Cardiovascular Research 40 (1998) 516–522
Effect of 5-HT 4 receptor stimulation on the pacemaker current If in human
isolated atrial myocytes
Roberto Pino a , Elisabetta Cerbai a , Giancarlo Calamai b , Franco Alajmo b , Alberto Borgioli b ,
b
b
b
a,
Lucio Braconi , Massimo Cassai , Gian Franco Montesi , Alessandro Mugelli *
a
Department of Pharmacology, University of Firenze, Viale G.B. Morgagni 65, 50134 Firenze, Italy
b
Department of Cardiosurgery, Azienda Ospedaliera Careggi, Firenze, Italy
Received 2 January 1998; accepted 22 April 1998
Abstract
Objective: 5-HT 4 receptors are present in human atrial cells and their stimulation has been implicated in the genesis of atrial
arrhythmias including atrial fibrillation. An If -like current has been recorded in human atrial myocytes, where it is modulated by
ß-adrenergic stimulation. In the present study, we investigated the effect of serotonin (5-hydroxytryptamine, 5-HT) on If electrophysiological properties, in order to get an insight into the possible contribution of If to the arrhythmogenic action of 5-HT in human atria.
Methods: Human atrial myocytes were isolated by enzymatic digestion from samples of atrial appendage of patients undergoing
corrective cardiac surgery. Patch-clamped cells were superfused with a modified Tyrode’s solution in order to amplify If and reduce
overlapping currents. Results and conclusions: A time-dependent, cesium-sensitive increasing inward current, that we had previously
described having the electrophysiological properties of the pacemaker current If , was elicited by negative steps (260 to 2130 mV) from a
holding potential of 240 mV. Boltzmann fit of control activation curves gave a midpoint (V1 / 2 ) of 288.962.6 mV (n514). 5-HT (1 mM)
consistently caused a positive shift of V1 / 2 of 11.062.0 mV (n58, p,0.001) of the activation curve toward less negative potentials, thus
increasing the amount of current activated by clamp steps near the physiological maximum diastolic potential of these cells. The effect
was dose-dependent, the EC 50 being 0.14 mM. Maximum current amplitude was not changed by 5-HT. 5-HT did not increase If amplitude
when the current was maximally activated by cAMP perfused into the cell. The selective 5-HT 4 antagonists, DAU 6285 (10 mM) and GR
125487 (1 mM), completely prevented the effect of 5-HT on If . The shift of V1 / 2 caused by 1 mM 5-HT in the presence of DAU 6285 or
GR 125487 was 0.361 mV (n56) and 1.060.6 mV (n55), respectively ( p,0.01 versus 5-HT alone). The effect of 5-HT 4 receptor
blockade was specific, since neither DAU 6285 nor GR 125487 prevented the effect of 1 mM isoprenaline on If . Thus, 5-HT 4 stimulation
increases If in human atrial myocytes; this effect may contribute to the arrhythmogenic action of 5-HT in the human atrium.  1998
Elsevier Science B.V. All rights reserved.
Keywords: Human atrial myocytes; Serotonin (5-HT); 5-HT 4 receptor subtype; Arrhythmia (mechanisms); Pacemaker current
1. Introduction
Serotonin (5-hydroxytryptamine, 5-HT) exerts multiple
effects on the cardiovascular system [1]. In the human and
pig atrium, 5-HT causes positive chronotropic, inotropic,
and lusitropic effects [2,3], through stimulation of the
5-HT 4 receptor subtype. The human atrial 5-HT 4 receptor
*Corresponding author. Tel.: 139-(55)-427-1264; fax: 139-(55)-4271285; e-mail: [email protected]
has been recently cloned; it appears to be different from
the 5-HT 4 receptor isoform present in the brain and
gastrointestinal tract [4]. Stimulation of the 5-HT 4 receptor
is associated with the increase in cAMP levels and
activation of cAMP-dependent protein kinase [2,3,5]. One
of the consequences is the 5-HT 4 receptor-mediated increase of L-type calcium current (ICa,L ) which has been
demonstrated in isolated human atrial cardiomyocytes [6].
Time for primary review 21 days.
0008-6363 / 98 / $ – see front matter  1998 Elsevier Science B.V. All rights reserved.
PII: S0008-6363( 98 )00198-9
R. Pino et al. / Cardiovascular Research 40 (1998) 516 – 522
This effect is similar to that exerted by ß-adrenoceptor
stimulation [6], and is the basis for the positive inotropic
response to 5-HT. This positive inotropic effect is, however, observed in the human atrium, but not in the human
ventricle [7]. Furthermore, the cardiac responses to 5-HT
appear to be present only in few mammalian species
(human, pig, monkey) and absent in common laboratory
animals, such as the rat, guinea pig and rabbit [2,6].
Due to its unique location in the human atrium, it has
been suggested that 5-HT 4 receptors may be implicated in
the control of atrial activity [8]. Since 5-HT has been
reported to cause arrhythmias in the isolated human atrium
[9], the suggestion that 5-HT released by platelets may be
involved in the genesis of atrial arrhythmias and particularly of atrial fibrillation [8], appears particularly
appealing. The mechanism for induction of the arrhythmia
could be an intracellular calcium overload resulting in
delayed afterdepolarizations and triggered activity, as
consequence of the activation of the cAMP-dependent
cascade induced by 5-HT through the stimulation of 5-HT 4
receptors and causing an increased calcium entry through
L-type calcium channels [8].
We recently reported that the pacemaker current If is
present in 82% of atrial myocytes isolated from the human
right appendage of patients undergoing cardiac surgery
[10]. The half maximal activation voltage (V1 / 2 ) of If was
around 286 mV which does not support a functional role
for this current under physiological conditions in human
atrial myocytes. In fact, the maximum diastolic potential in
human atrial myocytes has been reported to be generally
below this value (i.e., 270 / 275 mV) [11–16], implying
that the current undergoes little, if any, activation at these
potentials. There are at least two ways by which such a
current could become functionally important. First, its
properties are modified by disease (e.g. atrial dilatation),
similarly to what is observed for other ionic currents
during hypertrophy or failure [17–22], or chronic atrial
fibrillation [23], in such a way that it may become active
under pathological conditions. A second possibility is that
If activation curve is shifted toward less negative (and
more physiological) potentials by the stimulation of relevant receptors. Recently, it was found that the V1 / 2 of If in
human atrial myocytes was significantly shifted toward
less negative potentials by ß-adrenoceptor stimulation with
isoprenaline [10]. In rabbit sino-atrial node cells it has
been clearly demonstrated that the effect of cAMP on If is
largely direct and does not involve the phosphorylation of
the channel [24,25]. It is likely that a similar mechanism is
operative in the human heart; direct proof of the mechanism by which cAMP regulates If in the human atria is not
available.
To get an insight into the possible contribution of If to
the arrhythmogenic action of 5-HT in human atria, we
thought that it would be important to know if 5-HT,
through the stimulation of 5-HT 4 receptors, was able to
affect If electrophysiological properties (amplitude and
517
activation voltage), and if its effects on If were observed at
the concentrations previously demonstrated to increase
ICa,L .
Preliminary data have been presented in abstract form
[26].
2. Methods
2.1. Cell isolation and storage
The procedure used was similar to that described
previously [10]. Small samples from human right atrial
appendages were used to isolate single myocytes. The
specimens, obtained from the tissue routinely excised
during the preparation of extra-corporeal circulation, were
brought to the laboratory in cold transport saline (see
Section 2.4). In the laboratory the sample was placed in
Ca 21 -free oxygen-saturated solution. Epicardial fat and
connective tissue were removed from the muscle specimen. The tissue samples were cut into small pieces of
about 1 mm 3 in size and gently stirred in a thermostated
(358C) Ca 21 -free oxygen-saturated solution for 20 min in a
tissue dissociation vessel [27,28]. Cells were isolated by
enzymatic dissociation in Ca 21 -free solution plus collagenase type V (350 I.U. ml 21 , Sigma) and protease type XXIV
(4 I.U. ml 21 , Sigma) for 20 min. After this step was
completed, the samples were stirred for 20 min in a
Ca 21 -free solution containing collagenase type V (350
I.U. ml 21 ). The solution was then replaced by a new
enzymatic solution where the collagenase concentration
was reduced to 175 I.U. ml 21 . During this final step, the
supernatant was replaced by fresh collagenase solution
every 15 min and the dissociation was followed by
microscopic examination of the collected medium. Cells
¨ solution and allowed to
were then stored in a Kraft–Bruhe
sediment under gravity for 15 min or with gentle centrifugation (5 min at 1000 rpm). The supernatant was removed
and replaced by Ca 21 -free Tyrode’s solution. The Ca 21
concentration was raised stepwise (50, 100, 200, 500 mM).
The isolated cells were kept at room temperature in
Tyrode’s solution containing penicillin (50 I.U. ml 21 ) and
streptomycin (50 I.U. ml 21 ) and used within the day.
2.2. Patients
Tissue specimens were obtained from 18 patients of
both sexes (age: 67610 years, mean6S.D.) undergoing
cardiac surgery, for cardiac vascular disease (67%) or
aortic valve replacement (33%). All patients were in sinus
rhythm; they were under treatment with various combinations of drugs, including Ca 21 -antagonists, acetylsalicylic
acid and nitrate compounds. Our previous results [10]
suggested that the properties of If were not influenced by
the underlying pathology or pharmacological treatment in
518
R. Pino et al. / Cardiovascular Research 40 (1998) 516 – 522
any evident manner. However, patients chronically treated
with ß-adrenergic blockers were excluded due to the
reported sensitization of human atrial 5-HT 4 receptors
induced by such therapy [29]. All patients gave their
informed consent for the use of their tissue samples. The
design of this study conforms to the Helsinki declaration
[30] and was approved by the local ethical committee.
2.3. Electrophysiological recordings
The patch clamp technique (whole cell recording) was
used to measure the electrophysiological properties of the
isolated human atrial myocytes. Details of the equipment
have been previously described [10]. Cells were placed in
an experimental bath on the platform of an inverted
microscope (Nikon Diaphot TMD, Japan). Experiments
were performed using a patch amplifier (Axopatch-1B,
Axon Instruments, CA, USA) interfaced to a personal
computer by means of a general purpose DAC /ADC
interface (Labmaster Tekmar, Scientific Solutions). Data
were viewed on-line on an analogic oscilloscope and on
the computer screen. Data acquisition and preliminary
analysis were performed by means of the integrated
software package PCLAMP (Axon Instruments). Cells were
superfused with a normal Tyrode’s solution (see Section
2.4); a modified Tyrode’s solution was used during
measurements of the hyperpolarization-activated inward
current (If ). Temperature was maintained in the range
36–378C. Patch-clamp pipettes were prepared from glass
capillary tubes (Garner, CA, USA) by means of a twostage vertical puller (Hans Otchoski, Hamburg, Germany)
and fire-polished just before use. Pipettes had a resistance
of 1.5–2.5 MV when filled with the internal solution.
For the experiments with drug application, the patchclamped cell was superfused by means of a temperaturecontrolled micro-superfusor which allowed rapid changes
of the solution bathing the cell; this enabled us to minimize
current run-down and to reduce exposure of the other cells
present in the experimental chamber to drugs. 5-HT was
dissolved in distilled water to get a stock solution with a
final concentration of 10 mM containing 1 mg / ml ascorbic
acid. The stock solution was then diluted with Tyrode’s
solution to get the final 5-HT concentrations.
The selective 5-HT 4 antagonists, DAU 6285 (N-endo-8methyl-8-azabicyclol[3.2.1]-oct-3-yl)-2-3-dihydro-2-oxo1H-benzimidazole-1-carboxamide,hydrochloride) [31], and
GR
125487
([1-[2-[methylsulphonylamino]-ethyl]-4piperidinyl]-methyl-1-methyl-1H-indole-3-carboxylate) [32,33], were dissolved in distilled water to get the stock
solutions (both 10 mM) and then diluted in Tyrode’s
solution up to the final concentrations. In the experiments
aimed to test the role of cAMP in mediating 5-HT effect,
the tip of the patch pipette was filled with the internal
solution and then backfilled with the same solution also
containing 200 mM cAMP (Sigma).
2.4. Solutions
The composition of the solutions employed was as
follows (in mM): Transport saline: Lactic acid 286; NaCl
103; KCl 5.4; CaCl 2 1.8; pH adjusted to 7.0 with NaOH.
Isolation solution: NaCl 120, KCl 10, KH 2 PO 4 1.2, MgCl 2
1.2, D-(1)-glucose 10, Hepes 10, taurine 20; pH adjusted
¨ solution: KCl 85; K 2 HPO 4
to 7.2 with NaOH. Kraft-Bruhe
30; MgSO 4 5.0; glucose 20; pyruvic acid 5.0; creatine 5.0;
taurine 5.0; EGTA 0.5; b-hydroxybutyric acid 5.0; succinic acid 5.0; K 2 -ATP 2.0; pH adjusted to 7.0 with KOH.
Normal Tyrode’s solution: NaCl, 140; KCl, 5.4; CaCl 2 ,
1.0; MgCl 2 , 1.2; Hepes–NaOH pH 7.35, 5.0; glucose, 10.
Modified Tyrode’s solution for hyperpolarization-activated
inward current measurements: NaCl, 140; KCl 25; CaCl 2 ,
1.0; MgCl 2 , 1.2; Hepes–NaOH pH 7.35, 5.0; glucose, 10;
BaCl 2 1.0 (to block inward rectifier-like channels); MnCl 2
2.0 (to block ICa,L ); 4-aminopyridine 0.5 (to block transient
outward potassium current, Ito ). Internal solution (used to
fill the patch micro-pipettes): KCl 110, ATP/ K 1 4.3,
MgCl 2 2.0, CaCl 2 1.0, EGTA 11, Hepes / K 1 10, adjusted
to pH 7.4 with KOH (free [Ca 21 ] ,10 28 M).
2.5. Data analysis and statistics
Data analysis and fitting were performed by using the
4.1 program (MicroCal, MA, USA) running on a
Pentium personal computer. For fitting functions, nonlinear models of convergence to solutions were used.
To evaluate steady-state values of the hyperpolarizationactivated current, data were fitted to an exponential decay,
which gave the time constant of If activation (t ). The
difference between the extrapolated value and that of the
current at the end of the hyperpolarization step was usually
small (in the order of a few pA).
A Boltzmann model based on the partition theorem
according to the general equation:
ORIGIN
1
y 5 ]]]]
(V2V1 / 2 ) / k
11e
was fitted to the activation data, where V (mV) is the test
membrane potential, V1 / 2 (mV) is the fitted potential for
half activation and k (mV) is related to the slope of the
activation curve.
Cell membrane capacitance (Cm ) was measured by
applying a pulse of small amplitude (DV510 mV) starting
from a holding potential of 270 mV. The current transient
following this clamp protocol was fitted with a monoexponential model to compute series resistance (R s ) and
then Cm using the two equations given below:
DVm
3
R s 5 ]] ? 10
Ipeak
and
R. Pino et al. / Cardiovascular Research 40 (1998) 516 – 522
519
t
Cm 5 ]
Rs
where Ipeak is the maximum level of current (relative to the
holding current) following the depolarization and t is the
time constant of the exponential current decay. The
membrane capacitance values obtained have been used to
compute ionic current densities (If density in pA / pF).
Dose-effect curves were fitted by using the Hill equation:
x
y 5 Emax ? ]]
k1x
where Emax is the maximum effect, k corresponds to the
concentration at which 50% Emax is obtained (EC 50 ) and x
is the concentration of drug.
Given the relatively small number of observations and
the nature of the measured data that are likely to be
non-normally distributed, non-parametric statistics (Wilcoxon) on paired data (one- or two-tail, according to the
need) were used. All results are reported as
means6S.E.M.; p,0.05 was considered significant.
3. Results
Cells isolated from the right atrial appendages of human
hearts were relatively small, with an average membrane
capacitance (Cm ) of 62.163.6 pF (n5112). These data are
consistent with the previously reported values for capacitance and geometrical dimensions of the cells [10]. As
expected, following clamp steps to hyperpolarized membrane potentials, a time-dependent inward current was
present in about 80% of the cells; since we have previously
documented that this current has the electrophysiological
properties of If [10], we will call it If .
Fig. 1 (panel A) shows the hyperpolarization-activated
current (If ) recorded from a human atrial myocyte bathed
in the modified Tyrode’s solution used to amplify If [10].
The current traces were evoked by steps at potentials in the
range of the diastolic potential of normally polarized atrial
myocytes. After superfusion with 1 mM 5-HT (panel B),
the amplitude of the current at each potential was clearly
increased. The activation curve obtained by plotting current amplitude vs. membrane potential was fitted by a
Boltzmann function whose average midpoint was
288.962.6 mV (n514), and the average current density,
measured at 2120 mV, was 4.362.0 pA / pF. Panel C of
Fig. 1 shows that superfusion with 5-HT (1 mM) did not
affect maximum current amplitude, but shifted the activation curve of If toward less negative potentials. The mean
maximum current amplitude was 300646 pA in controls
(n58) and 275641 pA after 5-HT (n58); also the slope
factor k was not significantly changed being 10.461.2 mV
in controls and 10.161.3 mV in the presence of 5-HT. The
shift of the activation curve caused by 5-HT was 11.062.0
Fig. 1. Effect of 5-HT on If in single human atrial myocytes. (A and B)
Current traces recorded during steps to the potential indicated near each
trace, in the absence (A) and presence (B) of 1 mM 5-HT. Holding
potential5 240 mV; (C) corresponding I–V relationships for If , obtained
by plotting If amplitude versus the step potential (s5control; d51 mM
5-HT). Here and in the next figures, lines were obtained by fitting the
experimental data points with a Boltzmann function (see Section 2). (D)
Plot of time-constant reciprocals (1 /t ) versus membrane potential (s5
control; d51 mM 5-HT).
mV (n58, p,0.001). In the presence of 5-HT the activation kinetics of If was consistently faster at every potential
(Fig. 1, panel D).
Fig. 2 shows that the shift of the activation curve of If ,
measured with different concentrations of 5-HT, was dosedependent. For comparison the dose–response curve of the
effect of 5-HT on the shift of the voltage–activation
relationship for If (present results) is shown together with
that derived from the data regarding the effects of 5-HT on
ICa,L amplitude [6]. It is apparent that the two curves are
practically superimposable, the EC 50 for the effect on If
Fig. 2. Concentration–effect curve for the shift of the I–V relationship of
If (DV1 / 2 ) by 5-HT. Closed circles represent mean6S.E.M. of values
obtained in human atrial myocytes; each cell was exposed to a single
5-HT concentration. The number of cells tested is indicated in parenthesis. Open circles represent data derived from Fig. 1 of [6]. Lines were
obtained by fitting data with a Hill equation, which gave an EC 50 of 0.14
mM and 0.13 mM for the effect on If (solid line) and ICa,L (dashed line),
respectively.
520
R. Pino et al. / Cardiovascular Research 40 (1998) 516 – 522
being 0.14 mM, and that for the effect on ICa,L being 0.13
mM; in both cases the maximal effect of 5-HT was reached
at the concentration of 1 mM. The increase in ICa,L caused
by 5-HT in human atria involves an increase in intracellular cAMP [6].
Fig. 3 suggests that a cAMP dependent pathway is also
involved in the effect of 5-HT on If . In fact, increasing
intracellular cAMP by adding it to the patch pipette at a
concentration of 200 mM, causes a marked increase in If
amplitude (panel A). Panel B shows the time-course of the
cAMP effect: after rupturing the seal and dialysing the
intracellular milieu with cAMP, If amplitude increases
rapidly, reaching the steady-state in about 2 min (panel B).
When the maximal effect is reached, i.e. after 3 min of cell
dialysis, superfusion of the cell with a solution containing
5-HT (1 mM), does not cause any further increase in the
current amplitude. This is clearly shown in panel A, where
the current traces recorded before and after 5-HT are
superimposable and in panel B which clearly demonstrates
that superfusion with 5-HT (arrow) does not change the
amplitude of If , which remains stable over time. Similar
results were obtained in five cells. Panel C of Fig. 3 shows
the current–voltage relationship obtained in the same
myocyte before and after diffusion of cAMP into the cell,
in the presence or in the absence of 5-HT. It is apparent
that 5-HT does not exert any effect on the current–voltage
curve when intracellular cAMP is increased. To demonstrate that the effect of 5-HT on If was due to the
stimulation of the 5-HT 4 receptor subtype, cells were
exposed to the selective 5-HT 4 antagonist DAU 6285 [34].
As shown in Fig. 4 (upper panels), exposure to DAU
6285 (10 mM) for 2 min completely prevented the effect of
5-HT on If amplitude. If -current traces recorded before and
Fig. 3. Role of cAMP in mediating 5-HT effects on If . (A) current traces
recorded during a step to 270 mV, immediately after rupturing the
membrane under the patch pipette (s), at the steady state, i.e., 3 min after
seal rupture (h, 3 min), and after addition of 1 mM 5-HT to the external
solution (d). (B) time-course of the effect of cAMP on If amplitude and
lack of effect of 5-HT under this condition. Time 0 refers to the moment
at which the seal was disrupted; arrow indicates the start of 5-HT
superfusion. (C) Corresponding activation curves obtained before and
after dialysing the cell with cAMP (s and h, respectively) and in the
presence of 5-HT (d).
Fig. 4. 5-HT 4 receptor blockade by DAU 6285 prevents the effect of
5-HT on If . (A and B): current traces recorded during steps to the
potential indicated near each trace, in the absence (A) and presence (B) of
1 mM 5-HT, in a myocyte pretreated with 10 mM DAU 6285. Holding
potential5 240 mV; (C) corresponding I–V relationships for If , obtained
by plotting normalized If amplitude versus the step potential (s510 mM
DAU 6285; d5DAU 6285 plus 1 mM 5-HT; - - -5control curve).
after application of 5-HT 1 mM (in the presence of DAU
6285) during steps at 260, 270, and 280 mV were
identical. The activation curves obtained in control conditions, in the presence of DAU 6285 alone and of DAU
6285 plus 5-HT were fully superimposable. A similar
effect was observed with the more selective 5-HT 4 antagonist GR 125487 [32,33] (data not shown).
Fig. 5 summarizes the results obtained with the two
5-HT 4 antagonists. Both DAU 6285 and GR 125487
completely prevented the shift of the If activation curve
Fig. 5. Specificity of 5-HT 4 receptor blockade on the serotonin-induced
shift of If activation curve (DV1 / 2 ). (A) Summary of the effect of 5-HT on
If in the absence and presence of the 5-HT 4 selective antagonists, DAU
6285 or GR 125487. Each column represents the mean6S.E.M. of the
shift in V1 / 2 ; the number of cells tested is indicated in parenthesis,
* 5p,0.05 versus 5-HT alone. (B, C) Traces recorded during hyperpolarizing steps to the potential indicated near each trace, in the
absence (B) and presence (C) of 1 mM isoprenaline, in a myocyte
pretreated with 10 mM DAU 6285. Holding potential 5 240 mV.
R. Pino et al. / Cardiovascular Research 40 (1998) 516 – 522
caused by 5-HT: in fact, the shift of V1 / 2 caused by 1 mM
5-HT in the presence of 10 mM DAU 6285 or 1 mM GR
125487 was 0.361 mV (n56) and 1.060.6 mV (n55),
respectively ( p,0.01 versus 5-HT alone). The effect of
5-HT 4 receptor blockade was specific as demonstrated in
Fig. 5. In fact, exposure of the cells to DAU 6285 did not
prevent the effect of isoprenaline (ISO). A 1 mM concentration of ISO which increases If amplitude in human
atrial myocytes [10], also increased If amplitude in the
presence of the 5-HT 4 antagonist DAU 6285. Similar
results were obtained with the 5-HT 4 antagonist GR
125487 (not shown).
4. Discussion
The present results demonstrate that the pacemaker
current If in human atrial myocytes is modulated by 5-HT.
5-HT increases If amplitude through the stimulation of the
5-HT 4 receptor subtype and involves the activation of the
cAMP pathway. The effect of 5-HT is completely prevented by the selective antagonists of the 5-HT 4 receptor
subtype, GR 125487 and DAU 6285. These compounds
have a selectivity at least 30-fold higher versus the 5-HT 4
receptor subtype compared to other 5-HT receptor subtypes [31,32,34]. Furthermore, it has been clearly demonstrated that the 5-HT 4 subtype is the only 5-HT receptor
functionally present in the human atrium [4,9]. The
mechanism by which 5-HT 4 stimulation increase If amplitude is by shifting its activation voltage toward less
negative values. The voltage of half maximal activation for
If , i.e. the voltage at which the current is activated at half
of its maximal value, in the human atrial cells is approximately 290 mV. This property makes unlikely a functional
contribution of If under physiological conditions, since the
maximum diastolic potential of human atrial cells is rarely
so negative [11–16]. However, a shift of the activation
voltage of If toward less negative values may completely
change the scenario. Let us suppose that a shift occurs, as
experimentally observed with 5-HT in our study, of 11 mV
toward less negative values: this means that at 270 mV (a
likely maximum diastolic potential for an atrial cell) a
substantial fraction of the current (|40%) will now be
activated and possibly could play a functional role, i.e.
leading to the development of a diastolic depolarization
phase. The effect of 5-HT is similar to that previously
observed with ß-adrenoceptor stimulation [10]. This is also
confirmed in the present study; in fact when 5-HT 4
receptors are blocked, ISO causes an increase in If
amplitude, which is indistinguishable from that caused by
5-HT. Thus, as for the effect of ß-adrenoceptor stimulation,
it is likely that the effect occurs through the activation of
the adenylyl cyclase pathway, leading to an increase in
cAMP, that is the same pathway involved in the 5-HT 4
mediated increase in ICa,L [6]. This is strongly suggested
also by the observed increase of If amplitude when the
521
cAMP is diffused into the cell and by the lack of effect of
5-HT, when If is maximally activated by intracellular
cAMP. However the mechanisms by which cAMP regulates ICa,L and If are likely to differ. In fact the effect of
cAMP on If has been demonstrated to be largely direct and
independent of channel phosphorylation [24,25]. It is likely
that such a mechanism could regulates If also in human
atria, even if direct data are not available. It is worth
noting that the EC 50 for the increase in ICa,L and If
amplitude are similar ( [6] and present results). The
implications of these findings are quite interesting: a
certain concentration of 5-HT able to stimulate the 5-HT 4
receptor, will cause a concomitant increase in both ICa,L
and If in atrial cells. These two effects may cooperate in
the arrhythmogenesis of the human atrium. In fact, as
suggested by Kaumann [8], the increase in calcium current
may lead to calcium overload, causing the transient inward
current which underlies delayed afterdepolarizations
(DADs) [35]. Epinephrine-induced DADs have been recorded in human atrial trabeculae [36]. DADs are a well
known arrhythmogenic mechanism [37]; if their amplitude
is sufficient they may reach the threshold giving rise to a
spontaneous action potential (triggered activity). The
increase in If amplitude may contribute to this phenomenon. In fact, if If is responsible for a diastolic depolarization, an increase in its amplitude can lead to a diastolic
depolarization which, in its turn, may favor the attainment
of threshold by a subthreshold DAD, thus causing an
extrasystole, which may trigger the arrhythmia. In the
presence of an arrhythmogenic substrate, such as a dilated
atrium of an old patient [38], atrial fibrillation could
develop.
Acknowledgements
Supported in part by a grant from MURST (University
of Firenze). We wish to thank Dr. Carla Ghelardini and Dr.
Nicoletta Galeotti for useful discussion on 5-HT 4 receptor
pharmacology.
References
[1] Villalon CM, de Vries P, Saxena PR. Serotonin receptors as
cardiovascular targets. Drug Discovery Today 1997;2:294–300.
[2] Kaumann AJ, Sanders L, Brown AM, Murray KJ, Brown MJ. A
5-hydroxytryptamine receptor in human atrium. Br J Pharmacol
1990;100:879–885.
[3] Kaumann AJ. 5-HT 4 -like receptors in mammalian atria. J Neural
Transm 1991;34(suppl):195–201.
[4] Blondel O, Vandecasteele G, Gastineau M, et al. Molecular and
functional characterization of a 5-HT 4 receptor cloned from human
atrium. FEBS Lett 1997;412:465–474.
[5] Sanders L, Kaumann AJ. A 5-HT 4 -like receptor in human left
atrium. Naunyn Schmiedebergs Arch Pharmacol 1992;345:382–386.
[6] Ouadid H, Seguin J, Dumuis A, Bockaert J, Nargeot J. Serotonin
increases calcium current in human atrial myocytes via the newly
522
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
R. Pino et al. / Cardiovascular Research 40 (1998) 516 – 522
described 5-hydroxytryptamine-4 receptors. Mol Pharmacol
1992;41:346–351.
Jahnel U, Rupp J, Ertl R, Nawrath H. Positive inotropic response to
5-HT in human atrial but not in ventricular heart muscle. Naunyn
Schmiedebergs Arch Pharmacol 1992;346:482–485.
Kaumann AJ. Do human atrial 5-HT 4 receptors mediate arrhythmias? Trends Pharmacol Sci 1994;15:451–455.
Kaumann AJ, Sanders L. 5-Hydroxytryptamine causes rate-dependent arrhythmias through 5-HT 4 receptors in human atrium: facilitation by chronic b-adrenoceptor blockade. Naunyn Schmiedebergs
Arch Pharmacol 1994;349:331–337.
Porciatti F, Pelzmann B, Cerbai E, et al. The pacemaker current If in
single human atrial myocytes and the effect of b-adrenoceptor and
A 1 -adenosine receptor stimulation. Br J Pharmacol 1997;122:963–
969.
Bush HL, Gelband H, Hoffman BF, Malm JR. Electrophysiological
basis for supraventricular arrhythmias. Arch Surg 1971;103:620–
624.
Escande D, Loisance D, Planche C, Coraboeuf E. Age-related
changes of action potential plateau shape in isolated human atrial
fibers. Am J Physiol 1985;249:H843–H850.
Gelband H, Bush HL, Rosen MR, Myerburg RJ, Hoffman BF.
Electrophysiologic properties of isolated preparations of human
atrial myocardium. Circ Res 1972;30:293–300.
Singer DH. Automaticity in human cardiac tissue. In: Rosen MR,
Janse MJ, Wit AL, editors. Cardiac electrophysiology: a textbook.
Mount Kisco, NY: Futura, 1990:247–263.
Ten Eick RE, Singer DH. Electrophysiological properties of diseased human atrium: Low diastolic potential and altered cellular
response to potassium. Circ Res 1979;44:545–557.
Wang Z, Fermini B, Nattel S. Delayed rectifier outward current and
repolarization in human atrial myocytes. Circ Res 1993;73:276–285.
Beuckelmann DJ, Nabauer M, Erdmann E. Alterations of K 1
currents in isolated human ventricular myocytes from patients with
terminal heart failure. Circ Res 1993;73:379–385.
Cerbai E, Barbieri M, Mugelli A. Characterization of the hyperpolarization-activated current, If , in ventricular myocytes isolated
from hypertensive rats. J Physiol (Lond) 1994;481:585–591.
Cerbai E, Barbieri M, Li Q, Mugelli A. Ionic basis of action
potential prolongation of hypertrophied cardiac myocytes isolated
from hypertensive rats of different ages. Cardiovasc Res
1994;28:1180–1187.
Cerbai E, Barbieri M, Mugelli A. Occurrence and properties of the
hyperpolarization-activated current If in ventricular myocytes from
normotensive and hypertensive rats during aging. Circulation
1996;94:1674–1681.
Cerbai E, Pino R, Porciatti F, et al. Characterization of the
hyperpolarization-activated current, If , in ventricular myocytes from
human failing heart. Circulation 1997;95:568–571.
Nuss HB, Houser SR. T-type Ca 21 current is expressed in hy-
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
pertrophied adult feline left ventricular myocytes. Circ Res
1993;73:777–782.
Van Wagoner D, Pond AL, McCarthy PM, Trimmer JS, Nerbonne
JM. Outward K 1 current densities and Kv1.5 expression are reduced
in chronic human atrial fibrillation. Circ Res 1997;80:772–781.
Accili EA, Redaelli G, DiFrancesco D. Differential control of the
hyperpolarization-activated current (i(f)) by cAMP gating and
phosphatase inhibition in rabbit sino-atrial node myocytes. J Physiol
(Lond) 1997;500:643–651.
DiFrancesco D, Tortora P. Direct activation of cardiac pacemaker
channels by intracellular cyclic AMP. Nature 1991;351:145–147.
Pino R, Cerbai E, Alajmo F, et al. Effect of 5-hydroxytryptamine
(5-HT) on the pacemaker current If , in human atrial myocytes
(Abstract). Circulation 1996;94:2770.
Jacobson SL, Altschuld RA, Hohl H. Muscle cell cultures from
human heart. In: Piper HM, editor. Cell culture techniques in heart
and vessel research. Springer, Berlin / Heidelberg / New York,
1990:75–98.
¨
Pelzmann B, Schaffer P, Machler
H, Rigler B, Koidl B. Adenosine
inhibits the L-type calcium current in human atrial myocytes.
Naunyn Schmiedberg’s Arch Pharmacol 1995;351:293–297.
Sanders L, Lynham JA, Bond B, et al. Sensitization of human atrial
5-HT 4 receptors by chronic beta-blocker treatment. Circulation
1995;92:2526–2539.
World Medical Association Declaration of Helsinki. Recommendations guiding physicians in biomedical research involving human
subjects. Cardiovasc Res 1997;35:2–3.
Schiavone A, Giraldo E, Giudici L, Turconi M, Sagrada A. DAU
6285: a novel antagonist at the putative 5-HT 4 receptor. Life Sci
1992;51:583–592.
Schiavi GB, Brunet S, Rizzi CA, Ladinsky H. Identification of
serotonin 5-HT 4 recognition sites in the porcine caudate nucleus by
radioligand binding. Neuropharmacology 1994;33:543–549.
Ghelardini C, Galeotti N, Casamenti F, et al. Central cholinergic
antinociception induced by 5HT 4 agonists: BIMU 1 and BIMU 8.
Life Sci 1996;58:2297–2309.
Dumuis A, Golzan H, Sebben M, et al. Characterization of a novel
5-HT 4 receptor antagonist of the azabicycloalkyl benzimidazolone
class: DAU 6285. Naunyn Schmiedebergs Arch Pharmacol
1992;345:264–269.
Vassalle M, Mugelli A. An oscillatory current in sheep cardiac
Purkinje fibers. Circ Res 1981;48:618–631.
Hou ZY, Lin CI, Vassalle M, Chiang BN, Cheng KK. Role of
acetylcholine in induction of repetitive activity in human atrial
fibers. Am J Physiol 1989;256:H74–H84.
Wit AL, Rosen MR. Afterdepolarization ant triggered activity. In:
Fozzard H, Heber E, Jennings R, editors. The heart and cardiovascular system. New York: Raven Press, 1986:1449–1491.
Narayan SM, Cain ME, Smith JM. Atrial fibrillation. Lancet
1997;350:943–950.