вход по аккаунту


Antiepileptogenic effects of conventional anticonvulsants in the kindling model of epilepsy.

код для вставкиСкачать
Antiepileptogenic Effects
of Conventional Anticonvulsants
in the Kindling Model of Epilepsy
Jon M. Silver, MD,* Cheolsu Shin, MD,@ and James 0. McNamara, MD?$§
We sought to determine whether the clinically effective anticonvulsant drug valproate exhibited antiepileptogenic
properties in the kindling model (we use the term anticonvulsant to mean suppression of seizure, and antiepileptogenic
to mean suppression of development of epilepsy). We compared and contrasted valproate with two other anticonvulsant
drugs, phenobarbital and carbamazepine. We investigated the effects of these drugs on the development of kindling,
that is, the number of stimulation-induced afterdischarges required to induce enhanced seizure susceptibility in rats.
Valproate exhibited powerful antiepileptogenic effects as evidem in a dose-dependent increase in the number of
afterdischarges required to induce kindling. These effects were not due to retained valproate or an active metabolite
merely masking the expression of kindled seizures. By contrast, carbamazepine was devoid of any antiepileptogenic
effects despite exhibiting marked anticonvulsant effects. Like valproate, phenobarbital exhibited both antiepileptogenic and anticonvulsant properties, but its antiepileptogenic properties were significantly less pronounced. The
antiepileptogenic effects of valproate and phenobarbital strengthen the candidacy of these agents for the clinical
studies needed to investigate pharmacological prevention of the development of epilepsy in high-risk groups.
Silver JM, Shin C, McNamara JO. Antiepileptogenic effects of conventional anticonvulsants
in the kindling model of epilepsy. Ann Neurol 1991;29:356-363
The principal contemporary therapeutic approach to
epilepsy utilizes chronic administration of anticonvulsant drugs to inhibit the occurrence of seizures. This
therapy is often unsatisfactory since seizures persist in
at least 25% of patients and undesirable side effects
are common [ 1). Preventing epileptogenesis, the process of the development of epilepsy, would obviate the
need for chronic anticonvulsant therapy. In patients
who have a brain injury associated with a high risk of
epilepsy occurring months to years later [2], it may be
possible to intervene pharmacologically to prevent the
development of epilepsy. No drug has been demonstrated to exhibit antiepileptogenic efficacy in a randomized, double-blinded trial in humans [3}. Demonstration of antiepileptogenic properties in an animal
model could facilitate selection of an effective drug for
controlled clinical trials.
The kindling model provides a simple means of
quantitating epileptogenesis. Kindling refers to a phenomenon in which periodic application of an initially
subconvulsive electrical stimulus induces progressive
intensification of evoked electrographic and behavioral
seizures 14). Once established, this enhanced sensitivity to electrical stimulation appears to be permanent
for the life of the animal. A convenient measure of
epileptogenesis is the number of stimulation-induced
afterdischarges (ADS) or electrographic seizures required to induce enhanced seizure susceptibility. This
model has been exploited to demonstrate antiepileptogenic effects of the clinically used anticonvulsants clonazepam and phenobarbital as well as the experimental
drug gamma-vinyl gamma-aminobutyric acid (GABA)
I5-81. The goals of the present study were to determine whether valproate exhibited antiepileptogenic
properties in the kindling model and to compare and
contrast valproate with phenobarbital and carbamazepine.
From the Departments of "Surgery (Neurosurgery), tMedicine
(Neurology), and $Pharmacology, Duke University Medical Center,
and the Epilepsy Research Laboracorn, Veterans Administration
Medical Center, Durham, NC.
Received Jul 12, 1990, and in revised form Oct 10. Accepted for
publication Oct 11, 1990.
Address correspondellce Dr McNamara, Building 16, Room 20,
V.A. Medical Center, 508 Fulton St, Durham, NC 27705.
Materials and Methods
Animal Preparation
Male Sprague-Dawley rats (Charles River, Wilmington, DE)
weighing 300 to 430 gm were anesthetized with intraperitoneal injections of pentobarbital (50 mg/kg) and placed in a
stereotaxic frame. Bipolar electrodes (twisted Nichrome
wire, insulated except at the tips) were placed in the right
amygdala stereotaxically (anteroposterior - 0.8 mm from
bregma, mediolateral +4.8 mm, dorsoventral -8.5 rnm
from dura; nose bar + 5 mm). Four skull screws were placed
to serve as reference electrodes and the entire assembly was
356 Copyright 0 1991 by the American Neurological Association
attached to the skull with dental acrylic. Throughout the experiment, animals were housed with a 12-hour lighdl2-hour
dark cycle and allowed free access to rat chow and water.
Kindling Parameters
Stimulations were delivered to the right amygdala through
two Grass PSIU-6 constant-current sources and a Grass S-88
stimulator (Quincy, MA). Each stimulation consisted of a
1-second train of I-millisecond biphasic, rectangular pulses
at 60 Hz. The current intensity was monitored by measuring
the voltage drop across a 10,000-ohm resistor in series with
the animal. Afterdischarge duration (ADD) was determined
by analysis of the electroencephalograph (EEG) recorded
from the amygdala and reference electrodes. The seizures
were classified according to Racine [9]:class I, facial clonus;
2 , head nodding; 3, forelimb clonus; 4 , rearing with bilateral
forelimb clonus; and 5 , rearing and falling with bilateral forelimb clonus.
ADD, and seizure class were recorded. The same procedure
was repeated on days 5 and 7 so that each animal received
all three treatments.
The anticonvulsant effects of phenobarbital were measured
in a second group that consisted of animals that received the
vehicle in Group 3 above (vehicle for phenobarbital). A similar paradigm was used to determine the seizure threshold
with the following changes: After the initial stimulation on
day 1, rats were divided into two groups. Rats in each group
received either phenobarbital (40 mg/kg, n = 3) or vehicle
(PPG, EtOH, and water, n = 4 ) and were tested as above.
Following a 21-day rest period to permit complete clearance
of phenobarbital [lo], animals were again stimulated at 100
FA above the initial ADT to verify the persistence of a class
5 seizure. Forty-eight hours thereafter, either drug or vehicle
was administered 30 minutes prior to determination of ADT
so that all animals received both treatments.
Electrode Localization
Meusarement of Antiepileptogenic Effects
The antiepileptogenic effects of the various drugs were measured by examining their abilities to suppress the rate of
kindling development. After at least 1 week of postsurgical
recovery, afterdischarge thresholds (ADT) were determined
by administering a stimulation at 100 pA with increments of
100 p.A every minute until an A D was recorded. Following
this, on days 2 through 10,animals were given intraperitoneal
injections of a drug or its vehicle 30 minutes prior to stimulation at 100 pA above the ADT. The three different treatment groups were: Group 1: valproate in saline solution (with
p H corrected to 7.0-7.4) at 100 mg/kg (n = 6), 200 mgi
kg (n = 5), or 400 mdkg (n = 4 ) or saline solution alone
(n = 6 ) ;Group 2: carbamazepine in propylene glycol (PPG),
ethanol (EtOH), and water (40:10:50)at 25 mgikg (n = 7),
or the vehicle alone (n = 6); and Group 3: phenobarbital in
PPG, EtOH, and water (40:10:50) at 40 mgikg (n = 7 ) ,
or the vehicle alone (n = 7 ) . All drugs and vehicle were
delivered in a volume of 1 m u g . Following 9 days of drug
treatment, animals were given a 2-week rest, after which
stimulations were resumed at the same current without
administration of drug or vehicle. This experimental design
permitted distinguishing drugs truly acting to retard kindling
development from those simply masking the expression of
kindled seizures. Stimulations were continued until three
class 5 seizures were evoked.
At the completion of the experiments, animals were anesthetized with ether and decapitated. The brains were frozen
in isopentane chilled in a dry ice and methanol bath, and
16-prn-thick frozen sections were thaw mounted and stained
with methyl green-pyronine Y for histological examination.
Only those animals with electrodes in the amygdala complex
were included in the final analysis.
Valproate and carhamazepine were gifts from Abbott Laboratories (North Chicago, IL) and Ciba-Geigy (Summit,
NJ), respectively. Phenobarbital and pentobarbital were purchased from Sigma Chemicals (St Louis, MO).
Statistical Medwds
Statistical analyses were performed using one-way analysis of
variance (ANOVA) with post hoc Tukey’s test for multiple
treatment groups versus control and Student’s two-tailed t
tests for single treatment groups versus control, with the
exception of analysis of seizure class. For this nonparametric
analysis, a Kruskal-Wallis AN-OVA was used with post hoc
Mann-Whitney U test for multiple treatment groups versus
control and Mann-Whitney U test for single treatment
groups versus control. Statistical significance was defined as
p < 0.05. Values are expressed as mean standard error of
the mean in the text.
Measurement of Anticonvzllsunt Eflects
Two groups of animals were used in these experiments. The
anticonvulsant effects of valproate and carbamazepine were
measured in one group consisting of the 6 animals that received vehicle in Group 2 above (carbamazepine vehicle).
After their third class 5 seizure, these animals were allowed
at least 2 weeks of rest. Then, on day 1, each animal was
stimulated at 100 pA above the initial ADT to verify that a
class 5 seizure could still be elicited. On day 3, animals were
given intraperitoneal injections of either valproate (200 mg/
kg, n = 2), carbamazepine ( 2 5 mgikg, n = 2), or the PPG,
EtOH, and water vehicle (n = 2) 30 minutes prior to testing.
A 20-ll.A current was then delivered and increased by 20 pA
every minute until A D was elicited; the threshold current,
Antiepileptogenic Effect of Vulproate
Valproate profoundly inhibited kindling development
(Fig 1). Valproate inhibited the steady increase in
ADD observed in the vehicle-treated animals in a
dose-dependent manner during the 9 days of its administration (see Fig 1A). T h e effects o n ADD were
paralleled by profound and dose-dependent inhibitory effects o n behavioral seizure class during valproate
treatment (see Fig 1B).
The effects of valproate were not simply due t o inhibition of seizure expression, since stimulation 11, administered 2 weeks after the last dose, when the drug
Silver et al: Antiepileptogenic Drugs
100 80
4 -
4 -
3 -
Fig I . Effects of various drugs on afterdischarge (AD) duration
and on seizure class with each A D during kindling development.
During the period between two vertical dashed lines, animals were
treated with drug or vehicle. Following the last A D during drug
or vehicle treatment (stimulation lo), animals were allowed a
2-week rest to clear the drug fmm their system, before resumption
o f kindling without any treatment (stimulation 11 and onward.
(A) Effect of valgroate (100 (white triangles),200 (white diamonds), or 400 (white circles) mglkd or saline (white squares)
on A D duration. (B) Effect of valgroate (100 (white triangles),
200 (white diamonds), or 400 (white circles) mglkg) or saline
(white squares) on seizure class. (C)Effect of 40 mglkg of pheno-
barbital(white triangles)or vehicle (white squares)on A D duration. (D) Effect of 40 mglkg of phenobarbital (white triangles)
or vehicle (white squares) on seizure class. (E) Eflect of 25 mgl
kg of carbamazepine (white triangles) or vehicle (white
squares) on A D duration. (F) ERect of 25 mgikg of carhamazepine (white triangles)or vehicle (white squares) on seizure class.
BLcR symbols denote statistically significant differences compared
with vehicle control (p < 0.05; ANOVA with post hoc Tukey
test (A) or t test (C and E), Kruskal-Wallis ANOVA with post
hoc Mann-Whitney U test (B) or Mann-Whitnqy U test (D
and F)) .
presumably had been cleared from the animal, failed
to evoke class 5 seizures as exhibited by vehicle-treated
animals (see Fig 1B). Likewise the ADD tended to
remain suppressed on the first stimulation after drug
cessation (stimulation 11)in all three treatment groups,
although on subsequent stimulation ADD lengthened
rapidly to reach the duration seen in the vehicle group.
Thus, the behavioral and EEG patterns of kindling development following discontinuation of valproate mimicked those of the vehicle-treated animals.
Another way to analyze the antiepileptogenic effect of
valproate is to calculate the number of drug-free ADS
(i.e., for the drug-treated group, the initial AD used to
determine ADT before the start of drug plus the ADS
after the end of drug period) to reach a particular stage.
If the numbers are the same for both vehicle- and
drug-treated animals, then the drug is effective in suppressing epileptogenesis completely during its administration. If the number of drug-free ADS is less than
that in vehicle-treated animals, some epileptogenesis
took place during the drug administration. Comparison
of the number of drug-free ADS required to induce
358 Annals of Neurology
Vol 29 N o 4
April 1991
Carbamszepine and Kindling Development
Vrlproic Acid (mg/kg)
Phonobrrbitrl (mg/kg)
Fig 1. Continued.
class 2 seizures within three valproate dose groups disclosed a trend of dose-dependent increase (Fig 2A).
Remarkably, however, the similarity of these values in
the vehicle- and drugtreated groups indicates that the
ADS (stimulations 2 through 10) evoked in the presence of valproate (drug-treated ADS) had virtually no
epileptogenic effects.
Residual Eflect of Vu@route
In a separate group of animals, we addressed the possibility that significant retention of valproate 2 weeks
after the last dose may account for the profound antiepileptogenic effects that were observed. After determination of the initial ADT on day 1, animals received
either vehicle (n = 6) or valproate (200 mg/kg, n =
9) once daily on days 2 through 10 withont kindling
stimulation. Animals were then rested for 2 weeks. Following this 2-week drug-free interval, blood samples
were obtained from 3 of the 9 drug-treated animals.
Serum valproate levels were undetectable.
Since a functionally effective pool of retained valproate in the brain or an active metabolite persisting
after this drug-free interval will escape detection in
Crrbamrzrpina (rng/kg)
Fig 2. Number of drug-free aferdischarges (ADS)required to
reach the first class 2 seizure for different treatment groups s h w n
in Figure 1. The number of drug-free ADS in the drugtreated
animals was calculated by adding the initial A D used to determine the A D threshold t o the number ofADs evoked in the absence
of drug after the 2-week rest. The control values (0-mglkggroup)
rdect the number of ADs to reach thejrst class 2 seizure in the
vehide-treatedgmup. The asterisk refers to p < 0.05 persus control
(ANOVA with post hoc Tukey’s test (A) or Student’s two-tailed
t test (B and C)). NS = no significant difference vmsus control.
Silver et ai: Antiepiieptogenic Drugs
serum, we measured the rate of kindling development
in the vehicle- and drug-treated groups with stimulations commencing after the 2-week drug-free interval.
N o significant differences were found in the number
of stimulations required to elicit three class 5 seizures
(vehicle-treated group, 13.7 2 2.6; drug-treated group,
12.5 2.0). Thus, no residual effect could be detected
2 weeks after the last dose of valproate, even with this
functional assay.
Antiepileptogenic E f f t of Phenobarbital
Phenobarbital (40 mg/kg) also inhibited kindling development. The steady increase in ADD and behavioral
seizure class was markedly inhibited during the administration of phenobarbital (see Figs 1C and 1D). These
effects were most apparent in the 65% increase in the
number of stimulation-induced ADSrequired to evoke
class 2 seizures (13.7 k 0.4 versus 8.3 ? 0.5 in
vehicle-treated rats; p < 0.05, Student’s two-tailed t
The effects of phenobarbital were not due to masking seizure expression, since stimulation 11, administered 2 weeks after the last dose when the drug presumably had been effectively cleared from the animal
[lo], evoked abbreviated ADD and milder behavioral
seizures compared to the vehicle-treated animals. The
behavioral and EEG patterns of kindling development
following discontinuation of phenobarbital mimicked
that of the valproate-treated animals.
When the number of drug-free ADS to reach class 2
seizure was calculated, a modest (approximately 50%)
reduction was found in the phenobarbital-treated
group (see Fig 2B). This suggests that in contrast to
valproate, some degree of epileptogenesis may occur
even in the presence of phenobarbital.
The antiepileptogenic efficacies of valproate and
phenobarbital were compared further by analyzing the
number of ADS required to reach a particular kindled
class, in comparison with their respective control
groups. In order to reach the first class 2 seizure, animals treated with 200 mg/kg of valproate required 7.6
k 1.11more ADS than the vehicle-treated control animals, while those treated with phenobarbital (40 mg/
kg) required 5.4 ? 0.36 more than their vehicletreated controls @ > 0.05). On the other hand, in
valproate-treated animals, 7.6 ? 0.97 more ADS were
required to reach the third class 5 seizure, compared
with only 2.97 +- 0.75 more ADS for phenobarbitaltreated animals, than were required in the respective
controls @ < 0.01). This indicates a significantly
greater antiepileptogenic effect for valproate than for
Antiepileptogenic Efiect of Carbamazepine
In contrast to valproate and phenobarbital, carbamazepine exhibited no antiepileptogenic effects. There
360 Annals of Neurology Vol 29 No 4 April 1991
was no detectable inhibition of increase in ADD or
behavioral seizure class either during the carbamazepine treatment or after its discontinuation (see Figs 1E
and 1F). Similarly, no inhibitory effects were apparent
in the total number of stimulations required to elicit
the first class 2 (drug-treated group, 8.6 5 0.9;
vehicle-treated group, 8.7
1.4) or third class 5 seizure (drug-treated group, 13.6 2 1.1; vehicle-treated
group, 14.5 +. 2.1). With carbamazepine therapy, the
number of drug-free ADS required to induce a class 2
seizure was dramatically reduced (see Fig 2C), suggesting that epileptogenesis is not impeded at all during
carbamazepine administration.
Anticonvdsunt Effects
TOdemonstrate that the doses used in examining the
antiepileptogenic properties produced equivalent anticonvulsant effects once the animals were kindled, we
measured the effects of the three drugs on fully kindled seizures. Each of the three drugs exhibited powerful anticonvulsant effects as evident in the marked
inhibition of behavioral seizure class and ADD in comparison with vehicle-treated animals (Tables 1 and 2).
Phenobarbital produced the most profound anticonvulsant effect as apparent in its reduction of ADD by
93% (see Table 2). The effects of carbamazepine and
valproate were similar but less marked. None of the
three affected the ADT significantly. Of note, however, is the observation that the ADT in these fully
kindled animals were much lower than the initial ADT
determined prior to the start of the kindling protocol,
a phenomenon known to occur in kindling [111.
S i d Efiects
Powerful antiepileptogenic effects of valproate were
apparent with minimal or no persistent behavioral toxicity. The two lower doses (100 and 200 mg/kg) produced mild, transient lethargy or ataxia after the first
few doses, but thereafter animals exhibited no untoward effects. At the end of the 9-day treatment period,
animals in these groups had gained weight in a manner
indistinguishablefrom the vehicle group (Fig 3A), with
the group receiving 100 mg/kg gaining 10 gm and the
group receiving 200 mg/kg gaining 8 gm, versus a gain
of 22.5 gm in the vehicle-treated group (no significant
difference by ANOVA). The highest dose (400 mg/
kg), however, induced marked lethargy, ataxia, and a
42.5-gm weight loss (p < 0.05 versus control); two of
the animals receiving this dose died during the experiment and these were excluded from the final data
Phenobarbital (40 mg/kg administered intraperitoneally) usually produced mild ataxia with some lethargy.
Impairment of weight gain was not apparent in these
animals, compared with their control group (see Fig
3B). Carbamazepine resulted in minimal ataxia or leth-
Table I . Anticonvulsant Effects of Valproate and Carbamazepine
Threshold (FA)
Duration (sec)
Seizure Class
Valproate (200 mgikg)
190 t 23.5 (NS)
173.3 ? 15.2 {NS)
143.3 ? 12.0
37.7 i 9.2”
42.5 k 11.4”
77.2 6.5
1.3 t 0.9b
1.8 f 1.0b
5.0 2 0.0
ap < 0.05 versus control; one-way ANOVA with post hoc Tukey’s test.
‘p < 0.05 versus control; Kruskal-Wallis ANOVA with post hoc Mann-Whicney (Ianalysis.
NS = not signhcant; p > 0.05 versus control; one-way ANOVA.
Table 2. Anticonvulsant Effects of Phenobarbital
Threshold (PA)
Duration (sec)
40 mdkg
213.3 5 23.4 (NS)
216.7 2 17.4
4.8 2
Seizure Class
‘p < 0.05 versus control; Student’s two-tailed f test.
bp< 0.05 versus control; Mann-Whitney IJ analysis.
NS = not significant;p > 0.05 versus control; Student’s two-tailed t test
argy only with the first few doses and did not interfere
with weight gain (see Fig 3C).
The principal findings of this study are: (1) Valproate
prevented kindling development, an action not simply
due to valproate or an active metabolite masking the
expression of kindled seizures. (2) Phenobarbital also
inhibited kindling development to a modest extent that
was less pronounced than that of valproate. ( 3 ) In contrast, carbamazepine did not inhibit kindling development. (4)This antiepileptogenic effect of valproate was
apparent at a dose that provided anticonvulsant effects
equal to or even less than those of other drugs.
Our results confirm and extend previous findings
15, 121 that administration of valproate prior to each
stimulation inhibits the progressive intensification of
AD and behavioral seizure. Importantly, our results
demonstrate that this is truly an antiepileptogenic effect, by using an experimental design that excludes the
possibility that valproate or an active metabolite simply
masked the expression of kindled seizures through an
anticonvulsant action. We confirm the absence of detectable antiepileptogenic effects of carbamazepine 15,
13) despite the use of doses with anticonvulsant effects
at least equivalent to valproate. We also confirm the
antiepileptogenic effects of phenobarbital, but find less
robust effects than with valproate.
The cellular mechanism by which valproate exerts
this antiepileptogenic action is uncertain. One mechanism by which valproate may suppress seizures involves inhibition of sustained repetitive firing, an effect
mediated through an action on the voltage-dependent
sodium channel 114). This cannot readily explain its
antiepileptogenic action, since two other drugs that
also inhibit sustained repetitive firing, phenytoin and
carbamazepine, fail to inhibit kindling development 15,
7 , 8 , 13, 15, 161. An enhancement of inhibitory synaptic transmission 117) andor inhibition of excitatory
synaptic transmission [l8] by valproate may be responsible. The antiepileptogenic effects of drugs that selectively enhance inhibitory 16, 8) or inhibit excitatory
119 ) synaptic transmission support this idea.
If similar mechanisms underlie epileptogenesis in
kindling and some forms of human epilepsy, the present findings may have important implications for either
prevention of human epilepsy or attenuation of its progression. Prophylaxis may be possible in instances in
which epilepsy arises months to years after the initial
brain injury such as in penetrating head injury, abscess,
and embolic stroke [2, 201. Our findings may also
prove relevant to the choice of an anticonvulsant in a
patient with recent onset of epilepsy, since some patients become refractory to medication after an initial
period of responsiveness [ 2 11. Use of a drug with both
anticonvulsant and antiepileptogenic properties may
inhibit progression of the epileptic condition and thus
be preferable to a drug with only anticonvulsant properties in such patients.
The present findings underscore the need for properly designed clinical studies to examine the efficacy of
putative prophylactic agents to prevent epileptogenesis. Obviously one must consider the toxicity of the
agents to be used; that is, with valproate, there are
hematological, hepatic, and neurological side effects
1227. However, the risk of development of epikpsy,
Silver et al: Antiepileptogenic Drugs
200 mglkg
400 tng1tg
0 ,
with its medical and nonmedical implications, may be
high enough in a select group of patients (possibly
more than 30% in patients with penetrating head injury or brain abscess) to warrant prophylactic therapy
{a, 20). Only a randomized, placebo-controlled,
double-blind study can adequately evaluate the efficacy
and toxicity of the prophylactic regimen. One such
srudy tested the effect of phenytoin with therapeutic
serum levels and failed to detect any long-term antiepileptogenic effect { 31. Interestingly, phenytoin also
failed to inhibit development of kindling in animals
{ S , 8, 151. The powerful antiepileptogenic effects of
valproate and, to a lesser degree, phenobarbital in the
kindling model strengrhen the candidacy of these
agents for future clinical studies investigating pharmacological prevention of epilepsy in high-risk groups.
This study was supported by grants from the Veterans Administration and the National Inscitures of Health (NS17771).
1. Porter RJ. Epilepsy: 100 elementary principles. 2nd ed. London:
WB Saunders, 1989
2. Salazar AM, Jabbari H, Vance SC, et al. Epilepsy after penetrating head injury. I. Clinical correlates: a report of the Vietnam
Head Injury Study. Neurology 1985;35:1406-1414
3. Temkin NR, Dikmen SS, Wilensky AJ, et al. A randomized,
double-blind study uf phenyroin for the prevention of posttraumatic seizures. N Engl J Med 1990;323:497-502
4. Goddard GV, Mclnryre DC, Leech CK A permanent change
in brain function resulting from daily electrical stimulation. Exp
Neurol 1969;25:295-330
5. Schmutz M, Klebs K, Baltzer V. Inhibition or enhancement of
kindling evolution by ancicpileptics. J Neural Transm 1988;
6. Shin C, Rigsbee LC, McNamara JO. Anti-seizure and anti-
1 0.
0 0
Fig 3. Mean weight (gal daring drag treatment. (A) Valproate
(100 (white triangles), 200 (white diamonds), or 400 (white
circles) mglkg) or saline (white squares). (B) Phenobarbital, 40
mg/kg {white triangles) or vehicle (white squares). (C) Carbamazepine, 2S rnglkg (white triangles) or vehicle (white squares).
Solidsymbols refer t o p < 0.05 versus control (ANOVA with post
bor Ttlkq test iAj or Student's two-tailed t test (B and C)).
362 Annals of Neurology Vol 29 No 4 April 1991
epileptogenic effect of gamma-vinyl-GABA in amygdaloid kindling. Brain Res 1986;398:370-374
Wada JA. Pharmacological prophylaxis in the kindling model of
epilepsy. Arch Neurol 1977;34:389-395
Wise RA, Chinerman J. Effects of diazepam and phenobarbital
on electrically-inducedamygdaloid seizures and seizure development. Exp Neurol 1974;45:355-363
Racinc R]. Modification of seizure activity by electrical stimulation. 11. Motor seizure. Electroencephalogr Clin Neurophysiol
Toon S, Rowland M. Structure-pharmacokinecic relationships
among the barbiturates in the rat. J Pharmacol Exp Ther 1983;
k i n e RJ. Modification of seizure activity by electrical stimulation: I. After-discharge rhreshold. Electroencephalogr Clin Neurophysiol 1972;32:269-279
Leviel V, Naquet R. A study of the action of valproic acid on
the kindling effect. Epikpsia 1977;18:229-233
Albertson TE, Joy R, Stark LG. Carbamazepine: a pharmacological study in the kindling model of epilepsy. Neuropharmacology 1984;23:1117-1 123
Macdonald RL, Meldrum BS. Principles of antiepileptic drug
action. In: Levy R, Mattson R, Meldrum B, et al., eds. Antiepileptic drugs. 3rd ed. New York Raven Press, 198959-83
Racine RJ, Livingston K, Joaquin A. Effects of procaine hydrochloride, diazepam and diphenylhydantoin on seizure dcvelopment in cortical and subcortical structures in rats. Electroencephalogr Clin Neurophysiol 1975;38:355-365
16. Wada JA, Osawa T, Sat0 M, ct al. Acute anticonvulsant effects
of diphenylhydantoin, phenobarbital, and carbamazepine: a
combined electroclinical and serum level study in amygdaloid
kindled cats and baboons. Epilepsia 1976;17:77-88
17. Preisendtirfer U, Zeise ML, N e e MR. Valproate enhances inhibitory postsynaptic potentials in hippocampal neurons in vitro.
Brain Res 1987;435:213-219
18. Griffith WH, Taylor L. Sodium valproate decreases synaptic
potentiation and epileptiform activity in hippocampus. Brain
Res 1988;474:155-164
19. McNamara JO, Russell RD, Rigsbee LC, et al. Anticonvulsant
and antiepileptogenic actions of MK-801 in the kindling
and electroshock models. Neuropharmacology 1988;27:563568
20. Foy PM, Copeland GP, Shaw MDM. The natural history of
postoperative seizures. Acta Neurochir 1981;57:15-22
21. Elwes RDC, Johnson AL, Shorvon SD, Reynolds EH. The prognosis for seizure control in newly diagnosed epilepsy. N Engl J
Med 1984;311:944-947
22. Dreifuss FE. Valproate toxicity. In: Levy R, Mattson R, Meldrum B, et d., eds. Antiepileptic drugs. 3rd ed. New York:
h v e n Press, 1989643-65 1
In the February issue of Annals, Figures 3 and 4 (pp
197, 198) in the article by Kira and colleagues ( Kra
J-i, Koyanagi Y, Yamada T, Itoyama Y , Goto I, Yamamot0 N, Sasaki H, Sakaki Y . Increased HTLV-I proviral DNA in HTLV-I-associated myelopathy: a quantitative polymerase chain reaction study. Ann Neurol
1991;29:194-201) were inadvertently printed with
position labels on them. The figures are reprinted below. Ann& apologizes for this error.
Fig 3. Quantitative polymerase chain reaction analysis of the Pol
D N A of samples. (A)Thepmducts afer 25 lycles. 03) The prodacts afer 35 cycles. The same samples in Figure 2 were run in
the same or& as in this J5gu.e. The pol D N A in all patients
with HAMITSP und carriers without HAMITSP who had other
autoimmune or inJlammutory diseases were evident after 25 cycles
of amplijkation.
Fig 4. Polymerase chain reaction analy..ris of the pglobin D N A
of samples. The samples are the same as in Figures 2 and 3. The
amounts of the pglobin D N A in samples were determined, as
refmed to the intensity of the same gene in the serially diluted
MT2 DNA. Severalsamples with high D N A contents, including
MT2 DNA, sometimes gave an aditional band. However, the
intensity of the band of the predicted size in serially diluted M T 2
ivespective of an additionalband,
and dependedon the amount of D N A input.
Silver et al: Antiepileptogenic Drugs
Без категории
Размер файла
760 Кб
effect, mode, anticonvulsant, convention, epilepsy, kindling, antiepileptogenic
Пожаловаться на содержимое документа