Seizure 2002; 11: 285–302

doi:10.1053/seiz.2001.0644, available online at http://www.idealibrary.com on

Anticonvulsant profile and mechanism of action of

propranolol and its two enantiomers

W. FISCHER

Rudolf-Boehm-Institute of Pharmacology and Toxicology, University of Leipzig, Härtelstraße 16-18,

Leipzig, Germany

Correspondence to: Wolfgang Fischer, Rudolf-Boehm-Institute of Pharmacology and Toxicology, University of Leipzig,
Härtelstraße 16-18, D-04107 Leipzig, Germany. E-mail: fisw@medizin.uni-leipzig.de

The anticonvulsant properties of the ß-adrenoceptor antagonist propranolol and its two enantiomers were examined in various
screening tests in order to characterize the anticonvulsant profile as well as the possible molecular mechanism of action.
These compounds dose-dependently raised the threshold for tonic electroshock seizures in mice and were effective in the
traditional maximal electroshock test (ED50 s 15–20 mg kg−1 i.p.). In combination with clinically used antiepileptics, the
anticonvulsant effectiveness of the latter was significantly increased. In the pentylenetetrazol (85 mg kg−1 s.c.) seizure threshold
test, (±)- and (+)-propranolol were not effective in preventing clonic seizures. In unrestrained rats with chronically implanted
electrodes in the dorsal hippocampus, propranolol and its (+)-enantiomer equieffectively reduced the duration of electrically-
evoked hippocampal afterdischarges (10 and 20 mg kg−1 i.p.) and raised the focal stimulation threshold (20 mg kg−1 i.p.).
In amygdala-kindled rats, both drugs (≥10 mg kg−1 i.p.) reduced the seizure severity from stage 5 (generalized clonic–tonic)
to stage 3 (unilateral forelimb) seizures. Furthermore, whole-cell patch-clamp experiments showed that (+)- as well as (−)-
propranolol (10−6 to 10−4 M) depressed the fast inward sodium current in a concentration- and use-dependent manner in
cultured rat cardiomyocytes and inhibited picrotoxin-induced burst firing activity of mouse spinal cord neurones in culture. In
conclusion, propranolol and its two enantiomers have anticonvulsant effects in models for generalized tonic–clonic and complex
partial seizures which may be accounted for by the sodium channel blocking and not by the ß-adrenoceptor blocking activity.
c 2002 Published by Elsevier Science Ltd on behalf of BEA Trading Ltd.

Key words: propranolol; (+)- and (−)-propranolol; anticonvulsants; seizure models; whole-cell patch-clamp; sodium channel.

INTRODUCTION and carbamazepine block high-frequency repetitive

Propranolol is a prototype, non-selective ß1 /ß2 -
adrenoceptor antagonist, having an established indica-
tion in a variety of cardiovascular diseases including
essential hypertension, cardiac arrhythmias and angina
pectoris for more than 30 years 1, 2 . It is known that this
lipophilic drug exhibits marked central actions, such
as anxiolytic, antiaggressive and antipsychotic effects,
with possible indications in neurology and psychia-
try 3–5 . Interestingly, pronounced anticonvulsant prop-
erties in various seizure models were also documented
(for review, see Reference 6). Propranolol potently
prevents maximal electroshock (MES) seizures in
rodents and raised the after discharge threshold in
rats 7, 8 . These basic observations indicate that the
profile of anticonvulsant activity seems to be similar
to that of phenytoin and carbamazepine, established
antiepileptic drugs in the treatment of generalized
tonic–clonic and partial seizures in humans. Phenytoin
firing of action potentials and sodium channels are
the likely molecular targets of these antiepileptics 9, 10 .
Some authors described anticonvulsant effects of
propranolol against sound-induced seizures in DBA/2
mice or pentylenetetrazol (PTZ)-induced convulsions
in rats, in which the (−)-enantiomer was more potent
than the (+)-enantiomer 11, 12 . However, the majority
of other studies in several animal models suggests
that the anticonvulsant action of propranolol does not
result from ß-adrenoceptor blockade 13, 14 and may be
due to the ‘membrane stabilizing’ properties of this
drug 8, 15 .
Anecdotal reports show that propranolol may be
useful for treatment of seizures in humans 16, 17 .
In particular, this substance might also be used in
patients suffering from epilepsies associated with
increased blood pressure and cardiac dysfunctions.
The objective of the present work was to determine the
antiepileptic profile of propranolol and the mode of ac-

1059–1311/02/$22.00/0 c 2002 Published by Elsevier Science Ltd on behalf of BEA Trading Ltd.


286
tion in a series of appropriate screening models. Apart
from various, mostly not very extensive, studies using
electroshock- and pentylenetetrazol-induced seizure
tests, detailed investigations with different seizure
threshold and chemoshock tests as well as complex
partial seizure models like the amygdala kindling
are missing (see Reference 6). In order to clarify
whether the blockade of ß-adrenoceptors is involved
in the mechanism of action, experiments with (−)-
and (+)-propranolol (the latter is essentially devoid
of ß-adrenolytic effects but shares the membrane
stabilizing properties of the (−)-enantiomer) were
performed. Since propranolol may be useful as an
adjunctive drug in epilepsies, the modulation of the
protective efficacy of conventional antiepileptic drugs
by combined treatment was also analysed. Finally, the
inhibitory effects of propranolol on the fast inward
sodium current, which could be inferred from previous
studies, were examined in detail using whole-cell
patch-clamp technique.
METHODS
Animals
In most of the screening tests (MES, PTZ, Ro-
tarod; see later) male albino mice (strain 01,
Leipzig-Probstheida, Germany), weighing 19–25 g
(27- to 32-day-old) were used. Two special seizure
models (maximal N-methyl-DL-aspartate (NMDLA)
and quinolinate (QUIN) seizure test) were carried out
on male SHR mice (bred from Swiss strain, Rappolovo
farm, St Petersburg, Russia), weighing 18–22 g. (For
the patch-clamp experiments on spinal cord neurones,
foetal 12- to 13-day-old NMRI-Han mice (Hanover,
Germany) were used.) After adaptation to laboratory
conditions for some days, the experimental groups
were chosen by means of a randomized schedule.
Each mouse was used only for one experiment. The
EEG-studies (hippocampal afterdischarges) as well as
the pharmacokinetic investigations were carried out
with male Wistar rats (own breeding stock, formerly
‘Jelei: WIST’), weighing 250–300 g at the time
of surgery. The studies with amygdala-kindled rats
were performed on female Wistar rats (strain ‘Shoe:
WIST’, Schönwalde, Germany), weighing 150–220 g
at the beginning of the experiments. The animals
were kept in colony cages under standard laboratory
conditions on a natural light–dark cycle with free
access to commercial food pellets and tap water. The
screening experiments were carried out between 9
and 12 hours to avoid circadian influences. Animal
care and handling was conducted in compliance with
the German Animal Welfare Act and was approved by
the relevant local governmental body in Leipzig.
W. Fischer
Maximal electroshock seizure threshold (MES-T)
test (mice)
The stimulus train was applied via ear-clip electrodes
(sinusoidal pulses 5–10 mA, 50 Hz, 0.2 seconds
duration) by means of a constant current stimulator
(rodent-shocker type 221; Hugo Sachs Elektronik,
March-Hugstetten, Germany). The stimulus intensity
was varied by an up-and-down method in which the
current was raised or lowered in 1 (or 2) nA-steps
if the preceding animal did not or did show tonic
hindlimb extension, respectively (see Reference 18).
Groups of 15–20 mice were used. Current intensity-
effect curves were constructed on the basis of the
percentage of animals responding with the endpoint
at the corresponding current value. The calculation
of CC50 values (current intensity in mA, necessary
to induce tonic hindlimb extension in 50% of the
mice tested) and the statistical comparisons were
performed using a computer-supported probit analysis
according to the method of Litchfield and Wilcoxon 19 .
In the case of multiple comparisons between different
drug-treated and the corresponding control group,
α-correction was performed.
Maximal electroshock seizure test (mice)
Maximal electroshock seizure was induced in mice
via ear-clip electrodes by a constant suprathreshold
current (rectangular 20 msecond impulses, 50 mA,
35 Hz, 0.4 seconds duration) following the method
of Swinyard et al. 20 . The prevention of the hindlimb
tonic extensor component was regarded as the
endpoint of protection. The dose–response curves
were estimated by testing four to five doses and eight
(sometimes 10–16) animals per dose. The calculation
of ED50 values (dose that protects 50% of the animals
against MES-induced tonic hindlimb extension) and
the statistical analysis were performed according to
the traditional method of Litchfield and Wilcoxon 19 .
Pentylenetetrazol (PTZ) seizure threshold test
(mice)
In all chemically-induced seizure models the convul-
sants were given after the optimal pretreatment time
of the test drug (or vehicle). Unrestrained mice were
injected with the convulsant PTZ (85 mg kg−1 ) s.c.
in the neck. The animals were placed under separate
glass funnels and the appearance of the first general-
ized clonus (repeated clonic seizures of the fore- and
hindlimbs lasting ≥5 seconds with an accompanying
loss of righting reflex) was recorded during individual
observation for 30 minutes (see Reference 21). The
Anticonvulsant profile of propranolol
number of animals in the group (n = 12 mice) with
clonic seizures and the latency time were analysed for
statistical significance using Fisher’s exact probability
test and the Log rank test, respectively.
Maximal pentylenetetrazol (PTZ) seizure test
(mice)
Clonic–tonic convulsions were induced by admin-
istering PTZ (125 mg kg−1 ) s.c. in the neck of
mice (see Reference 22). In all chemically-induced
tonic seizure models the following criteria were
used for quantification of drug effects: the number
of animals in the group (n = 12) with tonic
hindlimb extension (1), the latency time to the first
tonic seizure (2) and the survival time (3). Results
were compared statistically by using Fisher’s exact
probability test (1) and the Log rank test (2, 3),
respectively.
Maximal N-methyl-DL-aspartate (NMDLA) and
quinolinate (QUIN) seizure test (mice)
NMDLA (2 µg/5 µl) and QUIN (25 µg/5 µl),
respectively, were injected i.c.v. in conscious mice by
a semiautomatic apparatus 23 . Animals were placed in
separate boxes and observed individually for 15 min-
utes. The doses of chemoconvulsants were chosen
from our own pilot studies to induce tonic seizures
and a short survival time. (It should be mentioned
that in the vehicle-pretreated control groups (n =
15) normally one to three animals exhibited no tonic
hindlimb extension.)
Strychnine (STR) seizure pattern test (mice)
Tonic seizures were induced by administering STR
(1.4 mg kg−1 ) s.c. in the neck of mice (see Refer-
ence 24). Preliminary studies showed that 1.2 mg kg−1
did not produce a clear tonic hindlimb extension in
all mice tested. As mentioned earlier, the occurrence
of the first tonic seizure (hindlimb extension) and
the survival time were registered after individual
observation for 30 minutes. In this test, groups of nine
mice were used.
Rotarod ataxia test (mice)
Sedation, decreased locomotor activity and ataxia
were quantified by the rotarod test 25 . Mice were
tested and trained before drug application. Evidence
for ‘minimal neurotoxicity’ was indicated when the
animals were unable to maintain their equilibrium on
the rotating rod (10/minute) in two of three subsequent
1 minute attempts. Respective TD50 values (dose
287
in which 50% of the animals show impaired motor
performance) were calculated with three to four doses
(10 mice per dose) on the basis of dose–response
curves using the method of Litchfield and Wilcoxon 19 .
Electrically-evoked hippocampal afterdischarges
(rats)
Anaesthetized rats (ketamine-HCl 100 mg kg−1 i.p./
xylazine-HCl 15 mg kg−1 i.p.) were placed in a
stereotaxic apparatus (TSE, Bad Homburg, Germany).
A bipolar deep electrode for intracerebral stimulation
and recording was implanted into the right dorsal
hippocampus (AP 2.5 to 3.0 posterior to bregma,
L = 1.8 mm lateral to the midline and V = 2.7–
3.0 mm ventral to the skull surface; stereotaxic coordi-
nates according to Fifková and Maršala 26 ). Addition-
ally, for EEG-recording six stainless-steel electrodes
were positioned epidurally above the bulbus olfacto-
rius, the sinus sagitalis superior (reference electrode;
2 mm frontomedially to the bregma), the sensomotor
and visual cortex. At the end, an antibiotic (Retacillin
compositum, 200 000 IE, Jenapharm, Jena, Germany)
was administered. Two weeks after surgery, the rats
were habituated to the recording setup and EEG-
recordings (Bioscript BST 1, Zwönitz, Germany) were
performed within the following 2–3 days. Constant
current stimulations were delivered to the deep
electrodes in the hippocampus, when a stable EEG re-
sponse has been established from the freely behaving
animals. The stimulus train (rectangular 1 msecond
current impulses, 60 to 300 µA, 50 Hz, 5 seconds
duration) was applied from a HSE-stimulator (type
215/1, HSE, March-Hugstetten, Germany) coupled
to a stimulus isolation unit and a constant current
unit. The individual stimulation threshold for hip-
pocampal rhythmic spike activity (‘afterdischarges’)
was estimated with a series of stimulations, com-
mencing with 60 µA and increasing in 20 µA steps
every 2 minutes until an afterdischarge was elicited
(for further details, see Reference 27). Selected drugs
could be studied when the duration and pattern of
the hippocampal afterdischarges remaining nearly
constant over three to four tests. Each animal received
no more than three drug applications (controls before
and after drug administration at intervals of 3–4 days;
at least 6 days were interposed between two drug
injections in order to avoid drug accumulation or
tolerance). Statistical differences between groups of
data were established by the paired Student’s t-test.
Amygdala-kindling (rats)
Surgery and kindling procedures were performed
as described elsewhere 28 . Briefly, rats were anaes-
288
thetized and implanted with one bipolar electrode
(stainless steel wires tightly twisted together) posi-
tioned stereotaxically into the right basolateral nucleus
of amygdala (AP 1.7 mm posterior to bregma, L =
4.0 mm lateral to the midline and V = 8.0 mm ventral
to the dura 26 ). Two weeks later, the animals were
stimulated once daily with rectangular 1 msecond
current impulses (100 to 200 µA depending on
the individual threshold for afterdischarges, 60 Hz,
1 second). The animals were considered to be kindled
after the achievement of at least three consecutive
generalized seizures with stages 4/5. The severity
of seizures was classified behaviourally according to
a modified ranking scale 6, 29 : 0 = no response to
stimulation; 1 = immobility, eye closure, ear and
facial twitches; 2 = head nodding, more severe facial
clonus, lacrimation; 3 = unilateral forelimb clonus;
4 = rearing, bilateral forelimb clonus with loss of
postural control; 5 = repeated backward falling,
accompanied by generalized clonic–tonic convulsions.
After completion of kindling, rats were grouped to six
to 12 animals and injected with saline (control values)
and 1 day later with the test substance i.p. before the
electrical stimulation. Results were expressed as mean
(scaled pre-drug and post-drug seizure behaviour) of
the tested animal group. Significance of differences
was calculated using the paired Student’s t-test.
Electrode placements were confirmed histologically
using the conventional Nissl-technique (cresyl violet-
stained frontal sections).
Plasma levels of phenobarbital and
carbamazepine (rats)
Plasma concentrations were determined after injec-
tion of the antiepileptic drug alone (phenobarbital
10 mg kg−1 i.p., 60 minutes before or carbam-
azepine 10 mg kg−1 i.p., 30 minutes before; saline
instead of propranolol, 45 minutes before) and in com-
bination with propranolol (10 mg kg−1 i.p., 45 minutes
before). Individual blood samples of 300–500 µl,
taken from the retro-orbital venous plexus under short
diethylether anaesthesia, were collected into Eppen-
dorf tubes and centrifuged at 10 800 rev minute−1
for about 1 minute. Subsequently, 100 µl of the
supernatant was pipetted into Abbott system cartridges
and the total plasma level of phenobarbital or
carbamazepine was determined by an Abbott TDx
analyser (Abbott, Irving, TX, USA), which is based
on a fluorescence polarization immunoassay (FPIA)
technique. The plasma levels were expressed in µM
as means ± SEM (six to 10 rats for each group). For
comparison of the plasma levels of phenobarbital and
carbamazepine in saline- or propranolol-treated rats,
the Mann–Whitney rank sum test was used.
W. Fischer
Heart cell culture (neonatal rats) and whole-cell
voltage-clamp experiments
Ventricular myocardiocytes from 1- to 3-day-old
Wistar rats were prepared by using a combined
enzymatic (0.2% trypsin) and mechanical dissociation
procedure and cultured as monolayers on coverslips
at 37 ◦ C for 5 days (for methodical details, see Ref-
erence 30). For the electrophysiological experiments,
coverslips with cultured cardiomyocytes (2–3 days
after plating) were transferred into a test chamber
(room temperature 20–22 ◦ C), perfused with external
solution and mounted on the stage of an inverted
microscope (Telaval; Carl Zeiss Jena, Germany).
The external (bathing) solution contained (in mM):
NaCl 130, KCl 4, MgCl2 1, CaCl2 1.8, HEPES 10,
glucose 5, buffered to pH 7.4 with NaOH. (In some
experiments the NaCl concentration was lowered to
40 mM (substitution with Tris-HCl 90 mM, CoCl2
5 mM) to reduce current amplitudes for optimizing
voltage-clamp control.) Patch pipettes were pulled
from filamented borosilicate glass capillaries (WPI,
Sarasota, USA) and heat-polished (tip resistances
of 2–4 M). The internal (pipette) solution was
composed of (in mM): CsCl 110, HEPES 10, TEA-
Cl 20, MgATP 5, EGTA 10, buffered to pH 7.0
with Tris-HCl. Sodium currents from selected car-
diomyocytes were studied using the whole-cell patch-
clamp technique 31 by means of an EPC-7 patch-clamp
amplifier (List Electronics, Darmstadt, Germany). The
experimental setup and the data acquisition have
been described in detail elsewhere 32, 33 . The drugs
tested, propranolol, its two enantiomers, valproate and
phenytoin (in DMSO), were dissolved and diluted
in external solution to the desired concentrations
immediately before use. The drug solutions were
applied with an automated application system.
Spinal cord neurone culture (neonatal mice) and
picrotoxin (PTX)-induced burst activity
Spinal cord neurones from 12- to 13-day-old mice
were prepared by using a combined enzymatic (0.25%
trypsin) and mechanical dissociation procedure and
cultured on collagen-coated plastic dishes using
methods described elsewhere 34 . For the electrophys-
iological studies, the cultures (14 days in vitro) were
placed on the stage of an inverted microscope and
superfused with control (HEPES-buffered saline) or
PTX-containing solution, respectively. All recordings
were carried out using the whole-cell patch-clamp
configuration in current-clamp mode (for further
details, see References 34,35). When picrotoxin
(10 µM) was applied to the cell culture, the neurones
within the neuronal network developed paroxysmal,
Anticonvulsant profile of propranolol 289

Fig. 1: Effect of (+)- and (−)-propranolol (2 and 10 mg kg−1 ) alone (left) or of combinations of (+)-propranolol (10 mg kg−1 ) with
various antiepileptic drugs on the threshold for tonic (hindlimb extension) electroshock seizures (MES-T) in mice. The columns
represent the CC50 -values (confidence limits for 95% probability) of drugs or drug combinations (15–20 animals per dose),
expressed in percent to the parallel estimated control MES thresholds (C = 100%, saline/vehicle 10 ml kg−1 i.p.; means of the
control MES thresholds 5.7–6.1 mA). Doses of drugs (in mg kg−1 i.p.) are indicated below the columns. (+)- and (−)-propranolol
((+)-P; (−)-P), carbamazepine (CBZ) and valproate-Ca (VP) were administered 30 minutes, phenobarbital (PB) 60 minutes and
phenytoin (PHT) 90 minutes before threshold determination, respectively. ∗ P < 0.05 (probit analysis).

burst-like firing activity within 10–15 minutes. The volumes of the vehicle (10 ml kg−1 in mice; 2 ml kg−1
effects of (+)- and (−)-propranolol (1–30 µM) were in rats) and were always tested together with the
tested by cumulative administration to the PTX- respective experimental group.
superfusion solution.

Drugs and solutions
The following substances were used: (±)-propranolol-
HCl, (+)-, (−)-propranolol-HCl (purity of the enan-
tiomers approximately 98%), N-methylpropranolol-
HCl (Isis-Chemie, Zwickau, Germany), carbam-
azepine, ethosuximide, phenobarbital-Na, valproate-
Ca (Arzneimittelwerk Dresden, Germany), phenytoin-
Na (Gödecke, Freiburg, Germany); N-methyl-DL-
aspartate (IEM, St. Petersburg, Russia), pentylenete-
trazol (Knoll, Ludwigshafen, Germany), picrotoxin
(Sigma, Deisenhofen, Germany), quinolinic acid
(Sigma, St Louis, USA), strychnine-SO4 (Merck,
Darmstadt, Germany). The drug solutions or sus-
pensions were prepared immediately before use.
All doses refer to the salts. For the screening
experiments, propranolol, its enantiomers and N-
methylpropranolol were dissolved in distilled water
(phenytoin-Na by means of 1–2 drops of 1N NaOH),
N-methyl-DL-aspartate (pH 7), quinolinic acid (pH
6-7), pentylenetetrazol, strychnine in 0.9% NaCl-
solution, carbamazepine, ethosuximide and valproate-
Ca in a 2% suspension of hydroxyethylcellulose.
Animals in the control groups received equivalent
RESULTS
Effects of (+)- and (−)-propranolol on the
threshold for electroshock-induced tonic seizures
Both (+)- and (−)-propranolol (10 mg kg−1
i.p.), administered 30 minutes before testing,
significantly raised the threshold for electrically-
induced tonic (hindlimb extension) seizures in mice
from 5.8 mA (mean of two control groups with saline)
to 7.7 and 7.4 mA, respectively (Fig. 1: presentation
in percent to the corresponding control group). At the
lower dose (2 mg kg−1 i.p.) no strong influence on
the electroconvulsive threshold could be observed.
In combination with phenobarbital, phenytoin,
carbamazepine or valproate, the tested (+)-enantiomer
significantly increased the effectiveness of the latter.
Anticonvulsant activity of (±)-, (+)- and
(−)-propranolol as well as N-methylpropranolol
in the maximal electroshock seizure test
(±)-Propranolol and its two enantiomers showed
anticonvulsant effects in the traditional MES test in
290 W. Fischer

Table 1: Anticonvulsant activity of propranolol, its two enantiomers as well as N-methylpropranolol in comparison with antiepileptic
drugs in the maximal electroshock seizure (MES) test (mice).

Substance Route Time of application ED50 Confidence limits ED50

(min) (mg kg−1 ) (mg kg−1 ) (µmol kg−1 )
(±)-Propranolol-HCl ip 30 15.2 (13.8–16.8) 51.4
45 16.7 (15.0–18.6) 56.5
60 20.6 (19.0–22.3) 69.6
sc 45 14.3 (12.2–16.8) 48.3
po 45 32.2 (26.7–38.8) 108.9
(+)-Propranolol-HCl ip 45 16.0 (14.1–18.2) 54.1
sc 45 13.0 (11.1–15.2) 43.9
po 45 33.5 (27.5–40.9) 113.3
(−)-Propranolol-HCl ip 45 19.2 (16.7–22.1) 64.9
sc 45 14.5 (12.6–16.7) 49.0
po 45 35.6 (31.5–40.2) 120.4
N-Methylpropranolol ip 45 16.4 (12.9–20.9) 52.9

Phenobarbital-Na ip 60 16.0 (14.5–17.6) 62.9

sc 60 17.6 (15.6–19.9) 69.2
po 60 21.2 (17.1–26.2) 83.4
Phenytoin ip 90 8.8 (8.1–9.6) 32.1
Carbamazepine ip 60 11.2 (10.1–12.4) 47.4
Valproate-Ca ip 30 256.0 (245.7–266.7) 1568.6

mice. The animals exhibited no post-ictal depression
(confusion and stupor) and quickly recovered after the
induced tonic seizure. N-methylpropranolol, which
does not possess ß-adrenolytic properties, exerted a
similar protective action. Table 1 summarizes the
estimated ED50 values of these drugs and some con-
ventional antiepileptics as reference substances after
different routes of administration. The anticonvulsant
activity of propranolol as well as its (+)-enantiomer
was roughly equipotent to phenobarbital in this test.
Interestingly, (−)-propranolol (the enantiomer with
the marked ß-adrenoceptor blocking action), tended
to have a lower anticonvulsant activity than the (+)-
enantiomer.
The efficacy of clinically established antiepileptics
was considerably increased when (±)- or (+)-
propranolol (10 mg kg−1 s.c.) were administered as
co-medication. The ED50 values were significantly
reduced in all cases (Fig. 2). Detailed studies with
phenobarbital and (±), (+)-, (−)-propranolol as well
as the N-methyl-derivative showed a dose-dependent
decrease of the ED50 values (see Fig. 2: left part of the
diagram).
Influence of (±)- and (+)-propranolol on
pentylenetetrazol-, N-methyl-DL-aspartate-,
quinolinate- and strychnine-induced tonic
seizures
As illustrated in Fig. 3, the racemic propranolol and
the (+)-enantiomer can suppress tonic (hindlimb ex-
tension) convulsions in different chemically-induced
seizure models in mice. For comparison, the effects
of some antiepileptics are also presented. When tonic
seizures were induced by systemic administration of
PTZ or central administration of NMDLA and QUIN,
respectively, both drugs dose-dependently inhibited
tonic convulsions (Fig. 3: upper diagram). In the re-
maining animals with tonic seizures, the latency to the
first tonic seizure showed a tendency to increase (not
demonstrated). The survival time was significantly
prolonged (lower diagram). On the other hand, in the
STR seizure test (±)- and (+)-propranolol revealed
only limited protective actions and, as expected,
the efficacy of conventional antiepileptics was also
limited.
Influence of (±)- and (+)-propranolol as well as
N-methylpropranolol in the pentylenetetrazol
seizure threshold test
Neither (±)- and (+)-propranolol nor the N-methyl-
derivative exerted protective effects in lower or higher
doses in the s.c.-PTZ seizure threshold test in mice
(Table 2). At 20 mg kg−1 i.p. the latency to the
first generalized clonus tended to decrease, indicating
possible proconvulsant actions. On the other hand, the
reference antiepileptics ethosuximide and valproate
exhibited significant anticonvulsant effects against
clonic seizures.
Anticonvulsant profile of propranolol 291

Fig. 2: Influence of (±)-, (+)- and (−)-propranolol and of N-methylpropranolol on the anticonvulsant effectiveness of phenobarbital
(PB), as well as of (±)- and (+)-propranolol on the anticonvulsant effectiveness of phenytoin (PHT), carbamazepine (CBZ) and
valproate-Ca (VP) in the maximal electroshock seizure (MES) test in mice. The columns represent the ED50 -values (confidence
limits for 95% probability) of drug combinations, expressed as percent of the controls with the antiepileptic drug alone (C = 100%,
combination with vehicle 10 ml kg−1 s.c.). Propranolol compounds were administered s.c. 45 minutes before MES, all
antiepileptic drugs i.p.; see Fig. 1). The average control ED50 -values were: PB 16.1, PHT 8.8, CBZ 11.0 and VP 260 mg kg−1
i.p., respectively. ∗ P < 0.05 19 ; (±)-P = racemic propranolol; (+)-P = (+)-propranolol; (−)-P = (−)-propranolol;
M-P = N-methylpropranolol.

Fig. 3: Effect of (±)- and (+)-propranolol in comparison with some standard antiepileptic drugs on tonic seizures (TS) induced by
pentylenetetrazol (PTZ 125 mg kg−1 s.c.), N-methyl-DL-aspartate (NMDLA 2 µg i.c.v.), quinolinate (QUIN 25 µg i.c.v.) or
strychnine (STR 1.4 mg kg−1 s.c.) in mice. Upper part: number of animals with tonic seizures (dotted boxes) within the groups.
+ P < 0.05; ++ P < 0.01; +++ P < 0.001 (Fisher’s exact probability test).
Lower part: Corresponding survival times (means ± SEM) as percent of the control groups. (Regarding the differences in the
absolute values, depending on a short or longer survival time of the control animals in these various seizure test and the number
of protected animals (calculated with censored 30 or 15 minutes, respectively) in the drug treated groups, it seems necessary to
limit all data to 300% for a clear presentation.) The doses of the tested drugs are indicated below the columns. (The times of
administration of the drugs before application of the chemoconvulsants were always 30 minutes for (±)-, (+)-propranolol and
valproate and 60 min for phenobarbital and carbamazepine.) The average control values ± SEM were: PTZ 11.4 ± 0.8 minutes
(n = 5 control groups); NMDLA 1.4 ± 0.3(n = 5); QUIN 1.5 ± 0.3(n = 4) and STR 5.7 ± 0.3(n = 5). ∗ P < 0.05; ∗∗ P < 0.01;
∗∗∗ P < 0.001 (Logrank test).
292 W. Fischer

Table 2: Effect of propranolol, its (+)-enantiomer as well as N-methylpropranolol on pentylenatetrazol (PTZ)-induced clonic
seizures (mice).

Substance Dose Time of PTZ seizure threshold test (85 mg kg−1 s.c.) % Control
(mg kg−1 i.p.) application Number of mice/ Latency to the
before PTZ with seizure 1. generalized clonus
(min) (min)
NaCl — 30 12/12 7.90 ± 0.97 100 ± 12.3
(±)-Propranolol-HCl 2 30 12/12 8.06 ± 1.01 102.0 ± 12.8
10 30 12/12 8.20 ± 1.68 103.8 ± 21.3
20 30 12/12 6.12 ± 0.96 77.5 ± 12.2
(+)-Propranolol-HCl 2 30 12/12 8.05 ± 1.09 101.9 ± 13.8
10 30 12/12 8.14 ± 1.01 103.0 ± 12.8
20 30 12/12 6.88 ± 0.99 87.1 ± 12.5
N-Methylpropranolol-HCl 10 30 12/12 7.34 ± 1.24 92.9 ± 15.7

NaCl-Cellulose — 30 12/12 8.83 ± 1.05 100 ± 11.9

Ethosuximide 50 30 12/11 12.20 ± 2.33 138.2 ± 26.4
150 30 12/2+++ 14.79 ± 3.38∗ 167.5 ± 38.5∗
Valproate-Ca 50 30 12/12 10.40 ± 1.24 117.8 ± 14.0
200 30 12/2+++ 19.54 ± 4.88∗∗ 221.3 ± 55.3∗∗

All data are given as means ± SEM. Statistical significance: +++ P < 0.001 (Fisher’s exact probability test); ∗ P < 0.05, ∗∗ P < 0.01
(Logrank test). NaCl-Cellulose = 0.9% NaCl in 2% hydroxyethylcellulose suspension.

Adverse effects of (±)- and (+)-propranolol
Impairment of motor function and an increasing
sedation could be observed in mice after application
of higher doses (≥30 mg kg−1 i.p.) of (±)- or (+)-
propranolol. In the rotarod ataxia test, TD50 values
of 41 and 42 mg kg−1 i.p. were determined, respec-
tively. The calculated protective indices (TD50 /MES-
ED50 ) for (±)- or (+)-propranolol were in the order
of 2.5 and 2.6 (for comparison: phenobarbital 4.6,
phenytoin 6.0, carbamazepine 4.7 and valproate 1.3,
respectively). Combination of relatively high doses
of both drugs (34 or 32 mg kg−1 ; i.e. twice
as high as their MES-ED50 s) with phenobarbi-
tal did not significantly alter the TD50 value of
the latter (TD50 phenobarbital/saline: 72 mg kg−1
i.p., phenobarbital/(±)- or (+)-propranolol: 72 and
69 mg kg−1 i.p., respectively).
effects of racemic propranolol and its (+)-enantiomer
on the duration of hippocampal afterdischarges (initial
phase) are shown in Fig. 5(a). The influence on
the stimulation threshold to induce afterdischarges is
illustrated in Fig. 5(b). Both drugs revealed activities
equalling or exceeding those of phenobarbital and
phenytoin.
Further studies with (+)-propranolol (10 or
20 mg kg−1 i.p.) given in combination with pheno-
barbital and phenytoin, respectively, showed that this
co-medication enhanced markedly the effectiveness
of the two antiepileptics (Fig. 6). The higher dose of
(+)-propranolol did not only completely inhibit the
afterdischarge activity, but in addition, significantly
elevated the stimulation threshold.
Effects of (±)- and (+)-propranolol in the
amygdala-kindling model

In amygdala-kindled rats, (±)- and (+)-propranolol

Effects of (±)- and (+)-propranolol upon
electrically-evoked hippocampal afterdischarges
In unrestrained rats stimulated by chronically im-
planted hippocampal electrodes, (±)- and (+)-
propranolol (10 and 20 mg kg−1 i.p.) reduced or
completely suppressed the initial phase of generalized
polyspike-wave and spike discharges. A representative
EEG-recording with (+)-propranolol is illustrated in
Fig. 4. The secondary (rebound) afterdischarge phase,
characterized by sharp-waves in the hippocampus and
visual cortex beginning about 80–90 seconds after the
stimulation, was also suppressed. The calculated mean
exerted anticonvulsant activity against secondarily
generalized clonic–tonic seizures (stage 4/5). At doses
of 10 mg kg−1 i.p. both drugs reduced seizure
severity to stage 3, but did not suppress the facial and
unilateral forelimb cloni (Fig. 7). The efficacy of some
antiepileptics was presented for comparison.
Effects of propranolol on the plasma level of
phenobarbital and carbamazepine
Propranolol (10 mg kg−1 i.p.) did not
alter the total plasma levels (µM) both of
phenobarbital (10 mg kg−1 i.p.) (alone 48.9 ± 1.1
Anticonvulsant profile of propranolol 293

Fig. 4: Electrically-evoked epileptiform spike activity (‘afterdischarges’) following hippocampal current stimulation in the
unrestrained rat (EEG-recording example). After habituation of the animal for 15 minutes (relaxed wakefulness) the stimulus was
delivered to the right dorsal hippocampus (region 4–5). Top panel: control recording before and after stimulation
(rectangular 1 msecond current impulses of 180 µA and 5 seconds duration) without drug treatment. The evoked epileptiform
activity has a complex pattern which consists of (1) initial polyspike-wave and spike discharges (without convulsive response),
associated with postural freezing and with some head-shakes (HS) at the end, (2) a period free of afterdischarges (post-ictal
voltage depression), with pronounced forward locomotion, sniffing and rearing activity, and (3) a second (rebound) phase
after 80–90 seconds characterized by sharp-waves (see hippocampus, visual cortex), with a short arrest, some head-shakes and
after that again increased locomotor activity for 3 (5) minutes. Bottom panel: effect of a single dose of (+)-propranolol
(10 mg kg−1 i.p.) on electrically-evoked hippocampal afterdischarges, recorded 30 minutes after drug application (3 days after
predrug control). (+)-Propranolol markedly reduced the initial spike-wave phase and the second (sharp-wave) phase was
completely suppressed. SMC = sensomotor cortex; DH = dorsal hippocampus; VC = visual cortex; BO = bulbus olfactorius.
Voltage calibration on the right: 60 µV.

(n = 10) vs. 48.4 ± 1.5 (n = 6)) or carbamazepine
(10 mg kg−1 i.p.) (alone 19.8 ± 2.4 (n = 10) vs.
18.8 ± 3.8 (n = 6)). There are no relevant pharma-
cokinetic interactions, in terms of total plasma levels,
between propranolol and the two tested antiepileptics.
Effects of (+)- and (−)-propranolol on fast
sodium inward current INa in cultured neonatal
rat cardiomyocyte cells
Current–voltage relationship of peak INa
Sodium currents (INa s) were evoked by a series of
25 msecond depolarization pulses (0.1–1 Hz; holding
potential of −100 mV) to different depolarizing
potentials. INa s started to activate at −60 mV reaching
a maximum amplitude around −20 mV, and then
declined towards the reversal potential near +40 mV.
The INa s were stable over a range of 15 minutes
with only minimal ‘run down’ effect. The current–
voltage relationship for the peak INa before and
after application of (+)-propranolol is shown in
Fig. 8(a). (The two insets on the left illustrate
the pulse protocol (above) and a family of sodium
currents elicited by voltage-steps after this protocol
from a selected experiment (below) under control
conditions and after exposure to 100 µM of the test
substance.) (+)-Propranolol and the (−)-enantiomer
(in the same magnitude; not documented) reduced the
INa amplitude over the entire membrane potential axis
without a shift of the curve maximum or a change
in the reversal potential (decrease of the maximum
INa amplitude by about 30% (10 µM) and 50–60%
(100 µM)). The onset of inhibition produced by (+)-
or (−)-propranolol occurred within 3 minutes, and the
block was reversible within 5 minutes of washout in
drug-free extracellular solution.
Frequency-dependent inhibition of peak INa
Under control conditions, repetitive depolarizing test
pulses (25 mseconds; from −120 mV to −30 mV)
produced little decrement in current amplitude (2–3%
at 10 Hz in a train of 10 test pulses). After application
of (+)- and (−)-propranolol, respectively, the drugs
led to a ‘tonic’ reduction of peak INa measured
after 3–5 minutes with some single test pulses. An
increase of the test pulse frequency to 1 or 10 Hz,
markedly increased the amount of current block by
both (+)- and (−)-propranolol (Fig. 8(b)). At 1 Hz,
the peak INa decreased about 30% with 100 µM (+)-
or (−)-propranolol and at 10 Hz up to 80%. Thus,
both enantiomers exhibited a pronounced frequency-
dependent inhibition of peak INa , especially at higher
drug concentrations. These additional blocking effects
always recovered after the reduction or cessation of
higher frequency stimulation.
Steady-state inactivation of peak INa
The voltage-dependence of INa block was estimated
by applying a conditioning 200 msecond prepulse
294
Fig. 5: Effect of (±)- and (+)-propranolol ((±)-P; (+)-P) in
comparison with phenobarbital (PB), phenytoin (PHT) and
carbamazepine (CBZ) on (a) the duration of
electrically-evoked hippocampal spike activity (initial phase)
and (b) the stimulation threshold for hippocampal
afterdischarges. The rats were tested 3 days before (control
values = 100%) and 25–30 minutes ((±)-P; (+)-P),
55–60 minutes (PB), 80–90 minutes (PHT) and 45–60
minutes (CBZ), respectively, after drug application
(intra-individual comparison). Each rat underwent maximal
three drug tests, separated by one (two) control trials. The
columns represent the means ± SEM of six to 11 animals.
The doses of drugs (in mg kg−1 i.p.) are given always below
the columns (at the base of the columns in the lower diagram:
number of animals with increased threshold/total number of
animals per group). The average control values (duration of
initial spike phase) lay between 25 and 39 seconds, the
means of control thresholds were between 162 and 226 µA.
∗ P < 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001 (paired t-test).
(varying from −120 mV to different depolarizing
potentials up to −10 mV in increasing 10 mV
steps) to the cell followed immediately by the
15 msecond test pulse (from −120 to −30 mV).
Both enantiomers shifted the voltage-dependence of
steady-state inactivation in a more hyperpolarized
direction without marked change in the slope of the
curves. Figure 8(c) shows the relationship between the
steady-state inactivation and the membrane potential
before and after exposure to (+)-propranolol from two
representative experiments. Under control conditions,
sodium currents began to inactivate at prepulse
potentials positive to about −90 mV and steady-
state inactivation was complete at about −50 mV.
W. Fischer
Fig. 6: Effect of co-administered (+)-propranolol ((+)-P) on
the anticonvulsant effectiveness of phenobarbital (PB) and
phenytoin (PHT), respectively, in the hippocampal
afterdischarge model. The rats were tested 3–4 days before
the drug application (control values). Each animal received
(+)-propranolol (10 or 20 mg kg−1 ), the antiepileptic drug
(PB or PHT) and the corresponding combination (COM) in a
random schedule. The interval between each drug
application was at least 6 days. Only animals with a constant
threshold over the whole series of experiments were included
in the analysis. (a) Duration of electrically-evoked
hippocampal afterdischarges (initial spike phase) before and
after drug application. Each column represents the mean ±
SEM of four to five animals. Significance level: ∗∗ P < 0.01
(paired t-test). (b) Modulation of the stimulation threshold for
hippocampal afterdischarges. Data (means ± SEM) are
shown as elevation over individual control threshold. The
doses of drugs are always given below the columns (at the
base of each column: number of animals with increased
threshold/total number of animals per group). Means of
control thresholds were for the three independent series of
experiments (in µA): PB/(+)-P 10 mg kg−1 = 169 ± 30;
PB/(+)-P 20 mg kg−1 = 202 ± 39 and PHT/(+)-P
20 mg kg−1 = 195 ± 27. ∗ P < 0.05 (paired t-test).
At the physiological membrane potential of −80 mV,
for example, the mean availability of INa was
about 80%. (+)-Propranolol (10−5 and 10−4 M)
reduced the current to 50 and 30%, respectively, at
this potential. The quantitative analysis revealed that
the (+)-enantiomer shifted V0.5 (membrane potential
where 50% of the channels are available; midpoint
of the inactivation curve) from −70.1 ± 2.9 mV to
−78.6 ± 2.7 (10−5 M) and −83.3 ± 3.3 mV (10−4 M;
mean ± SEM from five cells). On the other hand, the
(−)-enantiomer displaced V0.5 from −71.5 ± 2.5 mV
to −79.5 ± 2.9 (10−5 M) and −84.3 ± 4.1 mV
(10−4 M; n = 5 cells). After washout, the V0.5 values
in most cases recovered. Consequently, (+)- as well
Anticonvulsant profile of propranolol
Fig. 7: Effect of (±)- and (+)-propranolol ((±)-P; (+)-P) in
comparison with phenobarbital (PB), phenytoin (PHT),
carbamazepine (CBZ) and valproate-Ca (VP) in fully
amygdala kindled rats with reproducible seizures of
stages 4/5. Control responses (saline) were measured 1(2)
days prior to drug administration in the same group of
animals. Each double column represents the mean seizure
stage (scaled pre-drug and post-drug seizure behaviour
indicated as dashed line and solid line columns, respectively)
of tested groups (number of animals within the group at the
base of each column). The doses of drugs (in mg kg−1 i.p.)
and the pretreatment times are given below the columns.
Significance level: ∗ P < 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001
(paired t-test).
Fig. 8(a): Effect of (+)-propranolol (10 and 100 µM) on the
current–voltage (I–V) relationship for peak sodium inward
current (INa ) in the absence (◦-◦) and presence (•-•) of the
drug. Means ± SEM (at maximum values) of five rat
myocardial cells. The cells were voltage-clamped at a
membrane potential of −100 mV and sodium currents were
elicited by a series of 25 msecond test pulses at intervals of
3 seconds, in 10 mV steps to +40 mV. The I–V relationship
was obtained by plotting the peak INa as a function of the
test pulse potential. (Leak currents were subtracted using
average values of steady leak currents elicited by
hyperpolarizing pulses to −120 mV.) The insets on the left
illustrate the pulse protocol (top) and a family of INa curves
(bottom) from an individual cell before (control) and
after 3–5 minutes exposure to 100 µM (+)-propranolol
((+)-PROP).
as (−)-propranolol decreased to the same extent the
potential-dependent availability of sodium channels,
i.e. the number of channels for opening at the resting
potential.
295
Concentration-dependent inhibition of peak INa
The concentration-dependent inhibition of peak INa
by (+)- and (−)-propranolol in comparison with
phenytoin and valproate is shown in Fig. 9. The
concentration–response curves were obtained by
cumulative administration of drugs up to 10 µM;
before application of 100 µM, a washout period
was generally inserted to reduce long-lasting drug
influences and irreversible effects. The solid curves
represent the ‘tonic’ inhibition 3–5 minutes after
drug application (single depolarizing test pulses from
−120 to −30 mV). The dashed curves indicate the
additional, frequency-dependent reduction after the
last of 10 pulses with 10 Hz stimulation. A clear
reduction of peak INa is already observed at 1 µM to
10 µM. Half-maximal inhibition (IC50 ) values were in
the range of 100 µM for both enantiomers. Compared
to (+)- or (−)-propranolol, the inhibitory activity of
phenytoin (IC50 90 µM) was only somewhat higher.
In contrast, valproate exhibited a smaller reduction
of peak INa , even at high concentrations (about 15%
at 1 mM). The frequency-dependence was also not
very pronounced.
Effects of (+)- and (−)-propranolol on
picrotoxin-induced burst activity in cultured
neonatal mouse spinal cord neurones
After application of 10 µM PTX, spinal cord neurones
in culture developed a paroxysmal burst-like firing
activity (Fig. 10). The (+)- and (−)-enantiomers of
propranolol were nearly equieffective in inhibiting
this burst activity (concentration range 1, 10, 30 µM:
21.4/18.7, 45.1/42.6, 100/100% reduction of burst
duration, respectively; n = 4 cells). The results of
a representative experiment with (+)-propranolol are
shown in Fig. 10. The ß-blocker talinolol (100 µM),
without marked local anaesthetic properties, exhibited
no clear inhibitory influence (100 µM: 10% reduction
of burst duration; n = 2 cells) (not shown).
DISCUSSION AND CONCLUSION
The aims of the present study were to determine
(i) the anticonvulsant profile of propranolol and (ii) its
mechanism of action. The data show that propranolol
including its (+)-enantiomer with practically negligi-
ble ß-blocking activity but similar lipophilic and local
anaesthetic properties 36 , exhibits anticonvulsant prop-
erties in screening models for generalized tonic–clonic
seizures like the MES threshold test, the traditional
MES test as well as the maximal PTZ, NMDLA and
QUIN seizure test. Furthermore, in the hippocampal
296 W. Fischer

Fig. 8(b): Frequency-dependent inhibition of peak INa by (+)-propranolol (upper panels) and (−)-propranolol (lower panels) at
concentrations of 1–100 µM, respectively. Means ± SEM of four to seven rat myocardial cells. Voltage-clamp pulses from −120
to −30 mV of 25 msecond duration were applied at 1 and 10 Hz, respectively, in trains of 10 impulses. The peak current
amplitude for each pulse (In ) in the train was normalized to the current amplitude elicited by the first pulse (Ist ) and plotted as a
function of the pulse number. The insets on the left (top diagram) show the pulse protocol and a family of INa curves from a
representative cell before (control) and after 3–5 minutes exposure to 10 µM (+)-propranolol, stimulated with 10 Hz (note the
marked reduction of peak INa pulse by pulse).

afterdischarge model, predictive for complex partial
seizures (see Reference 33), (±)- and (+)-propranolol
reduced the duration of afterdischarges and increased
the afterdischarge threshold. It should be mentioned
that in agreement with the results in both MES
tests, the co-medication of (+)-propranolol markedly
increased the effectiveness of the tested standard
antiepileptics. Previous investigations revealed ad-
ditive anticonvulsant effects of (+)-propranolol in
combination with phenobarbital 37 . In fully kindled
rats, (±)- and (+)-propranolol reduced the seizure
severity down to stage 3 (unilateral forelimb seizures)
comparable to the effects of carbamazepine, whilst
phenobarbital and valproate reduced the seizure
severity up to stage 1 (facial twitches). On the other
hand, in the s.c.-PTZ (85 mg kg−1 ) seizure threshold
test, used as the standard model for myoclonic (and
absence) seizures (see Reference 38), propranolol
did not exhibit protective effects against generalized
clonic convulsions. Altogether, the profile of action of
propranolol seems to be comparable to that of the two
standard antiepileptics phenytoin and carbamazepine.
It is now well accepted that propranolol possesses
anticonvulsant activity in various seizure models,
Anticonvulsant profile of propranolol
Fig. 8(c): Effect of (+)-propranolol on the steady-state
inactivation of peak INa . Data (mean values) from two
representative experiments are presented. Channel
availability was modulated by varying the conditioning
prepulse (200 mseconds) between −120 and −10 mV in
increasing 10 mV steps prior to a test pulse (15 mseconds)
to −30 mV (see pulse protocol on the left). The cells were
held at a membrane potential of −120 mV (interval
2 seconds). The peak INa during the test pulse was
normalized to the control current amplitude for the first
stimulus, when the prepulse potential was the same as the
holding potential (−120 mV). The normalized current was
then plotted as a function of the prepulse potential and the
resulting data were fitted to the Boltzmann equation. Inset on
the right: families of INa curves from an individual cell before
(control) and after 3–5 minutes exposure to 10 µM
(+)-propranolol (numbers 4, 5, 6: prepulse potentials −90,
−80, −70 mV, respectively).
although conflicting results exist in the literature
regarding the effectiveness in diverse seizure
tests. Surprisingly, there are no detailed studies
on kindling models of epilepsy in the literature.
Only Gellman et al. 39 reported that propranolol
(2.5 mg kg−1 i.p.) did not alter the seizure response
in fully amygdala-kindled rats. The majority of
studies documented efficacy in electroshock seizure
tests like the MES and/or PTZ-induced tonic seizure
tests 7, 13–15, 40–45 . As in some papers, the (+)-
enantiomer is an equally or more potent drug in
this relationship, suggesting that the anticonvulsant
effects are not related to ß-adrenoceptor blockade 7, 14
but rather due to the ‘membrane stabilizing’
properties 8, 15, 46 . There are, however, two remarkable
exceptions. In DBA/2 mice, a genetically sensitive
strain for sound-induced generalized tonic–clonic
seizures 47 , a stronger anticonvulsant potency for the
(−)-enantiomer was observed (see References 11,48;
De Sarro et al., Eur. J. Pharmacol, 2002, in
press). Because of the high doses required, these
authors suggested that the ‘membrane stabilizing’
effects in this model also markedly contribute
to the anticonvulsant activity. Differences in the
297
Fig. 9: Concentration-dependent inhibition of peak INa by
(+)- and (−)-propranolol ((+)-P; (−)-P) as well as phenytoin
(PHT) and valproate-Ca (VP). Data are given as
means ± SEM (number of tested cells in parentheses). Solid
curves: ‘tonic’ inhibition 3–5 minutes after application of the
drug (single depolarizing test pulses from −120 to −30 mV
of 25 mseconds duration). Dashed curves: additive,
frequency-dependent inhibition (10 Hz stimulation, trains
of 10 impulses).
pharmacokinetic properties of the enantiomers or an
increase in the ß-adrenoceptor density of DBA/2-mice
may be responsible for the different anticonvulsant
potency of (−)- and (+)-propranolol 48 . Furthermore,
Louis et al. 12 and Papanicolaou et al. 49 reported
that the (−)-enantiomer of propranolol was seven
times more effective than the (+)-enantiomer (oral
administration of 0.05–1 mg kg−1 ) in decreasing the
seizure duration in PTZ(50 mg kg−1 i.p.)-induced
tonic–clonic seizures in rats. They suggested that
propranolol exert an anticonvulsant effect through
central ß-adrenoceptors, whereas at higher dose levels,
additional anticonvulsant activity is associated with
membrane stabilization.
As already discussed, the virtually identical
anticonvulsant potency of the non-receptor-active
(+)-enantiomer and the anticonvulsant activity of
N-methylpropranolol without marked ß-blocking
action 50 , argue against a relevant involvement
of central ß-adrenoceptors in the anticonvulsant
effectiveness of propranolol.
The following arguments also speak for this
statement.
298
1. The doses of propranolol causing anticonvul-
sant effects (10–20 mg range) are about 10-
fold higher than those required for adequate
ß-adrenoceptor blockade (1–2 mg range).
2. Pindolol and timolol, two lipophilic ß-blockers
with stronger ß-adrenolytic potency than pro-
pranolol 51, 52 , exhibited a smaller or no anticon-
vulsant effect, respectively.
3. A subchronic pre-medication with combi-
nations of desipramine/phenoxybenzamine or
desipramine/yohimbine, which induced a rapid
decrease in rat brain ß-receptor binding 53 ,
revealed no significant modulation of the an-
ticonvulsant effectiveness of propranolol (data
not shown, see Reference 6).
Fig. 10: Effect of (+)-propranolol ((+)-PROP) on picrotoxin
(PTX)-induced burst activity in cultured mouse spinal cord
neurones (representative tracings from a single experiment).
Within 10–15 minutes of superfusion with 10 µM PTX, the
spinal cord neurones developed a characteristic, burst-like
firing activity (a). Traces (b) and (c) show that (+)-PROP
dose-dependently inhibited or suppressed this paroxysmal
activity (see burst duration and number of action
potentials/burst). The burst activity recovered after 3–5
minutes of washout with (+)-PROP-free PTX-containing
solution (d). The membrane potential (−60 mV; scale bar on
the right) was unaffected by (+)-PROP.
Phenytoin and carbamazepine as well as a number
of promising new compounds block voltage-sensitive
sodium channels in a complex voltage- and frequency-
dependent manner 10, 54–56 . Since propranolol exhibits
W. Fischer
anticonvulsant activity in the same screening models
as phenytoin and carbamazepine, it was of interest to
investigate directly the possible influence of (+)- and
(−)-propranolol on the fast inward sodium current INa .
In earlier studies, Tarr et al. 57 found inhibitory effects
of propranolol on the fast inward sodium channel in
frog atrial muscle fibres using the sucrose-gap voltage-
clamp technique. Further investigations showed that
this drug decreased the maximum upstroke velocity
(Vmax ) and amplitude of the action potential in
heart muscle fibres 58–60 and inhibited (veratridine-
stimulated) Na+ influx as well as batrachotoxinin
binding in various brain membrane or cardiac tissue
preparations in the low µM-range, indicating Na+
channel blocking effects at the neurotoxin binding
site 2 61–63 . The present whole-cell patch-clamp exper-
iments were conducted primarily with cardiomyocytes
(see also Reference 27). Although sodium channels in
cardiac and nerve membranes are not fully identical,
sodium channels in the heart behave functionally very
much like those of neuronal cells 64, 65 . Interestingly,
the expression of cardiac sodium channel mRNA
was recently demonstrated in restricted areas of rat
and human brain, especially in limbic structures and
diencephalon 66, 67 . Hartmann et al. 66 suggested that
‘arrhythmias’ of heart and sensible brain regions
may be related and implicate these tetrodotoxin-
resistant sodium channels in some forms of primary
inherited epilepsy. The present studies demonstrate
that the two enantiomers of propranolol exhibit
equieffective sodium channel blocking effects and the
potency is very similar to that of phenytoin (10–
100 µM reduced the peak INa by approximately 20–
50%). Moreover, for both propranolol enantiomers
and phenytoin (see Reference 6), a clear voltage-
and frequency-dependence in blocking voltage-gated
sodium channels were observed. These phenomena
(decrease of the channel availability at stronger mem-
brane depolarizations, increase of channel inhibition
during higher stimulation frequency) may be very
important mechanisms in reducing the ability of cells
to fire trains of action potentials at high frequency. The
present findings are in agreement with previous reports
on the complex interactions of phenytoin and carba-
mazepine with sodium channels 9, 55 . Together, these
specific effects provide a good explanation for how
such drugs can reduce neuronal hyperexcitability and
suppress seizures by selectively inhibiting sustained
high-frequency firing in epileptic foci without altering
normal neuronal brain function 10, 68, 69 .
In addition, the whole-cell patch-clamp studies
on mouse spinal cord neurones in culture showed
that both (+)- and (−)-propranolol equieffectively
inhibited PTX-induced paroxysmal burst-like firing
activity in a concentration-dependent manner. The
Anticonvulsant profile of propranolol
potency of phenytoin and carbamazepine was in the
same order as determined by Binscheck et al. 35 . These
authors suggested that in this model for rapid firing of
action potentials within a neuronal network, the burst
duration might especially be a sensitive parameter
allowing quantification of anticonvulsant drug effects.
On the other hand, the peripheral ß-blocker talinolol,
without marked local anaesthetic properties 70 , showed
no clear inhibitory effects. Also these results underline
that the anticonvulsant activity of propranolol seems
to be related to the known local anaesthetic properties
and not to the antagonism of ß-adrenoceptors in the
CNS. Further studies must show if interferences of
propranolol with Ca2+ channels 71, 72 may play a role.
One might argue that the concentrations of propranolol
necessary to block Na+ channels are very high.
However, propranolol is concentrated in the brain of
animals and man. A single dose of (±)-propranolol
(22 mg kg−1 s.c.; mice), causes a brain concentration
of 20.3 µg g−1 (8×10−5 M; brain–plasma relationship
26 : 1) 73 . In humans (usual therapeutic dose range 80–
320 mg day−1 ) 74 , propranolol (80 mg p.o. twice
daily) causes brain concentrations of 0.8–5.7 µg g−1
(0.3–2 × 10−5 M; brain–plasma relationship 17 : 1) 75 .
Moreover, in the treatment of therapy-resistant
schizophrenic patients excessive propranolol doses
(means of 500–1600 mg day−1 ) were used 5, 76 .
The anticonvulsant effects of propranolol were
observed in a dose range in which approximately 10–
20% reduction in heart rate (unrestrained rats)
and 5–20% decrease in mean arterial blood pressure
(urethan-anaesthetized rats) was determined after (−)-
propranolol (10–20 mg kg−1 i.p., 30 minutes after
application) (for further details, see Reference 77).
The haemodynamic effects of the (+)-enantiomer
at higher doses were only somewhat smaller. In
another study with a lower dose of (±)-propranolol
(5 mg kg−1 i.p., daily for 10 days), Pessina et al. 78
reported no significant influence on the systolic and
diastolic blood pressure in normotensive as well as
spontaneously hypertensive rats and only a significant
reduction in heart rate in the latter. The possibility
that the changes of cardiovascular parameters can
be of relevance for the anticonvulsant properties of
propranolol, however, seems to be very unlikely. For
example, various potent peripheral ß-adrenoceptor
antagonists like bisoprolol, bunitrolol, talinolol and
timolol, which can be assumed to produce similar
patterns of haemodynamic effects in adequate doses,
revealed no anticonvulsant properties in the MES
test 6 .
To the knowledge of the author, there exist no
controlled clinical studies of antiepileptic effects of
propranolol. However, some findings in the literature
provide evidence for beneficial therapeutic effects
by the additive administration of this drug. For
299
example, in a case report of a young woman with
complex partial seizures following head trauma,
the combination of carbamazepine and propranolol
(120 mg twice daily) seems to produce synergistic
antiepileptic effects 16 . In the treatment of startle
epilepsy, a rare but severe seizure disorder with
predominantly tonic and tonic–myoclonic seizures,
propranolol (up to 240 mg day−1 ) was reported
to be an additional and safe drug 17 . Furthermore,
propranolol has been used effectively for the treat-
ment of valproate-induced tremor and relevant drug
interactions were excluded 79, 80 . Concerning the fact
that complex partial seizures are frequently associated
with an increase in blood pressure and cardiac
dysfunctions 81, 82 and ictal heart arrhythmias seem
to be an important risk factor of epilepsy 83, 84 , the
co-medication of propranolol or its (+)-enantiomer
with pronounced cardioprotective effects 85 might be
of interest. Principally, an investigation of seizure
incidence in epileptic patients taking ß-blockers like
propranolol for other indications should also be of
importance. In addition, as suggested by Nutt 86 ,
propranolol may have clinical value in patients
experiencing post-ictal phenomena like confusion and
stupor.
Taken together, the present results show marked
anticonvulsant properties of the lipophilic ß-
adrenoceptor antagonist propranolol and its two
enantiomers in experimental models of generalized
tonic–clonic and complex partial seizures. Moreover,
in combination with conventional antiepileptic drugs,
additive anticonvulsant effects can be observed.
With regard to the possible mode of action, these
compounds were found to depress the INa in cultured
rat cardiomyocytes in a concentration- and use-
dependent manner and inhibited picrotoxin-induced
burst firing activity of mouse spinal cord neurones in
culture. It can be suggested that the sodium channel
blocking properties and not the ß-receptor antagonistic
activity accounts for the anticonvulsant effects of
propranolol.
ACKNOWLEDGEMENTS
The author wishes to thank Dr G. Wallukat and
Dr R. Bodewei (Berlin-Buch) for supplying cultured
rat myocardial cells and for the excellent support in the
patch-clamp studies, respectively; Mrs Dr G. Keller
and Mrs T. Großmann (Dresden) for the good
collaboration in the amygdala-kindling experiments;
Dr I. V. Ryzov and Dr A. P. Garyaev (St Petersburg,
Russia) for skilful technical assistance in the QUIN-
and NMDLA-induced seizure models, respectively,
and Prof. Dr H. Bigalke and Dr R. Ties (Hanover)
for making it possible to carry out some studies on
300
cultured mouse spinal neurones. The author is also
indebted to Mrs U. Beier (deceased) and Mrs U.
Kermes (deceased) for technical assistance in the
screening experiments and Mrs M. Klausch (Leipzig)
for the determination of phenobarbital and carbam-
azepine plasma levels with the Abbott TDx analyser.
Furthermore, I would like to thank Prof. E. Schlicker
(Bonn) for critically reading the manuscript. The gift
of propranolol and its enantiomers by Isis Chemie
(Zwickau) is gratefully acknowledged.
REFERENCES
1. Fitzgerald, J. D. Propranolol. In: Pharmacology of Antihyper-
tensive Drugs (Ed. A. Scriabine). New York, Raven Press,
1980: pp. 195–208.
2. Slogoff, S. Beta-adrenergic blockers. In: Cardiac Anesthesia.
Vol. 2: Cardiovascular Pharmacology (Ed. J. A. Kaplan). New
York, Grune & Stratton, 1983: pp. 181–208.
3. Middlemiss, D. N., Buxton, D. A. and Greenwood, D. T. Beta-
adrenoceptor antagonists in psychiatry and neurology. Phar-
macology and Therapeutics 1981; 12: 419–437.
4. Pöldinger, W. Beta-rezeptorenblocker in der Neurologie
und Psychiatrie. Zeitschrift für Allgemeinmedizin 1985; 61:
7–12.
5. Bailly, D. The role of ß-adrenoceptor blockers in the
treatment of psychiatric disorders. CNS Drugs 1996; 5:
115–136.
6. Fischer, W. Pharmakologische Modulation noradrenerger und
membranstabilisierender Mechanismen—Möglichkeiten zur
Beeinflussung der Effektivität vonStandardantiepileptika, The-
sis (Habilitationsschrift). Halle-Wittenberg, Martin-Luther-
Universität, 1995.
7. Lasek, R., Fischer, W. and Müller, M. Zur antikonvulsiven
Wirksamkeit von β-Adrenolytika beim Elektrokrampf. Phar-
mazie 1983; 38: 202–203.
8. Fischer, W., Lasek, R. and Müller, M. Anticonvulsant effects
of propranolol and their pharmacological modulation. Polish
Journal of Pharmacology and Pharmacy 1985; 37: 883–896.
9. Macdonald, R. L. and Kelly, K. M. Antiepileptic drug
mechanisms of action. Epilepsia 1995; 36 (Suppl. 2): S2–S12.
10. Ragsdale, D. S. and Avoli, M. Sodium channels as molecular
targets for antiepileptic drugs. Brain Research Reviews 1998;
26: 16–28.
11. Anlezark, G., Horton, R. and Meldrum, B. The anticonvulsant
action of the (−)- and (+)-enantiomers of propranolol. Journal
of Pharmacy and Pharmacology 1979; 31: 482–483.
12. Louis, W. J., Papanicolaou, J., Summers, R. J. and Vajda,
F. J. E. Role of central ß-adrenoceptors in the control
of pentylenetetrazol-induced convulsions in rats. British
Journal of Pharmacology 1982; 75: 441–446.
13. Murmann, W., Almirante, L. and Saccani-Guelfi, M. Central
nervous system effects of four β-adrenergic receptor blocking
agents. The Journal of Pharmacy and Pharmacology 1966; 18:
317–318.
14. Jaeger, V., Esplin, B. and Čapek, R. The anticonvulsant effects
of propranolol and β-adrenergic blockade. Experientia 1979;
35: 80–81.
15. Khanna, N., Ray, A., Alkondon, M. and Sen, P. Effect of β-
adrenoceptor antagonists and some related drugs on maximal
electroshock seizures in mice. Indian Journal of Experimental
Biology 1989; 27: 128–130.
16. Renshaw, P. F., Ford, H. E. and Brotman, A. W. Possible syner-
gistic anticonvulsant effect of propranolol and carbamazepine.
American Journal of Psychiatry 1990; 147: 1687–1688.
17. Mayer, T. and Specht, U. Propranolol in startle induced
W. Fischer
epileptic seizures. Journal of Neurology, Neurosurgery, and
Psychiatry 1995; 58: 382–383.
18. Löscher, W., Fassbender, C. P. and Nolting, B. The role
of technical, biological and pharmacological factors in the
laboratory evaluation of anticonvulsant drugs. II. Maximal
electroshock seizure models. Epilepsy Research 1991; 8:
79–94.
19. Litchfield, J. T. Jr and Wilcoxon, F. A simplified method of
evaluating dose-effect experiments. Journal of Pharmacology
and Experimental Therapeutics 1949; 96: 99–113.
20. Swinyard, E. A., Brown, W. C. and Goodman, L. S. Compar-
ative assays of antiepileptic drugs in mice and rats. Journal
of Pharmacology and Experimental Therapeutics 1952; 106:
319–330.
21. Krall, R. L., Penry, J. K., White, B. G., Kupferberg,
H. J. and Swinyard, E. A. Antiepileptic drug development: II.
Anticonvulsant drug screening. Epilepsia 1978; 19: 409–428.
22. Jones, G. L. and Woodbury, D. M. Principles of drug
action: structure-activity relationships and mechanisms. In:
Antiepileptic Drugs (Eds D. M. Woodbury, J. K. Penry and
C. E. Pippenger). New York, Raven Press, 1982: pp. 83–109.
23. Lapin, I. P. and Ryzov, I. V. Effect of catecholaminergic drugs
on quinolinate- and kynurenine-induced seizures in mice.
Journal of Neural Transmission 1990; 82: 55–65.
24. Porter, R. J., Cereghino, J. J., Gladding, G. D., Hessie, B. J. H.,
Kupferberg, H. J., Scoville, B. and White, B. G. Antiepileptic
drug development program. Cleveland Clinic Quarterly 1984;
51: 293–305.
25. Dunham, N. W. and Miya, T. S. A note on a simple apparatus
for detecting neurological deficit in rats and mice. Journal of
American Pharmaceutical Association 1957; 46: 208–209.
26. Fifková, E. and Maršala, J. Stereotaxic atlases for the cat,
rabbit and rat. In: Electrophysiological Methods in Biological
Research, Appendix I, 3rd revised Edition (Eds J. Bureš,
M. Petráň and J. Zachar). Prague, Acad. Publ. House,
Czechoslovak. Acad. Sci., 1967: pp. 653–731.
27. Fischer, W., Kittner, H., Regenthal, R., Malinowska, B. and
Schlicker, E. Anticonvulsant and sodium channel blocking
activity of higher doses of clenbuterol. Naunyn-Schmiedebergs
Archives of Pharmacology 2001; 363: 182–192.
28. Schmidt, J. Comparative studies on the anticonvulsant
effectiveness of nootropic drugs in kindled rats. Biomedica
Biochimica Acta 1990; 49: 413–419.
29. Racine, R. J. Modification of seizure activity by electri-
cal stimulation. II. Motor seizure. Electroencephalography
and Clinical Neurophysiology 1972; 32: 281–294.
30. Wallukat, G. and Wollenberger, A. Differential α- and ß-
adrenergic responsiveness of beating rat heart myocytes after
stationary and non-stationary cultivation. Acta Biologica et
Medica Germanica 1980; 39: K7–K13.
31. Hamill, O. P., Marty, A., Neher, E., Sakmann, B. and
Sigworth, F. J. Improved patch-clamp techniques for high-
resolution current recording from cell and cell-free membrane
patches. Pflügers Archiv European Journal of Physiology
1981; 391: 85–110.
32. Hering, S., Kleppisch, T., Timin, E. N. and Bodewei,
R. Characterization of the calcium channel state transitions
induced by the enantiomers of the 1,4-dihydropyridine Sandoz
202 791 in neonatal rat heart cells. A nonmodulated receptor
model. Pflügers Archiv European Journal of Physiology 1989;
414: 690–700.
33. Fischer, W., Bodewei, R., VonVoigtlander, P. F. and Müller,
M. Anticonvulsant and related effects of U-54494A in various
seizure tests. Journal of Pharmacology and Experimental
Therapeutics 1993; 267: 163–170.
34. Netzer, R. and Bigalke, H. Memantine reduces repetitive
action potential firing in spinal cord nerve cell cultures.
European Journal of Pharmacology 1990; 186: 149–155.
35. Binscheck, T., Netzer, R. and Bigalke, H. Anticonvulsant
drugs: quantification of effects in convulsant treated spinal
Anticonvulsant profile of propranolol
cord neurons. Naunyn-Schmiedebergs Archives of Pharmacol-
ogy 1990; 342 (Suppl.): R3.
36. Barrett, A. M. and Cullum, V. A. The biological properties
of the optical isomers of propranolol and their effects on
cardiac arrhythmias. British Journal of Pharmacology 1968;
34: 43–55.
37. Fischer, W. and Müller, M. Pharmacological modulation
of central monoaminergic systems and influence on the
anticonvulsant effectiveness of standard antiepileptics in
maximal electroshock seizure. Biomedica Biochimica Acta
1988; 47: 631–645.
38. Löscher, W. and Schmidt, D. Which animal models should be
used in the search for new antiepileptic drugs? A proposal
based on experimental and clinical considerations. Epilepsy
Research 1988; 2: 145–181.
39. Gellman, R. L., Kallianos, J. A. and McNamara, J. O. Alpha-2
receptors mediate an endogenous noradrenergic suppression
of kindling development. Journal of Pharmacology and
Experimental Therapeutics 1987; 241: 891–898.
40. Leszkovszky, G. and Tardos, L. Some effects of propranolol
on the central nervous system. The Journal of Pharmacy and
Pharmacology 1965; 17: 518–519.
41. Yeoh, P. N. and Wolf, H. H. Effects of some adrenergic
agents on low frequency electroshock seizures. Journal of
Pharmaceutical Sciences 1968; 57: 340–342.
42. Singh, K. P., Bhandari, D. S. and Mahawar, M. M. Effects of
propranolol (a beta adrenergic blocking agent) on some central
nervous system parameters. The Indian Journal of Medical
Research 1971; 59: 786–794.
43. Iyer, K. S., Govindankutty, A. and Radha, M. Central nervous
system pharmacology of propranolol. Indian Journal of
Physiology and Pharmacology 1975; 19: 152–156.
44. Dashputra, P. G., Patki, V. P. and Hemnani, T. J. Antiepileptic
action of beta-adrenergic blocking drugs: pronethalol and
propranolol. Materia Medica Polona 1985; 17: 88–92.
45. Raju, S. S., Gopalakrishna, H. N. and Venkatadri, N. Effect
of propranolol and nifedipine on maximal electroshock-
induced seizures in mice: individually and in combination.
Pharmacological Research 1998; 38: 449–452.
46. Madan, B. R. and Barar, F. S. K. Anticonvulsant activity
of some β-adrenoceptor blocking agents in mice. European
Journal of Pharmacology 1974; 29: 1–4.
47. Kellogg, C. Audiogenic seizures: relation to age and mech-
anisms of monoamine neurotransmission. Brain Research
1976; 106: 87–103.
48. Lints, C. E. and Nyquist-Battie, C. A possible role for
beta-adrenergic receptors in the expression of audiogenic
seizures. Pharmacology, Biochemistry and Behavior 1985; 22:
711–716.
49. Papanicolaou, J., Vajda, F. J. E., Summers, R. J. and Louis,
W. J. Role of β-adrenoceptors in the anticonvulsant effect
of propranolol on leptazol-induced convulsions in rats. The
Journal of Pharmacy and Pharmacology 1982; 34: 124–125.
50. Crowther, A. F. and Smith, L. H. ß-Adrenergic blocking
agents. II. Propranolol and related 3-amino-1-naphthoxy-
2-propanols. Journal of Medicinal Chemistry 1968; 11:
1009–1013.
51. Scriabine, A., Torchiana, M. L., Stavorski, J. M., Ludden,
C. T., Minsker, D. H. and Stone, C. A. Some cardiovascular
effects of timolol a new ß-adrenergic blocking agent. Archives
Internationales de Pharmacodynamie et de Therapie 1973;
205: 76–93.
52. Waal-Manning, H. J. Hypertension: which beta-blocker?
Drugs 1976; 12: 412–441.
53. Scott, J. A. and Crews, F. T. Rapid decrease in rat brain beta
adrenergic receptor binding during combined antidepressant
alpha-2 antagonist treatment. Journal of Pharmacology and
Experimental Therapeutics 1983; 224: 640–646.
54. Macdonald, R. L. Antiepileptic drug actions. Epilepsia 1989;
30 (Suppl. 1): S19–S28.
301
55. Rho, J. M. and Sankar, R. The pharmacological basis of
antiepileptic drug action. Epilepsia 1999; 40: 1471–1483.
56. White, H. S. Comparative anticonvulsant and mechanistic
profile of the established and newer antiepileptic drugs.
Epilepsia 1999; 40 (Suppl. 5): S2–S10.
57. Tarr, M., Luckstead, E. F., Jurewicz, P. A. and Haas,
H. G. Effect of propranolol on the fast inward sodium
current in frog atrial muscle. Journal of Pharmacology and
Experimental Therapeutics 1973; 184: 599–610.
58. Ban, T. A kinetic study of effects of propranolol and N-
propylajmaline on the rate of rise of action potential in guinea
pig papillary muscles. Japanese Journal of Pharmacology
1977; 27: 865–880.
59. Courtney, K. R. Fast frequency-dependent block of action
potential upstroke in rabbit atrium by small anesthetics. Life
Sciences 1979; 24: 1581–1588.
60. Nakaya, H., Kimura, S., Nakao, Y. and Kanno, M. Effects
of nipradilol (K-351) on the electrophysiological properties
of canine cardiac tissues: comparison with propranolol and
sotalol. European Journal of Pharmacology 1984; 104:
335–344.
61. Matthews, J. C. and Baker, J. K. Effects of propranolol and
a number of its analogues on sodium channels. Biochemical
Pharmacology 1982; 31: 1681–1685.
62. Grima, M., Freyss-Beguin, M., Millanvoye-Van Brussel, E.,
Decker, N. and Schwartz, J. Effects of various antianginal
drugs on sodium influx in rat brain synaptosomes and
in rat heart muscle cells in culture. European Journal of
Pharmacology 1987; 138: 1–8.
63. Chidlow, G., Melena, J. and Osborne, N. N. Betaxolol, a
ß1 -adrenoceptor antagonist, reduces Na+ influx into cortical
synaptosomes by direct interaction with Na+ channels:
comparison with other ß-adrenoceptor antagonists. British
Journal of Pharmacology 2000; 130: 759–766.
64. Catterall, W. A. Structure and function of voltage-gated ion
channels. Annual Review of Biochemistry 1995; 64: 493–531.
65. Fozzard, H. A. and Hanck, D. A. Structure and function
of voltage-dependent sodium channels: comparison of brain
II and cardiac isoforms. Physiological Reviews 1996; 76:
887–926.
66. Hartmann, H. S., Colom, L. V., Sutherland, M. L. and
Noebels, J. L. Selective localization of cardiac SCN5A sodium
channels in limbic regions of rat brain. Nature Neuroscience
1999; 2: 593–595.
67. Donahue, L. M., Coates, P. W., Lee, V. H., Ippensen, D. C.,
Arze, S. E. and Poduslo, S. E. The cardiac sodium channel
mRNA is expressed in the developing and adult rat and human
brain. Brain Research 2000; 887: 335–343.
68. Willow, M. Pharmacology of diphenylhydantoin and carbam-
azepine action on voltage-sensitive sodium channels. Trends in
Neurosciences 1986; 9: 147–149.
69. Wakamori, M., Kaneda, M., Oyama, Y. and Akaike, N. Effects
of chlordiazepoxide, chlorpromazine, diazepam, diphenyl-
hydantoin, flunitrazepam and haloperidol on the voltage-
dependent sodium current in isolated mammalian brain neu-
rons. Brain Research 1989; 494: 374–378.
70. Femmer, K., Bartsch, R., Leonhardt, U., von Littrow,
C. and Riedel, H. Zur Pharmakologie von (±)-1-[4-Cyclo-
hexylureidophenoxy]-2-hydroxy-3-tert.-butylaminopropran
(Talinolol, Cordanumr , 02-115). Pharmazie 1975; 30:
642–651.
71. Akaike, N., Ito, H., Nishi, K. and Oyama, Y. Further analysis
of inhibitory effects of propranolol and local anaesthetics on
the calcium current in Helix neurones. British Journal of
Pharmacology 1982; 76: 37–43.
72. Bodewei, R., Fischer, W., Hering, S. and Wollenberger,
A. Wirkung von (−)- und (+)-Propranolol auf den Kalzium-
Einwärtsstrom klonierter Neuroblastom x Gliom-
Hybridzellen. Biomedica Biochimica Acta 1988; 47:
233–238.
302
73. Lasek, R. and Wehran, H.-J. Zur Pharmakokinetik von
Propranolol. Pharmazie 1984; 39: 430–431.
74. Haeusler, G. Pharmacology of ß-blockers: classical aspects
and recent developments. Journal of Cardiovascular Pharma-
cology 1990; 16 (Suppl. 5): S1–S9.
75. Cruickshank, J. M. The clinical importance of cardioselectiv-
ity and lipophilicity in beta blockers. American Heart Journal
1980; 100: 160–178.
76. Kelly, D. Betablocker in der Psychiatrie—eine klini-
sche Übersicht. In: Der therapeutische Zugang zur Psyche
über das betaadrenerge System (Ed. P. Kielholz). Bern, Huber,
1978: pp. 35–58.
77. Kittner, H., Fischer, W. and Poelchen, W. Einfluß von
(+)- und (−)-Propranolol in höheren Dosierungen auf aus-
gewählte Herz-Kreislauf-Parameter bei der Ratte. Pharmazie
1991; 46: 470–471.
78. Pessina, A. C., Corgnati, A., Palatini, P., Oneglia, C. and
Dal Palù, C. The effects of propranolol on spontaneous
and experimental hypertension in the rat. Clinical Science and
Molecular Medicine 1976; 51: 447s–450s.
79. Karas, B. J., Wilder, B. J., Hammond, E. J. and Bauman, A. W.
Treatment of valproate tremors. Neurology 1983; 1380–1382.
80. Nemire, R. E., Toledo, C. A. and Ramsey, R. E. A
pharmacokinetic study to determine the drug interaction
W. Fischer
between valproate and propranolol. Pharmacotherapy 1996;
16: 1059–1062.
81. Wilder-Smith, E. Complete atrio-ventricular conduction block
during complex partial seizure. Journal of Neurology, Neuro-
surgery, and Psychiatry 1992; 55: 734–736.
82. Frysinger, R. C., Engel, J. and Harper, R. M. Interictal heart
rate patterns in partial seizure disorders. Neurology 1993; 43:
2136–2139.
83. Leestma, J. E., Kalelkar, M. B., Teas, S. S., Jay, G. W. and
Hughes, J. R. Sudden unexpected death associated with
seizures: analysis of 66 cases. Epilepsia 1984; 25: 84–88.
84. Klenerman, P., Sander, J. W. A. S. and Shorvon, S. D. Mor-
tality in patients with epilepsy: a study of patients in long
term residential care. Journal of Neurology, Neurosurgery, and
Psychiatry 1993; 56: 149–152.
85. Hoque, A. N., Nasa, Y. and Abiko, Y. Cardioprotective effect
of d-propranolol in ischemic-reperfused isolated rat hearts.
European Journal of Pharmacology 1993; 236: 269–277.
86. Nutt, D. Changes in seizure threshold following ECS
and bicuculline-induced convulsions. In: Neurotransmitters,
Seizures, and Epilepsy (Eds P. L. Morselli, K. G. Lloyd,
W. Löscher, B. S. Meldrum and E. H. Reynolds). New York,
Raven Press, 1981: pp. 117–128.