). Isoproter-
enol was more effective in suppressing contracture tension in all instances (Fig. 8).
As was shown in Fig. 6 exposure of ventricular strips to small concentrations of
NiCl2 blocked the positive inotropic effect of adrenaline and seemed to unmask a
relaxing effect of the drug. Fig. 10 compares the effect of theophylline, cyclic AMP,
and dibutyryl cyclic AMP, with that of adrenaline in Ni2+-treated preparations.
598 M. MORAD, C. SANDERS AND J. WEISS
TABLE 2. The effects of DB cyclic AMP and cyclic AMP on the twitch and contracture tension
of theophylline pre-treated ventricular strips
0-8 mM 0-8 mm theo
Control theo + nucleotide % change
10/16* (mg) (mg) (mg)
Contraction 598 770, 689 732, 712 -1
Peak contracture 589 342, 276 152, 124 -55
10/22*
Contraction 180 256, 180 437, 446 +103
Peak contracture 256 256, 180 114,95 -52
11/1*
Contraction 247 504, 400 570, 560 +25
Peak contracture 370 247,171 95, 76 -59
11/7*
Contraction 266 808, 418 988, 965 +59
Peak contracture 608 370, 418 247, 266 -30
11/20**
Contraction 152 299, 195 432, 394 +67
Peak contracture 204 214, 176 90, 71 -59
2/23t
Contraction 142 228, 204 413, 332 +72
Peak contracture 347 309, 299 271, 247 -15
2/14t
Contraction 142 304, 209 408, 718 199
Peak contracture 285 209, 190 209, 133 -14
Mean percentage change in contraction was + 46-5 + 45 % (S.D.) and in contracture, = 49 + 13 %
(S.D.). The percentage change in each experiment was calculated on the average theo + DB cyclic
AMP or cyclic AMP values compared to the theo alone values. * 2 mM-DB cyclic AMP; ** 3 mM-DB
cyclic AMP; t 5 mM-cyclic AMP.

250F .1
1 scA
mg L
I
' N i2+ + DBcyclicA 1isec
1 sec

--O
,/7NJ Ni2+ + THEO 14 min

EPIT
5 min
I
I
L-~~~~A_
am Aec I
EPI washout

'2+
500 D
mg
A .

cvclicAMP 5 min
/-\I E

cyclicAMP washout S min 1 sec

 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 599
Similar to adrenaline, these agents all produced suppression of twitch tension in
Ni2+-containing solutions. Once again adrenaline seems to have a more prominent
relaxant effect on developed tension than the presumed mediators of its action.
The relaxant effects of dibutyryl cyclic AMP and theophylline were also investigated

Ringer DBcyclicAMP o

+20 mV
2 pA 20 J0~

-80mV 10mg mV
1 sec -80 50 Tension (mg)
50 0

E
C

0 ~~
~ ~ ~ ~ ~ ~~~~2m 0
25~~
~ ~ ~~~~~~~~~2

0 i sec

-80 -40 +40 -80 -40 0 40

Vm (mV) Vm (mV)

Fig. 11. Effect of DB cyclic AMP and theophylline on the tension-voltage relations in frog
ventricular strips. Using a single sucrose gap technique, frog ventricular strips were
voltage clamped to various membrane potentials for 3 sec. Final tension at the end of the
clamp has been plotted as a function of membrane potential. In the graph on the right
the final tension is lower in the presence of 2-0 mM-theophylline (0) when compared to
that in normal Ringer (0) over the range of membrane potentials tested (-70 to
+40 mV). In the graph on the left tension-voltage relations were the same in the absence
(@) and presence (0) of 2-0 mM-DB cyclic AMP. Insets show representative voltage
clamps in the presence and absence of these agents. Temp. 23 0C, strip diam. 0-3 mm,
[Ca]. = 0-2 mM.
under voltage clamp conditions. Fig. 11 shows that effect of DB cyclic AMP and
theophylline on the tension-voltage relation of two different ventricular strips bathed
in normal Ringer solutions. Dibutyryl cyclic AMP enhanced twitch tension by
enhancing the action potential (see inset of graph). However, developed tension
Fig. 10. Effects of theophylline, cyclic AMP, and DB cyclic AMP on twitch tension in
reserpinized frog ventricular strips exposed to Ni2+. In panel A, tension is monitored in
a frog ventricular strip bathed in Ringer solutions containing 2 mM-NiCl2. At the arrow,
1-0 mM-DB cyclic AMP is added to the perfusate and twitch tension gradually falls over
a period of 40 min. In panels B-E, the same experiment is performed in other strips using
other agents in place of DB cyclic AMP; in panel B, 2-0 mM-theophylline, in panel C,
5 x 10-6 M-adrenaline, in panels D and E, 2-0 mM-cyclic AMP. In panels C and E, the
recovery of twitch tension following washout of adrenaline and cyclic AMP are also shown
respectively. All four agents seem to suppress twitch tension in Ni2+-treated ventricular
strips. Temp. 23 IC, [Ca]. = 1.0 mM.
600 M. MORAD, C. SANDERS AND J. WEISS
accompanying a series of 3 see clamp steps to various potentials was not altered
significantly (compare open and filled circles, left graph). In some experiments, in fact,
a small but definite potentiation of tension at every clamped potential was observed.
Unlike DB cyclic AMP, theophylline produced suppression of the tension-voltage
relation at the end of a 3 see clamp pulse (open circles, right graph) even though
twitch tension was markedly potentiated (right graph insets). The effect oftheophylline
on the action potential, twitch tension, and tension-voltage relations was quite
similar to that found for adrenaline (compare with Figs. 1, 2, 3). Although DB cyclic
AMP potentiated twitch tension and altered action potential in a manner similar to
adrenaline, it failed to show any relaxant effect on twitch tension or tension-voltage
relation.
DISCUSSION

The major purpose of this study was to investigate the inotropic actions of
adrenaline on frog ventricular muscle independent of its effect on membrane
potential. The positive inotropic effect of adrenaline appears to be linked to the
alteration ofmembrane potential induced by the drug. When the mechanism by which
adrenaline changes action potential configuration is inhibited, the major effect of
adrenaline is to enhance relaxation and suppress developed tension.
The inotropic effects of adrenaline appear to be mediated via the beta adrenergic
receptor. While phosphodiesterase inhibitor, theophylline, mimics best the inotropic
actions of adrenaline, the relaxant effect could not be fully reproduced by cyclic AMP
or its derivatives alone. No major differences were found between cyclic AMP and
its more membrane permeable derivatives (mono- or dibutyryl cyclic AMP) in
potentiating twitch tension.
The relaxant effect of adrenaline. The results obtained from both KCl-induced
contractures and voltage clamp experiments suggest that during prolonged membrane
depolarization adrenaline suppresses the development of tension. The relaxant effect
of adrenaline occurs at all membrane potentials and seems to be independent of [K]o
or [Ca]o (Fig. 2 of Morad et al. 1978; Kavaler & Morad, 1966; Graham & Lamb, 1967).
Comparison of the voltage-tension relations at various times during voltage clamp
steps in the presence and absence of adrenaline indicate that approximately one
second of depolarization is required before the relaxant effect is sufficiently large to
suppress the development of tension (Fig. 3). Thus the relaxant effect of adrenaline
seems to require time to develop. However, Figs. 4 and 5 suggest that the relaxant
effect of adrenaline is present at least as early as 600 msec after the onset of the clamp
pulse. This is also supported by the observation that the time to peak of tension occurs
earlier relative to repolarization of the action potential in the presence of adrenaline
(Fig. 1). The suppressant effects of adrenaline on twitch tension in the presence of
Ni2+ (Fig. 6) are also consistent with the idea that when the positive inotropic effect
of the drug is blocked, the relaxant effect is apparent immediately. These results
suggest that Ca2+ sequestering system in the frog ventricle is facilitated directly by
adrenaline and that this effect is independent of the positive inotropic action of the
drug.
Considering the prominent role of sarcoplasmic reticulum in sequestering Ca2+, it
is postulated that adrenaline enhances relaxation by stimulation of the sequestering
 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 601
system through a mechanism which remains to be characterized (see section entitled
Mechanisms of Adrenergic Response).
The positive inotropic effects of adrenaline. It has been previously shown that in the
frog ventricle tension increases as the level and duration of membrane depolarization
are increased (Morad & Orkand, 1971). An important consideration therefore in
evaluating the positive inotropic effect of adrenaline is whether alteration of the
action potential plateau brought about by the drug is in itself sufficient to account
for the observed potentiation of tension. Examination of tension-voltage relations
at times comparable to the action potential duration suggest the presence of a positive
inotropic effect at potentials negative to + 30 mV (Fig. 3). At potentials positive to
+ 30 mV or with clamps of longer duration (> 500 msec) little or no potentiation of
tension was present (Fig. 2 inset, and Fig. 3). These findings suggest that the positive
inotropic effect of adrenaline occurs over a limited range of membrane potentials and
time. In fact, the range of membrane potential and times necessary are similar
to those described for the slow inward current in heart muscle (Reuter & Scholz,
1977a; Beeler & Reuter, 1970; Rougier et al. 1969; Horacikova & Vassort, 1976).
Adrenaline has been shown to increase the slow inward current (instrumental in
determining the level of the plateau) in both frog and mammalian heart (Vassort et
al. 1969; Reuter & Scholz, 1977 b). Thus it is likely that in the presence of adrenaline
a larger number of Ca2+ channels are activated (Reuter, 1974; Reuter & Scholz,
1977b), leading to both elevation and prolongation of the plateau and potentiation
of tension. The finding that Ni2+ blocks both the positive inotropic effect of adrenaline
(Fig. 6) and the TTX-insensitive slow inward current in frog ventricle (Morad &
Klitzner, 1978) suggests that a common mechanism may be responsible for the
adrenaline-induced alterations of the action potential and potentiation of tension.
Other interventions which inactivate or block the slow Ca2+ channel (such as
KCl-induced depolarization, long clamp pulses, and clamps to potentials positive to
+ 30 mV) also uniformly reduce or completely suppress the inotropic effect of
adrenaline. On the other hand, in mammalian ventricle it has been observed that the
increase in slow inward current in the presence of adrenaline clearly occurs before
the onset of positive inotropic effect, suggesting that the magnitude of calcium influx
due to the slow inward current in this preparation is not sufficient to potentiate
tension directly, but may contribute to the inotropic effect by facilitating the
releasable Ca2+ stores (Reuter, 1974). Although a similar mechanism may be at play
in the frog ventricular muscle, no evidence of post clamp potentiation, extrasystolic
potentiation, and rate staircase facilitation was found in the presence of adrenaline.
These observations suggest that internal recirculating stores, if present in frog
ventricle, play a minor role in mediating the positive inotropic action of the drug.
Mechanism of the adrenergic response. Adrenaline appears to act at least at two sites
in the frog ventricle: (1) the sarcolemma, where it increases calcium influx by
augmenting the action potential, and (2) the sequestering system, where it enhances
Ca uptake to facilitate relaxation of the myofilaments. The dose-response curve of
adrenaline in the frog ventricle suggests that the effects of catecholamines on the
action potential occur at a lower concentration than the effects on relaxation (Fig.
7). Our results suggest that in the frog ventricle, both the inotropic and relaxant
effects of adrenaline are mediated via the 8-adrenergic receptor. When either
602 M. MORAD, C. SANDERS AND J. WEISS
a-receptor agonists, or a-receptor antagonists in the presence of adrenaline, were
applied to the myocardium, no changes in electrical or contractile activity were
observed. a-adrenergic inotropic effects have been described in the frog atrium
(Niedergerke & Page, 1977). It is difficult at present to account for this discrepancy
with our own findings.
In attempting to evaluate the potential role of cyclic AMP as a mediator of the
adrenergic response in frog ventricle, several inconsistencies in our results became
apparent. Although both cyclic AMP and dibutyryl cyclic AMP mimicked the
positive inotropic effect of adrenaline as well as its effect on the action potential,
neither agent was effective in suppressing KCl-induced contractures or tension
generated by long voltage clamp steps. In the presence of Ni2+, on the other hand,
both cyclic AMP and dibutyryl cyclic AMP reproduced the adrenaline-induced
suppression of twitch tension. Theophylline was most effective in reproducing both
the relaxant and positive inotropic effects of adrenaline on frog ventricular muscle.
In fact, in the presence of theophylline, even cyclic AMP and DB cyclic AMP
demonstrated a relaxant effect on KCl-induced contractures. Since cyclic AMP and
dibutyryl cyclic AMP are reported to mimic the relaxant effects of adrenaline in
mammalian heart (Meinertz, Nawrath & Scholz, 1976, Fabiato & Fabiato, 1975) it
is puzzling why the relaxant effect is not easily demonstrated in the frog heart. Since
the relaxant effect of adrenaline requires a sixty-fold higher concentration to be seen
than the positive inotropic effect, it may be argued that a similar dose differential
may exist for the relaxant effect of cyclic AMP. This argument, however, is not
consistent with the experimental findings of this report, where tenfold increases in
concentration of cyclic AMP or dibutyryl cyclic AMP over the maximal positive
inotropic doses of these agents still did not unmask a relaxant effect. Also, in all
preparations which were exposed to a combination of theophylline and cyclic AMP
or its derivatives, subsequent exposure to isoproterenol caused an increase in
contracture tension and irreversibly damaged the preparation (unpublished obser-
vations). The explanation of this phenomenon in terms of a cyclic AMP-mediated
mechanism of the adrenergic response is not obvious. However, if theophylline acts
primarily as a phosphodiesterase inhibitor, its effectiveness in reproducing both the
positive inotropic and relaxant effects of adrenaline lends strong support for a unitary
cyclic AMP-mediated mechanism of the adrenergic response. Since all of the
pharmacologic experiments above were carried out in reserpinized or fl-blocked
preparations, it is difficult to attribute the actions of theophylline to its potential
effects on endogenous catecholamine stores. Theophylline, however, has been reported
to have actions besides phosphodiesterase inhibition (Johnson & Inesi, 1969;
Kukovetz, Poch & Wurm, 1975), and caution must be exercised in ascribing all of
the effects of theophylline to its influence on intracellular levels of cyclic AMP. It
is conceivable that the relaxant effect of theophylline may result from other as yet
to be described specific consequences of phosphodiesterase inhibition.
We wish to thank Dr James Maylie for help with some of these experiments. Supported by N.I.H.
grants HL16152 and HL17702.
 RELAXANT AND INOTROPIC ACTION OF ADRENALINE 6003
REFERENCES
BASSINGTHWAIGHTE, J. B. & REUTER, H. (1972). Calcium movements in excitation contraction
coupling in cardiac cells. In Electrical Phenomenon in the Heart, ed. DEMELLO, W. C., pp. 353-395.
BEELER, G. W. & REUTER, H. (1970). Membrane calcium current in ventricular myocardial fibres.
J. Physiol. 207, 191-209.
CORABOEUF, E., DISTEL, R. & BOISTEL, J. (1955). Electrophysiologie 616mentaire des tissues
cardiaques. In Colloque International de Microphysiologie du CNRS, Paris-Gif sur Yvette, pp.
123-145. Paris:
FABIATO, A. & FABIATO, F. (1975). Relaxing and inotropic effects of cyclic AMP on skinned cardiac
cells. Nature, Lond. 253, 556-558.
GARGOUIL, Y. M., TRICOCHE, R., FROMENTY, D. & CORABOEUF, E. (1958). Effets de lFadr6naline
sur l'activit, 6lectrique du coeur de Mammiferes. C. r. hebd. Seances Acad. Sci., Paris 246, 334-336.
GOLDMAN, Y. & MORAD, M. (1977a). Measurement of membrane potential and current in cardiac
muscle: A new voltage clamp method. J. Physiol. 268, 613-654.
GOLDMAN, Y. & MORAD, M. (1977b). Ionic membrane conductance during the time course of the
cardiac action potential. J. Physiol. 268, 655-695.
GRAHAM, J. A. & LAMB, J. F. (1967). The effect of adrenaline on the tension developed in
contractures and twitches of the ventricle of the frog. J. Phy8iol. 197, 479-509.
HORA6KOVA, M. & VASSORT, G. (1976). Calcium conductance in relation to contractility in frog
myocardium. J. Physiol. 259, 597-616.
JOHNSON, P. H. & INESI, G. '(1969). The effect of methylxanthines and local anesthetics on
fragmented sarcoplasmic reticulum. J. Pharmac. exp. Ther. 169, 308-314.
KAVALER, F. & MORAD, M. (1966). Paradoxical effects of epinephrine on excitation-contraction
coupling in cardiac muscle. Circulation Res. 18, 492-501.
KUKOVETZ, W. R., P6CH, G. & WURM, A. (1975). Quantitative relations between cAMP and
contraction as affected by stimulators of adenylate cyclase and inhibitors of phosphodiesterase.
Adv. cyclic nucleotide Res. 5, 393-414.
MEINERTZ, T., NAWRATH, H. & SCHOLZ, H. (1976). Possible role of cAMP in the relaxation process
of mammalian heart: Effects of dibutyryl cAMP and theophylline on potassium contractures in
cat papillary muscles. Naunyn-Scmiedebergs Arch. exp. path. Pharmacol. 293, 129-137.
MORAD, M. (1969). Contracture and catecholamines in mammalian myocardium. Science, N. Y. 166,
505-506.
MORAD, M. & GOLDMAN, Y. (1973). Excitation-contraction coupling in heart muscle: Membrane
control of development of tension. Prog. Biophys. molec. Biol. 27, 257-313.
MORAD, M. & KLITZNER, T. (1978). A Ca-transport system for activation of tension in frog
ventricular muscle. In Biophysical Aspects of Cardiac Muscle, ed. MORAD, M., pp. 285-305. New
York: Academic.
MORAD, M. & ORKAND, R. (1971). Excitation-contraction coupling in frog ventricle: Evidence from
voltage clamp studies. J. Physiol. 219, 167-189.
MORAD, M. & ROLETT, E. L. (1972). Relaxing effects of catecholamines on mammalian heart. J.
Physiol. 224, 537-558.
MORAD, M., WEISS, J. & CLEEMANN, L. (1978). The inotropic action of adrenaline on cardiac muscle:
Does it relax or potentiate tension. Eur. J. Cardiol. 7 (suppl.), 53-62.
NIEDERGERKE, R. & PAGE, S. (1977). Analysis of catecholamine effects on single atrial trabeculae
of the frog heart. Proc. R. Soc. B 197, 333-362.
REUTER, H. (1966). Strom-Spannungsbeiziehungen von Purkinje-Fasern bei verschieden extra-
cellularen Calcium-Konzentrationen und unter Adrenalineinwirtung. Pflugers Arch. ges. Physiol.
287, 357-367.
REUTER, H. (1967). The dependence of slow inward current in Purkinje fibers on the extracellular
calcium concentration. J. Physiol. 192, 479-492.
REUTER, H. (1974). Localization of beta adrenergic receptors, and effects of noradrenaline and cyclic
nucleotides on action potentials, ionic currents, and tension in mammalian cardiac muscle. J.
Physiol. 242, 429-451.
REUTER, H. & SCHOLZ, H. (1977a). A study of the ion selectivity and the kinetic properties of the
calcium dependent slow inward current in mammalian cardiac muscle. J. Physiol. 264, 17-47.
604 M. MORAD, C. SANDERS AND J. WEISS
REUTER, H. & SCHOLZ, H. (1977b). The regulation of the calcium conductance of cardiac muscle
by adrenaline. J. Physiol. 264, 49-62.
ROBISON, G. A., BUTCHER, R. W., 0YE, I., MORGAN, H. E. & SUTHERLAND, E. W. (1965). The effect
of epinephrine of adenosine 3-,5--phosphate levels in the isolated perfused rat heart. Molec.
Pharmacol. 1, 168-177.
ROUGIER, O., VASSORT, G., GARNIER, O., GARGOUIL, Y. & CORABOEUF, E. (1969). Existence and
role of a slow inward current during the frog atrial action potential. Pfluigers Arch. 308, 91-1 10.
VASSORT, G., ROUGIER, O., GARNIER, D., SAUVIAT, M. P., CORABOEUF, E. & GARGOUIL, Y. M. (1969).
Effects of adrenaline on membrane inward currents during the cardiac action potential. Pfliiger8
Arch. 309, 70-81.