Epilepsy Research 31 (1998) 175 – 186
Limbic epileptogenesis alters the anticonvulsant efficacy of phenytoin in Sprague–Dawley rats Wolfgang Lo¨scher *, Sybille Cramer, Ulrich Ebert Department of Pharmacology, Toxicology and Pharmacy, School of Veterinary Medicine, Bu¨nteweg 17, D-30559 Hanno6er, Germany Received 23 February 1998; received in revised form 23 April 1998; accepted 27 April 1998
Abstract Studies on the anticonvulsant efficacy of the major antiepileptic drug phenytoin in kindled rats have often reported inconsistent effects. It has been proposed that technical and genetic factors or poor and variable absorption of phenytoin after i.p. or oral administration may be involved in the lack of consistent anticonvulsant activity of phenytoin in this model of temporal lobe epilepsy. We examined if kindling itself changes the anticonvulsant efficacy of phenytoin by testing this drug before and after amygdala kindling in male and female Sprague – Dawley rats. To exclude the possible bias of poor and variable absorption, blood was sampled in all experiments for drug analysis in plasma. The threshold for induction of focal seizures (afterdischarge threshold; ADT) was used for determining phenytoin’s anticonvulsant activity. Before kindling, phenytoin, 75 mg/kg i.p., markedly increased ADT in both genders, although the effect was more pronounced in males. Following kindling, the anticonvulsant activity obtained with phenytoin, 75 mg/kg, before kindling was totally lost, and female rats even exhibited a proconvulsant effect upon administration of this dose, indicating that kindling had dramatically altered the anticonvulsant efficacy of phenytoin. Plasma levels of phenytoin were comparable before and after kindling, and were within or near to the ‘therapeutic range’ known from epileptic patients. When the dose of phenytoin was reduced to 50 or 25 mg/kg i.p., significant anticonvulsant effects on ADT were obtained. When phenytoin, 50 mg/kg, was administered i.p. or i.v. in the same group of fully kindled rats, both anticonvulsant activity and plasma drug levels were comparable with both routes, indicating that the i.p. route is suited for such studies. The data indicate that kindling alters the dose – response of phenytoin in that a high anticonvulsant dose becomes ineffective or proconvulsant after kindling, possibly by an increased sensitivity of the kindled brain to proconvulsant effects of phenytoin which normally only occur at much higher doses. If similar alterations evolve in humans during development of chronic epilepsy, this may be involved in the mechanisms leading to intractability of temporal lobe epilepsy. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Kindling; Epilepsy; Antiepileptic drugs; Seizures
* Corresponding author. Tel.: +49 511 8568721; fax: + 49 511 9538581. 0920-1211/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0920-1211(98)00029-1
176
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
1. Introduction
2. Materials and methods
The pathophysiological mechanisms that lead to chronic, often intractable epilepsy are not known (Schmidt and Morselli, 1986; Heinemann et al., 1994; Johannessen et al., 1995; Lo¨scher, 1997). Particularly in temporal lobe epilepsy, many patients become pharmaco-resistant during treatment or are resistant to available antiepileptic drugs from the onset of first treatment (Schmidt and Morselli, 1986; Johannessen et al., 1995). Thus, there is a need for development of more effective antiepileptic drugs for this most common type of human epilepsy (Lo¨scher and Schmidt, 1994); however, this is a difficult goal in the absence of any information of the underlying causes why temporal lobe epilepsy is so often intractable. We have previously proposed that the amygdala kindling model in rats might be suited to study mechanisms of drug resistance in temporal lobe epilepsy, because (a) kindled rats are less sensitive to anticonvulsant effects of antiepileptics than non-kindled rats (Lo¨scher et al., 1986; Lo¨scher and Schmidt, 1988), and (b) a subgroup of kindled rats in which focal seizures are resistant to phenytoin and other major antiepileptic drugs can be selected from large groups of kindled Wistar rats (Lo¨scher and Rundfeldt, 1991; Lo¨scher et al., 1993). This indicates that both kindling itself and individual genetic factors are involved in the low susceptibility of kindled rats to antiepileptics. In order to address the influence of kindling on the susceptibility of rats to anticonvulsant drug effects, we undertook experiments in which phenytoin was administered in the same rats before and after kindling. We chose Sprague–Dawley rats for this purpose, because this outbred rat strain has been used in most previous experimental studies on the kindling model. In addition to comparing phenytoin’s anticonvulsant efficacy before and after kindling, we also examined the influence of route of administration (i.p. versus i.v.), because McNamara et al. (1989) previously reported that phenytoin is much more effective by the i.v. route, which could explain inconsistent results obtained with i.p. administration of phenytoin in the kindling model.
2.1. Animals Female and male Sprague–Dawley rats (Harlan–Winkelmann, Borchen, Germany), weighing 200–250 g (females) or 240–360 g (males), were used. The animals were purchased from the breeder at an age of 10 weeks. Following arrival in the animal colony, the rats were kept under controlled environmental conditions (ambient temperature 24–25°C, humidity 50–60%, 12/12-h light/dark cycle, light on at 07:00 h) for at least 1 week before being used in the experiments. Standard laboratory chow (Altromin 1324 standard diet) and tap water were allowed ad lib. All experiments were done in the morning to minimize the bias of circadian variations. All animal care and handling was conducted in compliance with the German Animal Welfare Act and was approved by the responsible governmental agency in Hannover.
2.2. Electrode implantation For implantation of kindling electrodes, rats were anesthetized with chloral hydrate (360 mg/kg i.p.) and a bipolar electrode was implanted into the right hemisphere aimed at the basolateral amygdala using the following coordinates (in mm relative to bregma) derived from the atlas of Paxinos and Watson (1986): AP − 2.2, L − 4.7, V − 8.7; incisor bar at − 3.9. These coordinates were based on preliminary experiments in Sprague–Dawley rats of both sexes in which subsequent histological verification of electrode placement showed a correct location of electrode tip in the basolateral amygdala under these coordinates. The electrode consisted of two twisted Teflon-coated 0.2-mm diameter stainless steel wires separated by 0.5 mm at the tip. A stainless steel screw, which served as neutral electrode, was positioned over the contralateral parietal cortex. Bipolar and ground electrodes were connected to plugs and the electrode assembly and anchor screws were held in place with dental acrylic cement applied to the exposed skull surface.
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
2.3. Kindling procedure and testing of phenytoin Electrical stimulation of the amygdala was initiated following a recovery period of 2 weeks after surgery. At the first day of the stimulation period, the threshold for induction of amygdalar afterdischarges (ADT) was determined 1 h after i.p. injection of saline (eight males, nine females; group 2) or phenytoin, 75 mg/kg (nine males, 11 females; group 1), i.e. a dose which has previously been shown to significantly increase the ADT 1 h after i.p. injection (Rundfeldt et al., 1990). ADT was determined by a staircase procedure, i.e. by administering a series of stimulations at 1-min intervals beginning at 10 mA and increasing in steps of about 20% of the previous current (Freeman and Jarvis, 1981). The threshold was defined as the lowest intensity producing afterdischarges with a duration of at least 3 s. Seizure severity, seizure duration and afterdischarge duration occurring at the seizure threshold current were recorded at all ADT determinations. Seizure severity was classified behaviorally according to Racine (1972): (1) immobility, eye closure, twitching of vibrissae, sniffing, facial clonus; (2) head nodding associated with more severe facial clonus; (3) clonus of one forelimb; (4) rearing, often accompanied by bilateral forelimb clonus; (5) all of the above plus loss of balance and falling, accompanied by generalized clonic seizures. Seizure duration was the duration of limbic (stage 1 – 2) and motor seizures (stage 3 – 5); limbic seizure activity (immobility associated with lowamplitude afterdischarges and occasional facial clonus or head nodding) often occurring after termination of motor seizures was not included in seizure duration. Afterdischarge duration was the total duration of amygdala EEG spikes with an amplitude of at least twice the amplitude of the prestimulus recording and a frequency greater than 1/s. After each ADT determination, i.e. about 60 min after injection of phenytoin, blood was sampled by retroorbital puncture from each rat (after local anesthesia of the eye with a 2% solution of lidocaine). Blood was used for drug analysis in plasma by a gas chromatographic method previously used by us for determination of primidone and phenobarbital (Frey et al.,
177
1984), using 5-(p-methylphenyl-)-5-phenylhydantoin (MPPH) as internal standard. Four days later, ADT was redetermined in all rats 1 h after saline injection. From the next day on, constant current stimulations (500 mA, 1 ms, monophasic square-wave pulses, 50/s for 1 s) were delivered to the amygdala at intervals of 1 day until at least 10 consecutive fully kindled (stage 5; seizure severity classified according to Racine (1972)) seizures were elicited. In these fully kindled rats, ADT was repeatedly determined at intervals of at least 2 days until all rats exhibited a reproducible seizure threshold. Since almost all fully kindled rats exhibited generalized (stage 4–5) seizures in addition to focal (stage 1–3) seizures at the afterdischarge threshold current, it was not necessary to determine the generalized seizure threshold separately. In most animals, there was a tendency of ADTs to decrease during repeated determinations (see Section 3). After reproducible ADTs were obtained, phenytoin was administered at a dose of 75 mg/kg i.p. Current application at 1-min intervals was started 58 min after drug injection at a current that was two 20%-steps below the individual predrug control threshold. The current was elevated at 1-min intervals in steps of about 20% of the previous current until an afterdischarge of at least 3 s duration was recorded. Seizure severity, seizure duration and afterdischarge duration occurring at the seizure threshold current were recorded and used for comparison with respective readings at control threshold. In subsequent experiments, other doses of phenytoin were tested in the kindled rats. Control thresholds (determined 1 h after i.p. injection of saline) were determined 2–3 days before and after each phenytoin injection. At least 1 week was allowed between 2 drug injections to avoid drug accumulation or development of tolerance due to too frequent drug injections. Comparison of predrug and postdrug control ADTs of each drug trial indicated that single doses of phenytoin exerted no long-term effects on ADT. Following seizure threshold determination, i.e. about 60 min after injection of phenytoin, blood was withdrawn by retroorbital puncture and analyzed by gas chromatography as described above. Experiments (less than 10%) in
178
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
which drug analysis indicated insufficient drug absorption and/or erroneous drug injection were repeated. With respect to the repeated experiments with phenytoin in female rats, it should be noted that we recently showed that effects of phenytoin on ADT are not affected by estrous cycle (Rundfeldt et al., 1990). In a group of six fully kindled male rats, the anticonvulsant efficacy of phenytoin was compared after i.p. and i.v. administration, using a dose of 50 mg/kg and a pretreatment time of 1 h. Time interval between the two phenytoin injections in the same rat was 1 week. Data after injection of phenytoin were compared with vehicle control data recorded 2 – 3 days before drug injection, using the same route of administration. As in all other experiments with phenytoin, blood was withdrawn shortly after ADT determinations for drug analysis in plasma. In all experiments with phenytoin, animals were observed in an open field for adverse effects shortly before the ADT determination (for details see Ho¨nack and Lo¨scher, 1995). Furthermore, all rats were subjected to the rotarod test shortly before ADT determination (see Ho¨nack and Lo¨scher, 1995); since all rats passed the test in all drug trials, these data are not illustrated here.
2.4. Drugs Phenytoin (kindly provided by Desitin Arzneimittel GmbH; Hamburg, Germany) was freshly dissolved in distilled water by means of dilute NaOH (final pH of the solution was about 11). Except some experiments with i.v. injection, all injections were made i.p. with an injection volume of 3 ml/kg. For control recordings, rats received saline with the same volume and time interval to amygdala stimulation as in the drug experiment.
2.5. Statistics Significance of differences between individual control recordings and recordings after drug treatment was calculated by the Wilcoxon signed rank test for paired replicates. Significance of differences in recordings between different groups was
calculated by the U-test of Mann and Whitney. All tests were performed two-sided and P B 0.05 was considered significant.
3. Results
3.1. Anticon6ulsant effect of phenytoin on prekindling ADT When the ADT was determined twice at an interval of 4 days before kindling, there was a slight but statistically significant decrease in ADT at the second determination in saline-injected rats of both genders (group 2 in Fig. 1). Injection of phenytoin, 75 mg/kg, in group 1 markedly increased ADT in male rats both compared to the first ADT determination in group 2 and the subsequent ADT determination in group 1 (Fig. 1). A less marked ADT increase was seen after phenytoin in female rats, although plasma concentrations at time of ADT determination in female rats were higher than in male rats (see below). Recording of seizure parameters (severity and duration of seizures) at ADT currents showed focal seizures of short duration after saline injections, whereas an increase in seizure severity and duration was observed at the increased ADT currents after phenytoin injection, which has recently been reported by our group (Ebert et al., 1997) and is therefore not illustrated here.
3.2. Anticon6ulsant effect of phenytoin on postkindling ADT When the same dose of phenytoin, i.e. 75 mg/ kg, was administered after kindling development, the anticonvulsant effect observed at this dose before kindling was lost in both genders (Figs. 2 and 3). In kindled females, phenytoin even decreased ADT, suggesting a proconvulsant effect (Fig. 3). The loss of anticonvulsant efficacy of phenytoin, 75 mg/kg, after kindling was not due to a change in pharmacokinetics, because plasma levels of phenytoin determined at this dose after kindling were similar or even higher than plasma levels determined before kindling (Fig. 4). Adverse effects (slight ataxia) seen after phenytoin at
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
179
Fig. 1. Effect of phenytoin on prekindling ADT in groups of male and female Sprague – Dawley rats. In group 1 (nine males, 11 females), phenytoin, 75 mg/kg, was injected i.p. and ADT was determined after 1 h. Four days later, ADT was redetermined after saline injection. In group 2 (eight males, nine females), saline was injected instead of phenytoin at time of first ADT determination, followed by a second ADT determination 4 days later. Data are shown as means + S.E. Within group significance of ADT difference between first and second determination is indicated by asterisk (PB 0.05, at least), while between group difference in the first ADT determination is indicated by circle (P B 0.01).
time of ADT determination were similar before and after kindling (not illustrated). In order to examine whether the apparent loss of anticonvulsant efficacy was due to a shift in the dose –response curve of phenytoin, we tested phenytoin at lower doses in fully kindled rats. At 50 mg/kg, ADT was markedly increased in both genders (Figs. 2 and 3), although plasma levels were lower compared to 75 mg/kg (Fig. 4). Further lowering the dose to 25 mg/kg led to a loss of anticonvulsant activity in males (Fig. 2), while the ADT was similarly increased in females as seen with 50 mg/kg (Fig. 3). Plasma concentrations in both genders were lower at 25 compared to 50 mg/kg (Fig. 4). However, because females in gen-
eral had higher plasma levels than males, administration of phenytoin, 25 mg/kg, in females resulted in similar phenytoin plasma concentrations than 50 mg/kg in males (Fig. 4), which is explained by the lower elimination rate of phenytoin in female compared to male rats (Jones and Wimbish, 1985). No adverse effects were seen after phenytoin, 25 or 50 mg/kg, in the two genders. In order to evaluate whether the repeated testing of phenytoin in kindled rats had changed their response to phenytoin, we again administered 75 mg/kg in both genders, showing that at this dose phenytoin was still ineffective in kindled Sprague–Dawley rats (Figs. 2 and 3). Further-
180
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
more, we repeated the experiments with phenytoin in other groups of fully kindled Sprague – Dawley rats, substantiating the significant anticonvulsant
Fig. 2. Effect of phenytoin on ADT before and after kindling in male Sprague – Dawley rats. Data before kindling have already been shown in Fig. 1 and are included for comparison with postkindling data. Determination of phenytoin’s postkindling effects on ADT was started about 3 weeks after completion of kindling, i.e. about 6 weeks after the prekindling phenytoin injection, the time intervals between pre- and postkindling drug applications being due to the kindling process and the repeated postkindling ADT determinations before the first postkindling drugs administration. Data are shown as means plus S.E. of nine rats. Interval between two phenytoin injections in the postkindling period was at least 1 week. Phenytoin was either injected at 75, 50, or 25 mg/kg i.p., and ADT was determined after 1 h. Each postkindling drug trial was preceded by a saline control trial, which was performed 2–3 days before each drug experiment. As described in the legend to Fig. 1, saline control trials before kindling were performed 4 days after drug trials, which explains the reversed order of the white and black bars for the pre- and postkindling data illustrated in the figure. Significant difference between each drug trial and its control trial is indicated by asterisk (P B 0.05, at least).
Fig. 3. Effect of phenytoin on ADT before and after kindling in female Sprague – Dawley rats. Data before kindling have already been shown in Fig. 1 and are included for comparison with postkindling data. Data are shown as means plus S.E. of 10 rats. Significant difference between each drug trials and its control trials is indicated by asterisk (PB 0.05, at least). For further details see legend to Fig. 2.
effects of 50 or 25 mg/kg but lack of anticonvulsant effect of 75 mg/kg (not illustrated). In both genders, control ADTs decreased during the period of drug testing in fully kindled rats (Figs. 2 and 3), indicating that repeated amygdala stimulations lowered ADT. However, this wellknown decrease of ADT in response to repeated amygdala stimulations obviously did not affect the anticonvulsant effect of phenytoin, as indicated by the two drug trials with phenytoin, 75 mg/kg, in the post-kindling period. Thus, although control ADT before the second drug trial with phenytoin, 75 mg/kg, had decreased by more than 50% compared to the control ADT recorded before the first post-kindling drug trial with this dose of phenytoin, no anticonvulsant effects were
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
181
Fig. 4. Plasma concentrations in male and female Sprague–Dawley rats after i.p. administration of phenytoin as determined shortly after recording of ADTs as shown in Figs. 2 and 3. Data are means +S.E. of nine (males) or 10 (females) rats. For further details see legends to Figs. 2 and 3.
seen in the two drug trials. Seizure parameters recorded at ADT currents in fully kindled rats were not significantly changed by phenytoin in most experiments (not illustrated).
thermore, plasma levels were not significantly different 1 h after i.p. or i.v. injection (Fig. 5).
4. Discussion
3.3. Anticon6ulsant effect of phenytoin on postkindling ADT after i.p. 6ersus i.6. administration Because it has been reported previously that i.v. administration of phenytoin is more effective than i.p. injection in male Sprague – Dawley rats (McNamara et al., 1989), we compared the two routes in fully kindled male rats, using a dose of 50 mg/kg. As shown in Fig. 5, 1 h after injection phenytoin increased ADT by almost the same extent with both routes of administration. Fur-
It has previously been shown that phenytoin increases the threshold for induction of focal afterdischarges in the amygdala of nonkindled rats of different strains, including Sprague–Dawley rats (Albright, 1983; Jurna, 1985; Ebert et al., 1997). This was confirmed in the present study in the latter strain, but phenytoin was clearly more effective in male than in female rats, although plasma levels in females were higher. Subsequent kindling of the animals led to a total loss of anticonvulsant efficacy of phenytoin when the
182
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
Fig. 5. Comparison of anticonvulsant efficacy and plasma levels of phenytoin after i.p. and i.v. administration in a group of six fully kindled male Sprague – Dawley rats. Data are shown as means +S.E. Phenytoin was injected at a dose of 50 mg/kg either i.p. or i.v. in the same six rats with an interval of 1 week between the two injections. ADT was determined 1 h after injection. Control trials with i.p. or i.v. injection of saline were performed 2–3 days before each drug injection. Significance of difference in ADT between control and drug trial is indicated by an asterisk (PB 0.05). The ADT increase after i.v. injection was not significantly different from the increase seen after i.p. injection. Furthermore, plasma levels determined immediately after ADT recording did not differ between the two routes.
same dose, 75 mg/kg i.p., was given both before and after kindling. This was unexpected, because previous experiments in fully kindled Wistar rats had shown that phenytoin dose-dependently increases ADT at doses of 12.5 – 75 mg/kg, the highest ADT increase seen after the highest dose (Rundfeldt et al., 1990). The present data on Sprague–Dawley rats provide evidence that epileptogenesis, as induced by kindling, can dramatically alter the anticonvulsant efficacy of an antiepileptic drug. One possible explanation for the loss of anticonvulsant activity of phenytoin, 75 mg/kg, after kindling of Sprague– Dawley rats would be a shift of the dose–response curve to the left, so that lower doses of phenytoin become effective. It is
known that the dose–response of anticonvulsant effects of phenytoin is limited, because at high doses phenytoin may produce proconvulsant rather than anticonvulsant effects in seizure models, including kindled rats (Racine et al., 1975; Wauquier et al., 1979; Callaghan and Schwark, 1980; Jones and Wimbish, 1985; Jurna, 1985; White et al., 1985; Lo¨scher et al., 1991). Such proconvulsant effects of high doses of phenytoin have also been reported in patients (Levy and Fenichel, 1965; Lerman, 1986; Oscorio et al., 1989). Thus, if kindling lowers the threshold to proconvulsant drug effects, which has been shown for several convulsants (Cain, 1986), this might explain the finding that phenytoin was no longer anticonvulsant when administered in a high dose
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
that was anticonvulsant before kindling. Indeed, in females phenytoin, 75 mg/kg, significantly decreased ADT after kindling. Reduction of the dose from 75 to 50 mg/kg or, in females, even 25 mg/kg, resulted in significant anticonvulsant effects in fully kindled rats. Thus, in contrast to non-kindled rats, in which dose-related increases in ADT has been seen up to doses of 150 mg/kg i.p. (Albright, 1983), fully kindled Sprague – Dawley rats exhibited a change in dose – response characteristics of phenytoin with a loss of anticonvulsant efficacy of high doses. In this respect, however, one should consider that, in contrast to humans, phenytoin at 75 mg/kg was not particularly high or toxic in rats but led to plasma concentrations within or nearby the ‘therapeutic range’ of phenytoin in epileptic patients (Hvidberg, 1985; Schmidt et al., 1986) because of the differences in pharmacokinetics between humans and rats (Jones and Wimbish, 1985). To clearly demonstrate a shift in phenytoin’s anticonvulsant dose – response curve after kindling, a formal dose – response comparison with different doses of phenytoin before and after kindling would be needed. However, such an experiment would be biased by the kindling effect of repeated ADT determinations needed for such experiments in the same group of rats. As shown in Fig. 1, two subsequent control ADT determinations already led to a significant reduction in ADT, i.e. an indication of a kindling effect. Furthermore, seizure severity and duration slightly increased at the second ADT determination (not illustrated), again indicating a kindling effect. Thus, in order to compare phenytoin’s effect before and after kindling in the same group of rats, we decided to test phenytoin at only one dose, 75 mg/kg, before kindling. As mentioned above, previous experiments of Albright (1983) in non-kindled rats have demonstrated dose-related increases in ADT after i.p. administration of phenytoin at doses of 12.5 – 150 mg/kg, so that we chose one dose within this dose range for the present experiments before kindling acquisition. It has previously been shown that kindling alters the sensitivity of rats to drug effects. For instance, the noncompetitive N-methyl-D-aspartate (NMDA) receptor antagonist memantine
183
produced seizures in kindled rats at doses that were anticonvulsant in non-kindled rats (Lo¨scher and Ho¨nack, 1990), demonstrating a lowering of the threshold to proconvulsant drug effects by kindling. The finding that NMDA antagonists such as MK-801 (dizocilpine) exert potent anticonvulsant (or antiepileptogenic) effects during kindling, but are almost ineffective in fully kindled rats (Lo¨scher and Schmidt, 1994) may also indicate that anticonvulsant effects of these drugs are changed by kindling. Further evidence of altered drug effects after acquisition of kindling comes from the observation that competitive NMDA antagonists induced adverse effects, e.g. stereotyped behaviors, in kindled rats at doses that were devoid of such effects in nonkindled rats (Lo¨scher and Ho¨nack, 1991). Similarly, some antiepileptic drugs produced more intense adverse effects in kindled than in nonkindled rats (Ho¨nack and Lo¨scher, 1995). In the present study with phenytoin, no difference in adverse effects was observed before and after kindling in the same rats, suggesting that the alterations in phenytoin’s anticonvulsant efficacy produced by kindling were not a reflection of a general shift in the brain’s susceptibility to phenytoin. In this respect it is interesting to note that we recently found that a subgroup of kindled Wistar rats in which phenytoin exerts no effects on kindled seizures (phenytoin nonresponders) still responds to anticonvulsant effects of phenytoin when generalized seizures are induced by transcorneal electroshock application, thus indicating that the lack of anticonvulsant efficacy of phenytoin in such rats is restricted to kindled focal seizures (Lo¨scher and Rundfeldt, 1991). It will be interesting to examine whether, in such kindled phenytoin nonresponders, ADTs determined in the amygdala are not responsive to phenytoin already before kindling or if kindling leads to the loss of anticonvulsant efficacy. Phenytoin is often considered as a ‘stabilizer’ of excitable membranes, because it prevents or depresses repetitive electrical activity (Jurna, 1985; Rogawski and Porter, 1990). However, in some preparations phenytoin has found to produce burst-like activity in a subpopulation of neurons, substantiating that, depending on the state of the
184
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
neurons, this drug can exert both anticonvulsant and proconvulsant actions (Jurna, 1985). During early postnatal development, phenytoin has been found to exert excitatory effects by blocking both pre- and postsynaptic inhibition. At later stages of development and in adulthood this excitatory effect is counteracted by the activation by phenytoin of inhibitory systems (Jurna, 1985). Numerous findings indicate that the effects of phenytoin on excitable membranes might be due to an action on the movement of sodium and calcium ions (Jurna, 1985; Rogawski and Porter, 1990; DeLorenzo, 1995), but more recent reports indicate that phenytoin exerts also effects on some receptor types and biochemical processes involved in GABAergic and glutamatergic neurotransmission (DeLorenzo, 1995; Lo¨scher, 1998), which might explain previous electrophysiological findings that phenytoin potentiates GABA-mediated inhibition and depresses excitatory transmission (Rogawski and Porter, 1990). Thus, it is only fair to state that the electrophysiological and biochemical mechanisms underlying the anticonvulsant and proconvulsant actions of phenytoin are presumably more complex than often proposed and are still not fully understood (DeLorenzo, 1995; Lo¨scher, 1998). In view of the complex electrophysiological and biochemical alterations occurring during kindling (McNamara, 1988; Bolwig and Trimble, 1989; Sato et al., 1990; Lo¨scher, 1993; McNamara, 1994), it is conceivable that the excitatory effects that are produced by phenytoin in subsets of normal neurons (cf., Jurna, 1985) are enhanced by kindling, thus leading to a proconvulsant activity at phenytoin concentrations which normally are anticonvulsant. As indicated by the present data, kindling seems to be an appropriate epileptic model to elucidate this effect. To our knowledge, only few previous studies compared the anticonvulsant efficacy of antiepileptic drugs before and after kindling (Albertson et al., 1981). In a study in male Sprague – Dawley rats, the effects of various doses of phenobarbital and diazepam were determined on prekindled and kindled amygdaloid seizures in the same rats. However, in contrast to the present study, only one fixed, suprathreshold current (400 mA) was used for drug testing. Diazepam, 0.5 – 4
mg/kg i.p., was ineffective against the prekindled focal seizures, but demonstrated significant effects on fully kindled seizures. Phenobarbital, 7.5–60 mg/kg i.p., reduced afterdischarge duration of both prekindled and kindled seizures, but was more effective in reducing seizure severity in the postkindled state (Albertson et al., 1981), which is consistent with another study of Albertson et al. (1978). The authors suggested that the increase in anticonvulsant effectiveness found with diazepam and phenobarbital against the kindled seizure may reflect an increased sensitivity of the neuronal pathways which have undergone change during the kindling process. Alternatively, the increased efficacy in fully kindled rats may simply be due to the fact that secondarily generalized (stage 4/5) seizures are more sensitive to these drugs than focal seizures (Lo¨scher et al., 1986; Lo¨scher and Schmidt, 1988). Interestingly, following chemical kindling with pentylenetetrazol, a decrease in anticonvulsant efficacy of diazepam and phenobarbital was found (Albertson et al., 1978). McNamara’s group (McNamara et al., 1989) has previously proposed that the inconsistent results obtained by many groups with phenytoin in kindled rats might be due to the route of administration used by most groups, which is the i.p. route, because of low and variable plasma levels achieved by this route. In order to exclude that the differences in phenytoin’s efficacy before and after kindling were at least in part due to such variation, plasma levels were determined in all of the present experiments, showing that the i.p. route is not involved in this finding. Indeed, less than 10% of all i.p. injections resulted in erroneously low plasma levels, e.g. by crystal formation in the peritoneal cavity leading to incomplete absorption (McNamara et al., 1989). Furthermore, we directly compared anticonvulsant efficacy and plasma levels of phenytoin after i.p. and i.v. administration in male Sprague–Dawley rats and could not confirm the findings of McNamara et al. (1989). Instead, both routes gave the same anticonvulsant effect and similar plasma levels of phenytoin. In conclusion, the present data demonstrate a marked change in phenytoin’s anticonvulsant efficacy in response to kindling, resulting in a trun-
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186
cated dose–response with total loss of efficacy of a high dose, but significant anticonvulsant effects at lower doses. The most likely explanation for the loss of anticonvulsant activity of phenytoin, 75 mg/kg, is an increased sensitivity of kindled rats to proconvulsant drug action, thereby shifting the loss of anticonvulsant effect, which normally only occurs at high, toxic doses, to lower doses of phenytoin. If the same phenomenon of epileptogenesis-induced shifts in drug responses occur in human temporal lobe epilepsy, it might at least in part explain why anticonvulsant activity of a given dose of phenytoin or other antiepileptic drugs with proconvulsant activity at high doses may be lost during development of chronic epilepsy. Neurophysiological experiments are under way to examine which cellular mechanisms underlie the change in phenytoin’s anticonvulsant activity induced by kindling. Furthermore, we plan to undertake similar experiments in other rats strains, including the Wistar strain. Acknowledgements We thank Dr C. Rundfeldt for the suggestion to test phenytoin before kindling and M. Weissing, M. Gramer and C. Bartling for technical assistance. The study was supported by a grant (Lo 274/5-2) from the Deutsche Forschungsgemeinschaft (Bonn, Germany).
References Albertson, T., Peterson, S., Stark, L., 1978. Effects of phenobarbital and SC-13504 on partially kindled seizures in rats. Exp. Neurol. 61, 270–280. Albertson, T.E., Peterson, S.L., Stark, L.G., 1981. The anticonvulsant effects of diazepam and phenobarbital in prekindled and kindled seizures in rats. Neuropharmacology 20, 597 – 603. Albright, P.S., 1983. Effects of carbamazepine, clonazepam, and phenytoin on seizure threshold in amygdala and cortex. Exp. Neurol. 79, 11–17. Bolwig, T.G., Trimble, M.R., 1989. The Clinical Relevance of Kindling. John Wiley & Sons, Chichester. Cain, D.P., 1986. The transfer phenomenon in kindling. In: Wada, J.A. (Ed.), Kindling 3. Raven Press, New York, pp. 231 – 248.
185
Callaghan, D.A., Schwark, W.S., 1980. Pharmacological modification of amygdaloid-kindled seizures. Neuropharmacology 19, 1131 – 1136. DeLorenzo, R.J., 1995. Phenytoin. Mechanisms of action. In: Levy, R.H., Mattson, R.H., Meldrum, B.S. (Eds.), Antiepileptic Drugs, 4th ed. Raven Press, New York, pp. 271 – 282. Ebert, U., Cramer, S., Lo¨scher, W., 1997. Phenytoin’s effect on the spread of seizure activity in the amygdala kindling model. Naunyn-Schmiedeberg’s Arch. Pharmacol. 356, 341 – 347. Freeman, F.G., Jarvis, M.F., 1981. The effect of interstimulation interval on the assessment and stability of kindled seizure threshold. Brain Res. Bull. 7, 629 – 633. Frey, H.-H., Lo¨scher, W., Reiche, R., Schultz, D., 1984. Anticonvulsant effect of primidone in the gerbil. Time course and significance of the active metabolites. Pharmacology 28, 329 – 335. Heinemann, U., Draguhn, A., Ficker, E., Stabel, J., Zhang, C.L., 1994. Strategies for the development of drugs for pharmacoresistant epilepsies. Epilepsia 35 (Suppl. 5), S10 – S21. Ho¨nack, D., Lo¨scher, W., 1995. Kindling increases the sensitivity of rats to adverse effects of certain antiepileptic drugs. Epilepsia 36, 763 – 771. Hvidberg, E.F., 1985. Monitoring antiepileptic drug levels. In: Frey, H.-H., Janz, D. (Eds.), Antiepileptic Drugs. Springer, Berlin, pp. 725 – 764. Johannessen, S.I., Gram, L., Sillanpa¨a¨, M., Tomson, T., 1995. Intractable Epilepsy. Wrightson Biomedical Publishing Ltd., Petersfield. Jones, G.L., Wimbish, G.H., 1985. Hydantoins. In: Frey, H.-H., Janz, D. (Eds.), Antiepileptic Drugs. Springer, Berlin, pp. 351 – 420. Jurna, I., 1985. Electrophysiological effects of antiepileptic drugs. In: Frey, H.-H., Janz, D. (Eds.), Antiepileptic Drugs. Springer, Berlin, pp. 611 – 658. Lerman, P., 1986. Seizures induced or aggravated by anticonvulsants. Epilepsia 27, 706 – 710. Levy, L.L., Fenichel, G.M., 1965. Diphenylhydantoin activated seizures. Neurology 15, 716 – 722. Lo¨scher, W., 1993. Basic aspects of epilepsy. Curr. Opin. Neurol. Neurosurg. 6, 223 – 232. Lo¨scher, W., 1997. Animal models of intractable epilepsy. Prog. Neurobiol. 53, 239 – 258. Lo¨scher, W., 1998. New visions in the pharmacology of anticonvulsion. Eur. J. Pharmacol. 342, 1 – 13. Lo¨scher, W., Schmidt, D., 1988. Which animal models should be used in the search for new antiepileptic drugs? A proposal based on experimental and clinical considerations. Epilepsy Res. 2, 145 – 181. Lo¨scher, W., Ho¨nack, D., 1990. High doses of memantine (1-amino-3,5-dimethyladamantane) induce seizures in kindled but not in non-kindled rats. Naunyn-Schmiedeberg’s Arch. Pharmacol. 341, 476 – 481. Lo¨scher, W., Ho¨nack, D., 1991. Responses to NMDA receptor antagonists altered by epileptogenesis. Trends Pharmacol. Sci. 12, 52.
186
W. Lo¨scher et al. / Epilepsy Research 31 (1998) 175–186 Racine, R.J., 1972. Modification of seizure activity by electrical stimulation: II. Motor seizure. Electroencephalogr. Clin. Neurophysiol. 32, 281 – 294. Racine, R.J., Livingston, K., Joaquin, A., 1975. Effects of procaine hydrochloride, diazepam, and diphenylhydantoin on seizure development in cortical and subcortical structures in rats. Electroencephalogr. Clin. Neurophysiol. 38, 355 – 365. Rogawski, M.A., Porter, R.J., 1990. Antiepileptic drugs: pharmacological mechanisms and clinical efficacy with consideration of promising developmental stage compounds. Pharmacol. Rev. 42, 223 – 286. Rundfeldt, C., Ho¨nack, D., Lo¨scher, W., 1990. Phenytoin potently increases the threshold for focal seizures in amygdala-kindled rats. Neuropharmacology 29, 845 – 851. Sato, M., Racine, R.J., McIntyre, D.C., 1990. Kindling: basic mechanisms and clinical validity. Electroencephalogr. Clin. Neurophysiol. 76, 459 – 472. Schmidt, D., Morselli, P., 1986. Intractable Epilepsy: Experimental and Clinical Aspects. Raven Press, New York, NY. Schmidt, D., Einicke, I., Haenel, F., 1986. The influence of seizure type on the efficacy of plasma concentrations of phenytoin, phenobarbital, and carbamazepine. Arch. Neurol. 43, 263 – 265. Wauquier, A., Ashton, D., Melis, W., 1979. Behavioural analysis of amygdaloid kindling in Beagle dogs and the effects of clonazepam, diazepam, phenobarbital, diphenylhydantoin, and flunarizine on seizure manifestations. Exp. Neurol. 64, 579 – 586. White, H.S., Chen, C.F., Kemp, J.W., Woodbury, D.M., 1985. Effects of acute and chronic phenytoin on the electrolyte content and the activities of Na + , K + , Ca2 + , Mg2 + , and HCO3 − -ATPases and carbonic anhydrase of neonatal and adult cerebral cortex. Epilepsia 26, 43 – 57.
Lo¨scher, W., Rundfeldt, C., 1991. Kindling as a model of drug-resistant partial epilepsy: selection of phenytoin-resistant and nonresistant rats. J. Pharmacol. Exp. Ther. 258, 483 – 489. Lo¨scher, W., Schmidt, D., 1994. Strategies in antiepileptic drug development: is rational drug design superior to random screening and structural variation? Epilepsy Res. 17, 95 – 134. Lo¨scher, W., Ja¨ckel, R., Czuczwar, S.J., 1986. Is amygdala kindling in rats a model for drug-resistant partial epilepsy? Exp. Neurol. 93, 211 –226. Lo¨scher, W., Ho¨nack, D., Fassbender, C.P., Nolting, B., 1991. The role of technical, biological and pharmacological factors in the laboratory evaluation of anticonvulsant drugs. III. Pentylenetetrazol seizure models. Epilepsy Res. 8, 171– 189. Lo¨scher, W., Rundfeldt, C., Ho¨nack, D., 1993. Pharmacological characterization of phenytoin-resistant amygdala-kindled rats, a new model of drug-resistant partial epilepsy. Epilepsy Res. 15, 207–219. McNamara, J.O., 1988. Pursuit of the mechanisms of kindling. Trends Neurosci. 11, 33–36. McNamara, J.O., 1994. Cellular and molecular basis of epilepsy. J. Neurosci. 14, 3413–3425. McNamara, J.O., Rigsbee, L.C., Butler, L.S., Shin, C., 1989. Intravenous phenytoin is an effective anticonvulsant in the kindling model. Ann. Neurol. 26, 675–678. Oscorio, I., Burnstine, T.H., Remler, B., Manon-Espaillat, R., Reed, R.C., 1989. Phenytoin-induced seizures: a paradoxical effect at toxic concentrations in epileptic patients. Epilepsia 30, 230 – 234. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates. Academic Press, Sydney.
.
.