Epilepsy Research (2014) 108, 675—683
journal homepage: www.elsevier.com/locate/epilepsyres
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model K. Leclercq, A. Matagne, R.M. Kaminski ∗ Neuroscience TA, UCB Pharma, Braine-l’Alleud, Belgium Received 6 June 2013; received in revised form 10 February 2014; accepted 23 February 2014 Available online 5 March 2014
KEYWORDS Kindling; Seizures; Epilepsy; Levetiracetam; 6 Hz; Pharmacoresistance
∗
Summary Corneal kindling is a useful alternative to electrically induced amygdala or hippocampal kindling, which requires advanced surgical and EEG techniques that may not be easily available in many laboratories. Therefore the first aim of this study was to evaluate whether repeated 6 Hz corneal stimulation in mice would lead to an increased and persistent seizure response as described for higher frequency (50/60 Hz) corneal kindling. Male NMRI mice stimulated twice daily (except weekends) for 3 s with 6 Hz electrical current at 44 mA displayed robust kindling development, i.e. a progressive increase in seizure severity. The majority of the animals (about 90%) developed a fully kindled state, defined as at least 10 consecutive stage 3—5 seizures within 5 weeks of corneal stimulation. Afterwards, the fully kindled state was maintained for at least 8 weeks with only two days of stimulations per week. Next, the protective efficacy of four mechanistically different antiepileptic drugs (AEDs; clonazepam, valproate, carbamazepine and levetiracetam) was assessed and compared between 6 Hz and 50 Hz fully kindled mice. All tested AEDs showed a relatively lower potency in the 6 Hz kindling model and a limited efficacy against partial seizures was observed with carbamazepine and levetiracetam. We can conclude that 6 Hz kindling may be more advantageous than the previously described 50/60 Hz corneal kindling models due to its robustness and persistence of the fully kindled state. Furthermore, the observed low potency and limited efficacy of AEDs in 6 Hz fully kindled mice suggest that this model could be a useful tool in the discovery of novel AEDs targeting treatment resistant epilepsy. © 2014 Elsevier B.V. All rights reserved.
Abbreviations: Hz, hertz; EEG, electroencephalogram; mA, milliampere; ms, millisecond; i.p., intraperitoneal. Corresponding author at: UCB Pharma, Chemin du Foriest, R9, B-1420 Braine-l’Alleud, Belgium. Tel.: +32 2 386 6499; fax: +32 2 386 3339. E-mail addresses: [email protected], rafal [email protected] (R.M. Kaminski).
http://dx.doi.org/10.1016/j.eplepsyres.2014.02.013 0920-1211/© 2014 Elsevier B.V. All rights reserved.
676
Introduction Despite the introduction of several novel antiepileptic drugs (AEDs) there is still a sizable population of patients suffering from treatment resistant seizures (Potschka and Brodie, 2012). This creates a strong need for the development of highly differentiated AEDs, which would specifically target this patient population. Innovative approaches for target identification and preclinical testing of candidate molecules are required in order to achieve this goal (Löscher and Schmidt, 2011). In line with this, the epilepsy research community came forth with a consensus proposal and guidance concerning preclinical AEDs discovery. This guidance makes a particular emphasis on the need for implementation of novel models of treatment resistant seizures (Galanopoulou et al., 2012). Traditionally, preclinical AEDs discovery involved relatively simple models, where seizures were acutely induced in naïve animals, which did not reflect underlying pathophysiology of the disease. Therefore chronic epilepsy models, such as seizure kindling, have been increasingly used in drug testing (Löscher, 2011). In kindling, an initially nonconvulsive stimulus applied repeatedly produces a clear progressive change in seizure response leading to so-called fully kindled state, characterized by stable seizure response to each stimulation (Goddard et al., 1969). The kindling process is considered as a relevant model of brain plasticity that may lead to epileptogenesis and progression of epilepsy (Sutula, 2004; Morimoto et al., 2004). There are, in fact, several ways to induce seizure kindling; electrical, chemical or even acoustic stimuli could be used for this purpose (Fisher, 1989; Kupferberg, 2001). Among those paradigms the best characterized is amygdala kindling, where repeated electrical stimulation is applied through an electrode implanted in amygdala (Goddard, 1983; Löscher and Schmidt, 1988). The amygdala kindling model has been widely used for AEDs testing; however it requires invasive stereotactic surgery and EEG equipment, which may not be easily accessible in many labs. An alternative that does not require electrode implantation could be the corneal kindling model in mice (Matagne and Klitgaard, 1998; Potschka and Löscher, 1999). In this model, a low intensity current (2—3 mA) 50 Hz electrical stimulation is applied twice daily for 3 s via corneal electrodes in contrast to a higher current (25—50 mA) single stimulation typically used in the maximal electroshock model (MES). A progressive seizure response develops over a few weeks of stimulations and a large proportion of animals reach fully kindled state and can be used for AEDs testing. The 50 Hz corneal kindling is indeed a useful alternative, but it does not seem to reproduce some of the treatment resistance aspects that had been reported in the amygdala kindling model (Löscher et al., 1986; Matagne and Klitgaard, 1998; Potschka and Löscher, 1999; Rowley and White, 2010). The 6 Hz model of partial seizures, originally developed in the 1950s (Toman, 1951; Brown, 1953; Swinyard, 1972), has been re-introduced in 2001 and proposed as a model of treatment resistant epilepsy (Barton et al., 2001). Since that time the 6 Hz model has been increasingly used for testing of new AEDs in naïve animals (Bialer and White, 2010) and has become a standard screening model in the NINDS
K. Leclercq et al. anticonvulsant screening program (Löscher, 2011). The acute 6 Hz electrical stimulation at 32—44 mA current intensity lasting 3 s produces a partial seizure, stages 1—2 according to Racine’s scale (Racine, 1972) and secondarily generalized seizures (stages 3—5) are never observed (Barton et al., 2001). The first aim of this study was to evaluate whether repeated 6 Hz corneal stimulation in mice would also lead to increased seizure responses and a persistent fully kindled state as described for repeated 50 Hz corneal stimulation (Matagne and Klitgaard, 1998; Potschka and Löscher, 1999). Having confirmed robust and reproducible 6 Hz kindling development, we next compared the effect of four mechanistically diverse AEDs on fully kindled seizures in both the 50 Hz and 6 Hz kindling paradigms. In the course of these experiments we have demonstrated that the efficacy and potency of AEDs in the 6 Hz kindling model is lower than in the 50 Hz kindling paradigm, which is akin to their limited efficacy in the amygdala kindling model. Therefore we propose that 6 Hz corneal kindling may be a novel, convenient model for treatment resistant seizures.
Materials and methods Animals Male NMRI mice (Charles River, France) weighing 19—29 g were used in all experiments. They were maintained on a 12/12-h light/dark cycle with lights on at 06:00 h and had free access to food and drinking water. The temperature in the husbandry was maintained at 20—21 ◦ C and humidity about 40%. For each experiment, the mice were housed in groups of 10 per cage and habituated for at least one hour in the experimental room. All procedures were carried out according to the Helsinki declaration and conducted according to the guidelines of the European Community Council directive 86/609/EEC. A local Ethics Committee approved all performed experiments.
Corneal kindling 6 Hz stimulations The mice (N = 140) were stimulated through corneal electrodes connected to a stimulator (ECT Unit 57800, UgoBasile, Comerio, Italy) using a current intensity of 44 mA, 0.2 ms monopolar pulses at 6 Hz and 3 s duration. A drop of Unicaine (0.4% oxybuprocainum in saline) was placed on the eyes before stimulation to induce local anesthesia and ensure good conductivity. Stimulations were performed twice daily (Monday—Friday), with a 4 h minimum interval between each stimulation, for 3 consecutive weeks. After each stimulation, the animal was gently placed in a plastic bowl (diameter = 19 cm) and observed for 30 s. At the beginning of the kindling process, the mice displayed a range of behaviors including stereotypy, Straub tail, stunned posture and sometimes mild myoclonus. After few days of stimulations approximately 50—60% of the mice began to exhibit secondarily generalized convulsions (stages 3—5) according to Racine’s scale (Racine, 1972). The mice were considered fully kindled when they displayed at least 10 consecutive,
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model
677
Figure 1 Proportion of generalized seizures in mice developing fully kindled state (panel A) and mean seizure severity score with S.D. (panel B) as a function of kindling stimulation days. Male NMRI mice (N = 140, N = 120 in 6 Hz and 50 Hz kindling, respectively) were used. The mice were stimulated through corneal electrodes using current intensity of 44 mA, 0.2 ms monopolar pulses at 6 Hz and 3 s duration. Stimulations were performed twice daily (Monday—Friday), with a 4 h minimum interval between each stimulation, for 3 consecutive weeks. In the 50 Hz kindling mice were stimulated daily (including weekends) using 3 mA current intensity, at 50 Hz pulse frequency for 3 s duration (Matagne and Klitgaard, 1998). Seizures were scored according to Racine’s scale (Racine, 1972). The mice were considered as fully kindled when they displayed at least 10 consecutive, secondarily generalized convulsions (stages 3—5). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
secondarily generalized convulsions. At the end of the third week of stimulations, 60—70% of the mice reached the fully kindled state, which was maintained by twice daily stimulation (Monday, Thursday) every week. The kindling procedure was pursued for 1 or 2 more weeks with remaining nonkindled animals before including them for maintenance. Finally, about 90% of mice reached a fully kindled state within 5 weeks of chronic stimulations. 50 Hz stimulations The 50 Hz stimulations were conducted as described by Matagne and Klitgaard (1998). Briefly, male NMRI mice (N = 120) were stimulated twice daily through corneal electrodes connected to a stimulator (WITT Industrie Elektronik, Berlin, Germany) using 3 mA current intensity, at 50 Hz pulse frequency for 3 s duration. A drop of Unicaine (0.4% oxybuprocainum in saline) was placed on the eyes
Table 1
before stimulation to induce local anesthesia and ensure good conductivity. Stimulations were performed twice daily (including weekends), with a 4 h minimum interval between each stimulation, for 4 consecutive weeks. After each stimulation, the animal was gently placed in a plastic bowl (diameter = 19 cm) and observed for 30 s. At the beginning of the kindling process, the mice displayed jaw movements and facial clonus, head nodding (stages 1—2) according to Racine’s scale (Racine, 1972). After few days of stimulations approximately 20—30% of the mice began to exhibit secondarily generalized convulsions (stages 3—5) according to Racine’s scale (Racine, 1972). The mice were considered fully kindled when they displayed at least 10 consecutive, secondarily generalized convulsions. At the end of the fourth week of stimulations, 50—60% of the mice reached the fully kindled state, which was maintained by twice daily stimulation (Monday, Thursday) every week.
The effects of antiepileptic drugs (AEDs) against partial and generalized seizures in 50 Hz or 6 Hz fully kindled mice.
Drugs
Clonazepam Valproate Carbamazepine Levetiracetam
50 Hz kindling
6 Hz kindling
Generalized
Partial
Generalized
Partial
0.032 (0.030—0.034) 66.1 (65.6—66.5) 5.4 (2.9—10.2) 6.5 (4.8—8.7)
0.048 (0.015—0.15)# 85.0 (64.1—112.8)# MAD 7.6 10.7 (5.8—19.9)
0.054 (0.032—0.089) 92.4 (82.7—103.2)* 8.2 (6.0—11.4) 18.7 (10.4—33.6)**
0.195 (0.094—0.40)*,## 168.5 (89.4—317.4)* 23.6 (10.0—56.0)## 108.0 (35.7—326.1)**,##
The ED50 values, i.e. doses protecting 50% of the animals against partial seizures (stages 1—2) and secondarily generalized seizures (stages 3—5) together with their 95% confidence intervals were calculated and compared using non-linear regression analysis (GraphPad Prism 5, San Diego, CA). MAD, minimal active dose. * P < 0.05. ** P < 0.01 — comparison between 50 Hz and 6 Hz kindling models. # P < 0.05. ## P < 0.01 — comparison between generalized and partial seizures.
678
K. Leclercq et al.
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model
Drug testing Carbamazepine (Sigma), valproate (Sigma) and clonazepam (Hoffman-La Roche; Sequoia Res.) were suspended in 0.9% saline containing 0.1% Tween 80. Levetiracetam (UCB Pharma S.A., Belgium) was dissolved in 0.9% saline. All drugs were administered i.p. at the following pre-administration times (time to peak effect) consistent with their pharmacokinetic profiles in mice: carbamazepine (15 min), valproate (15 min), clonazepam (30 min) and levetiracetam (60 min) (Barton et al., 2001). The drugs were tested in separate groups of kindled animals (N = 9—11) allowing one week wash-out between each trial.
Data analysis and statistics Seizure severity scores were expressed as medians and comparison between pre- and post-drug treatment was performed for each drug and dose using a Wilcoxon signed rank test for paired replicates; comparison between seizure scores in 50 Hz and 6 Hz kindling models was performed for each drug using a Kruskal—Wallis test followed by Dunn’s comparison. The ED50 values (dose protecting 50% of the animals) for partial seizures (stages 1—2) and secondarily generalized seizures (stages 3—5) with 95% confidence intervals were calculated using a non-linear regression analysis (GraphPad Prism 5, San Diego, CA). The dose—response curves and the ED50 values obtained in both kindling paradigms were statistically compared with the same software.
Results 6 Hz kindling development Initially we explored different current intensities (ranging from 14 to 44 mA) with different durations of stimulation (3—5 s) applied once or twice daily for at least 12 consecutive days (data not shown). Based on these initial experiments we chose 44 mA, 3 s stimulation as optimal to achieve robust kindling development within three weeks in the majority (60—70%) of the animals. Fig. 1A illustrates the 6 Hz kindling acquisition (generalized seizures) in those mice which developed a fully kindled state within three weeks. The 50 Hz kindling acquisition data within the same time frame are shown for comparison (Fig. 1A). Furthermore, the 6 Hz kindling procedure was continued for additional 1 or 2 weeks with remaining non-kindled animals and about 90% of mice finally reached the fully kindled state (data not shown).
679
There was a clear progression in seizure severity score from a mean of 1.8 on the first day of 6 Hz stimulations to a mean of 4.3 after three weeks of daily stimulations (except on weekends) (Fig. 1B). In comparison, 50 Hz kindling stimulation resulted in an increase of seizure severity from the mean score of 0, on the first day, up to the mean score of 4 after three weeks of daily stimulations (Fig. 1B). The 6 Hz fully kindled state was maintained for at least 8 weeks, using twice daily maintenance stimulation for two days per week, as described in the methods section. During this time the mean seizure score remained stable at approx. 4.3—4.6 (data not shown).
Comparison of AEDs activity in the 6 Hz and 50 Hz corneal kindling models Even though all tested AEDs displayed protective effects against fully kindled seizures in both kindling models (Fig. 2A—H), the dose—response curves for protection against generalized (scores 3—5) and partial (scores 1—2) seizures revealed interesting differences between the two models (Fig. 2A—H). These also allowed calculation of ED50 values against both generalized and partial seizures as well as comparison between the models (Table 1). Clonazepam (Fig. 2A and B) and valproate (Fig. 2C and D) afforded nearly complete protection against both partial and generalized seizures in both kindling models. Carbamazepine also showed dose-dependent and nearly complete protection against generalized seizures in both models (Fig. 2E), but had only limited efficacy (max. 50—65% protection) against partial seizures (Fig. 2F). In comparison, levetiracetam displayed a different profile in the two kindling models (Fig. 2G and H). A significant difference in the slopes of dose—response curves was observed, slopes which were much steeper in the 50 Hz kindling compared to the 6 Hz kindling (Fig. 2G and H). Levetiracetam displayed only limited protection (75%) against partial seizures in the 6 Hz kindling model (Fig. 2H). In the 50 Hz kindling model, the potency of carbamazepine and levetiracetam against generalized or partial seizures was comparable. Clonazepam and valproate were significantly less potent against partial seizures (Table 1). In the 6 Hz kindling paradigm, we observed a striking decrease in potency of all tested AEDs against partial seizures when compared to generalized seizures (Table 1). Moreover, ED50s against partial seizures in the 6 Hz kindling model were significantly higher when compared to ED50s against partial seizures in the 50 Hz kindling model (Table 1). The median seizure score is often used in the amygdala kindling model to assess the efficacy of AEDs. Consistently, we compared the effects of all tested AEDs on the
Figure 2 Dose—response curves for clonazepam, valproate, carbamazepine and levetiracetam in 6 Hz or 50 Hz fully kindled mice. Left panel illustrates protective effects of antiepileptic drugs against generalized seizures (stages 3—5) according to Racine’s scale (Racine, 1972). Right panel shows protective effects of antiepileptic drugs against partial seizures (stages 1—2). Solid red lines illustrate sigmoidal fitting of the dose—response curves in the 6 Hz kindling model, while dotted blue lines pertain to the 50 Hz kindling model. Red square symbols show percent protection against 6 Hz seizures and blue square symbols illustrate percent protection against 50 Hz seizures. N = 9—11 mice per group. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
680
K. Leclercq et al.
Figure 3 Effects of clonazepam, valproate, carbamazepine and levetiracetam on median seizure severity score in the 50 Hz (left panel) and 6 Hz (right panel) fully kindled mice. Open bars represent median seizure score before drug treatment and solid bars indicate median seizure score post-drug treatment. N = 9—11 mice per group. *P < 0.05, **P < 0.01, ***P < 0.001 — Wilcoxon signed rank test for paired replicates.
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model median seizure score in both 50 Hz and 6 Hz kindling models (Fig. 3A—H). Clonazepam dose-dependently reduced the seizure severity in both kindling models (Fig. 3A and B). Valproate was also efficacious in reducing the median seizure score, but it produced ‘‘all or none’’ response in the 50 Hz kindling model from the dose of 93 mg/kg (Fig. 3C), while the effects in the 6 Hz kindling were dose-dependent (Fig. 3D). Carbamazepine displayed a very flat dose—response curve in the 6 Hz model (Fig. 3F) and also showed ‘‘all or none’’ response at the dose of 7.6 mg/kg in the 50 Hz kindling model (Fig. 3E). Levetiracetam dose-dependently reduced the median seizure scores in both 50 Hz and 6 Hz kindling models, but it was more potent in reducing seizure score in the 50 Hz kindling model (Fig. 3G and H).
Discussion The present study demonstrates for the first time the development of seizure kindling induced by repeated 6 Hz corneal stimulations; a fully kindled state was easily achieved and maintained for several weeks. The potency and efficacy of four mechanistically different AEDs was established in the 6 Hz kindling model and compared with their effects in previously described 50 Hz corneal kindling model (Matagne and Klitgaard, 1998) at optimal pre-administration time (Barton et al., 2001). The tested AEDs showed relatively lower potency and limited efficacy in protection against partial seizures induced with the 6 Hz kindling paradigm suggesting that this model could be a useful tool in the discovery of novel AEDs targeting treatment resistant epilepsy. Corneal kindling models have been described in the literature, but so far only 50 or 60 Hz stimulation frequencies were used (Sangdee et al., 1982; Wlaz et al., 1998; Matagne and Klitgaard, 1998; Potschka and Löscher, 1999). These models were proven to be useful alternatives to the more laborious amygdala kindling model, particularly for drug screening purposes. However, the corneal kindling models described thus far have several limitations. For example, variability in the proportion of mice developing kindling, lack of persistence of the fully kindled state and relatively high mortality rate (10—20%) were observed (Sangdee et al., 1982; Wlaz et al., 1998; Matagne and Klitgaard, 1998; Potschka and Löscher, 1999); the mortality rate could reach up to 40% of kindled mice during longer maintenance phase (Matagne and Klitgaard, 1998). In that context the 6 Hz kindling model described in the present study appears more advantageous, even though the rate of kindling acquisition appears comparable between the two paradigms and partial seizures (scores 1—2) are already present during the first 6 Hz kindling stimulation. The observed differences between the 50 Hz and 6 Hz kindling models may be due to activation of different brain structures following single stimulation with these frequencies as evidenced by differential expression of an immediate early gene (c-fos) (Barton et al., 2001). Expression of c-fos in brainstem areas was only observed after 50 Hz stimulation, which might explain higher lethality observed after 50 Hz kindling. Furthermore, there are obvious differences in seizure semiology between acute MES and 6 Hz models (Barton et al., 2001). Tonic hindlimb extension is the primary endpoint in the MES model and is never observed in the acute 6 Hz model, which is more akin to a
681
focal limbic seizure, despite similar current intensity used in both of these acute models. Consequently, tonic hindlimb extension is occasionally observed at the beginning of the 50 Hz kindling development (Matagne and Klitgaard, 1998; Potschka and Löscher, 1999), which could lead to higher lethality, while these seizures are never observed in the 6 Hz kindling (present study). The proportion of mice developing 6 Hz kindling is very consistent (80—90%) and the fully kindled state persists for at least eight weeks, even after stopping the stimulations during the weekends. In addition, no mortality is associated to low frequency chronic stimulations. Overall the behavioral seizure responses during the development and maintenance of 6 Hz kindling resemble the traditional amygdala kindling which is well described in the literature and claimed to be a suitable model of treatment resistance (Löscher et al., 1986; Löscher, 2002a). In the present study we used four mechanistically different AEDs, i.e. clonazepam, valproate, carbamazepine and levetiracetam (Rogawski and Löscher, 2004). Clonazepam acts as a positive allosteric modulator of GABAA receptor (Whiting, 2003), valproate is a multi-mechanistic drug (Löscher, 2002b), and carbamazepine inhibits voltage-gated sodium channels (Ambrosio et al., 2002). Finally, levetiracetam binds to the synaptic vesicle protein 2A (SV2A), which mediates its anticonvulsant activity (Lynch et al., 2004; Kaminski et al., 2009). The differences in their mechanisms of action could have been responsible for their differential efficacy and potency in the 6 Hz and 50 Hz kindling models. Indeed, the potency of these AEDs was generally lower in the 6 Hz versus 50 Hz kindling model. This was particularly evident when potencies in protection against partial seizures were compared between the two models. Furthermore, the potency loss against partial seizures was much more pronounced in the 6 Hz kindling model. Levetiracetam did not afford complete protection against partial seizures in the 6 Hz kindling, while carbamazepine had limited efficacy against partial seizures in both models. It is worth noting that in the present study the ED50 of levetiracetam for protection against partial seizures was 108 mg/kg, whereas Barton et al. (2001) found that it was 1089 mg/kg. Even though the same stimulation parameters were used, these researchers have performed only acute 6 Hz stimulations in CF1 mice, while we determined it in fully kindled NMRI mice after repeated 6 Hz stimulations. More recently, Rowley and White (2010) have compared the potency of several AEDs in 60 Hz corneal kindling mode (CF1 mice) and found a good correlation between this model and a range of acute seizure models in mice such as the maximal electroshock (MES), the acute 6 Hz model, and the subcutaneous pentylentetrazole (scPTZ) test, as well as with the hippocampal kindled rat model. The present study is therefore the first one to compare the pharmacology of AEDs in two corneal kindling models in mice, i.e. 50 Hz and 6 Hz kindling. It is also the first study to suggest the usefulness of 6 Hz kindling in the context of therapy resistant epilepsy. In the rat amygdala kindling model, the ED50s of clonazepam and carbamazepine against generalized seizures were comparable to those calculated in the 6 Hz kindling model (Löscher and Schmidt, 1988). For carbamazepine and levetiracetam, the ED50s were higher in the rat amygdala kindling model (Löscher and Schmidt, 1988; Matagne et al., 2008), but complete seizure suppression was not achieved in
682 the 6 Hz kindling model despite administration of high doses of these AEDs. Median seizure severity score reduction is another important endpoint in the amygdala kindling model (Racine, 1972). Consequently, we assessed the ability of all four tested AEDs to reduce the median seizure severity score. We observed that the highest tested doses, going beyond clinically relevant therapeutic dose level, were able to reduce the median severity score to 0, but in general the AEDs displayed lower potency in the 6 Hz kindling model and there were still few animals not protected against partial seizures. In amygdala kindling models, depending on the stimulation protocol, it is often difficult to reduce the median seizure score to 0 and to protect all the animals against partial seizures (Löscher et al., 1986). Based on these observations the amygdala kindling model had been proposed as a model of treatment resistant seizures (Löscher et al., 1986). As described above, since persistence of partial seizures despite treatment with high doses of AEDs is present in the 6 Hz corneal kindling model, we suggest the latter can also be considered as a model of treatment resistant seizures. In the present study we have established a robust protocol for 6 Hz kindling in male NMRI mice, but as a next step this should also be confirmed in female mice to assess a potential effect of gender (Potschka and Löscher, 1999). Some modifications of the 6 Hz kindling protocol may be required if other strains of mice are used since there is a well described variability in seizure susceptibility among different mice strains (Frankel et al., 2001; Ferraro et al., 2002). In fact, differences in 6 Hz seizure thresholds have been observed by various groups depending on the mouse strains used (Barton et al., 2001; Hartman et al., 2008; Thomas et al., 2012), but head to head comparisons under the same experimental conditions have not been reported so far. Kindling models may also be useful to study epileptogenesis since they are characterized by a progressive increase in seizure incidence or severity. Consequently, antiepileptogenic therapies have been tested in kindling models (Dudek et al., 2008); for example, levetiracetam showed a robust delay of kindling development in both amygdala and corneal kindling models (Löscher et al., 1998; Matagne et al., 2008; Dudek et al., 2008). It would be interesting to confirm this effect of levetiracetam and other AEDs on 6 Hz kindling development. The usefulness of kindled animals to study epilepsy associated comorbidities (depression, anxiety, cognition, etc.) or treatment adverse effects has also been reported. In fact, amygdala kindled rats display a higher sensitivity to the side effects induced by anticonvulsants, in particular NMDA receptor antagonists (Hönack and Löscher, 1995), which also extends to corneally kindled mice (Klitgaard et al., 2002). Therefore the 6 Hz kindling model may be also useful for the assessment of side effects profiles and determination of therapeutic index as demonstrated in the 50 Hz corneal kindling model (Klitgaard et al., 2002). The present study reports the development and pharmacological validation of the 6 Hz corneal kindling model. The obtained data indicate the robustness of this model and seizure resistance to treatment with several AEDs. Finally, the 6 Hz corneal kindling model may also be used in future
K. Leclercq et al. studies aiming at the development of antiepileptogenic therapies.
Acknowledgments We thank Colette Chaussée, Fabienne Coddens and MarieChristine Tordeur for helpful technical assistance and Prof. Ilse Smolders (Center for Neurosciences, C4N, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Belgium) for critical review of the manuscript.
References Ambrosio, A.F., Soares-da-Silva, P., Carvalho, C.M., Carvalho, A.P., 2002. Mechanisms of action of carbamazepine and its derivatives, oxcarbazepine, BIA 2-093, and BIA 2-024. Neurochem. Res. 27, 121—130. Barton, M.E., Klein, B.D., Wolf, H.H., White, H.S., 2001. Pharmacological characterization of the 6 Hz psychomotor seizure model of partial epilepsy. Epilepsy Res. 47, 217—227. Bialer, M., White, H.S., 2010. Key factors in the discovery and development of new antiepileptic drugs. Nat. Rev. Drug Discov. 9, 68—82. Brown, W.C., 1953. Properties and alterations of electricallyinduced seizures in mice. Epilepsia 2, 127—137. Dudek, F.E., Bertram, E.H., Staley, K.J., 2008. Antiepileptogenesis therapy with levetiracetam: data from kindling versus status epilepticus models. Epilepsy Curr. 8, 28—30. Ferraro, T.N., Golden, G.T., Smith, G.G., DeMuth, D., Buono, R.J., Berrettini, W.H., 2002. Mouse strain variation in maximal electroshock seizure threshold. Brain Res. 936, 82—86. Fisher, R.S., 1989. Animal models of the epilepsies. Brain Res. Rev. 14, 245—278. Frankel, W.N., Taylor, L., Beyer, B., Tempel, B.L., White, H.S., 2001. Electroconvulsive thresholds of inbred mouse strains. Genomics 74, 306—312. Galanopoulou, A.S., Buckmaster, P.S., Staley, K.J., Moshé, S.L., Perucca, E., Engel Jr., J., Löscher, W., Noebels, J.L., Pitkänen, A., Stables, J., White, H.S., O‘Brien, T.J., Simonato, M., 2012. Identification of new epilepsy treatments: issues in preclinical methodology. Epilepsia 53, 571—582. Goddard, G.V., McIntyre, D.C., Leech, C.K., 1969. A permanent change in brain function resulting from daily electrical stimulation. Exp. Neurol. 25, 295—330. Goddard, G.V., 1983. The kindling model of epilepsy. Trends Neurosci. 6, 275—279. Hartman, A.L., Lyle, M., Rogawski, M.A., Gasior, M., 2008. Efficacy of the ketogenic diet in the 6-Hz seizure test. Epilepsia 49, 334—339. Hönack, D., Löscher, W., 1995. Kindling increases the sensitivity of rats to adverse effects of certain antiepileptic drugs. Epilepsia 36, 763—771. Kaminski, R.M., Gillard, M., Leclercq, K., Hanon, E., Lorent, G., Dassesse, D., Matagne, A., Klitgaard, H., 2009. Proepileptic phenotype of SV2A-deficient mice is associated with reduced anticonvulsant efficacy of levetiracetam. Epilepsia 50, 1729—1740. Klitgaard, H., Matagne, A., Lamberty, Y., 2002. Use of epileptic animals for adverse effect testing. Epilepsy Res. 50, 55—65. Kupferberg, H., 2001. Animal models used in the screening of antiepileptic drugs. Epilepsia 42, S7—S12. Löscher, W., 2002a. Animal models of epilepsy for the development of antiepileptogenic and disease-modifying drugs. A comparison of the pharmacology of kindling and post-status epilepticus models of temporal lobe epilepsy. Epilepsy Res. 50, 105—123.
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model Löscher, W., Hönack, D., Rundfeldt, C., 1998. Antiepileptogenic effects of the novel anticonvulsant levetiracetam (ucb L059) in the kindling model of temporal lobe epilepsy. J. Pharmacol. Exp. Ther. 284, 474—479. Löscher, W., Jäckel, R., Czuczwar, S.J., 1986. Is amygdala kindling in rats a model for drug-resistant partial epilepsy? Exp. Neurol. 93, 211—226. Löscher, 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. Löscher, W., Schmidt, D., 2011. Modern antiepileptic drug development has failed to deliver: ways out of the current dilemma. Epilepsia 52, 657—678. Löscher, W., 2002b. Basic pharmacology of valproate: a review after 35 years of clinical use for the treatment of epilepsy. CNS Drugs 16, 669—694. Löscher, W., 2011. Critical review of current animal models of seizures and epilepsy used in the discovery and development of new antiepileptic drugs. Seizure 20, 359—368. Lynch, B.A., Lambeng, N., Nocka, K., Kensel-Hemmes, P., Bajjalieh, S.M., Matagne, A., Fuks, B., 2004. The synaptic vesicle protein SV2A is the binding site for the antiepileptic drug levetiracetam. Proc. Natl. Acad. Sci. U.S.A. 101, 9861—9866. Matagne, A., Klitgaard, H., 1998. Validation of corneally kindled mice: a sensitive screening model for partial epilepsy in man. Epilepsy Res. 31, 59—71. Matagne, A., Margineanu, D.G., Kenda, B., Michel, P., Klitgaard, H., 2008. Anti-convulsive and anti-epileptic properties of brivaracetam (ucb 34714), a high-affinity ligand for the synaptic vesicle protein, SV2A. Br. J. Pharmacol. 154, 1662—1671. Morimoto, K., Fahnestock, M., Racine, R.J., 2004. Kindling and status epilepticus models of epilepsy: rewiring the brain. Prog. Neurobiol. 73, 1—60.
683
Potschka, H., Brodie, M.J., 2012. Pharmacoresistance. Handb. Clin. Neurol. 108, 741—757. Potschka, H., Löscher, W., 1999. Corneal kindling in mice: behavioral and pharmacological differences to conventional kindling. Epilepsy Res. 37, 109—120. Racine, R.J., 1972. Modification of seizure activity by electrical stimulation. II. Motor seizure. Electroencephalogr. Clin. Neurophysiol. 32, 281—294. Rogawski, M.A., Löscher, W., 2004. The neurobiology of antiepileptic drugs. Nat. Rev. Neurosci. 5, 553—564. Rowley, N.M., White, H.S., 2010. Comparative anticonvulsant efficacy in the corneal kindled mouse model of partial epilepsy: correlation with other seizure and epilepsy models. Epilepsy Res. 92, 163—169. Sangdee, P., Turkaris, S.A., Karler, R., Portera Sanchez, A., 1982. Kindling-like effect induced by repeated corneal electro shock in mice. Epilepsia 23, 471—479. Sutula, T.P., 2004. Mechanisms of epilepsy progression: current theories and perspectives from neuroplasticity in adulthood and development. Epilepsy Res. 60, 161—171. Swinyard, E.A., 1972. Electrically induced convulsions. In: Purpura, D.B., Penry, J.K., Tower, D., et al. (Eds.), Experimental Models of Epilepsy. Raven Press, New York, pp. 443—458. Thomas, N.K., Willis, S., Sweetman, L., Borges, K., 2012. Triheptanoin in acute mouse seizure models. Epilepsy Res. 99, 312—317. Toman, J.E.P., 1951. Neuropharmacologic considerations in psychic seizures. Neurology 1, 444—460. Whiting, P.J., 2003. The GABAA receptor gene family: new opportunities for drug development. Curr. Opin. Drug Discov. Dev. 6, 648—657. Wlaz, P., Potschka, H., Löscher, W., 1998. Frontal versus transcorneal stimulation to induce maximal electroshock seizures or kindling in mice and rats. Epilepsy Res. 30, 219—229.
journal homepage: www.elsevier.com/locate/epilepsyres
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model K. Leclercq, A. Matagne, R.M. Kaminski ∗ Neuroscience TA, UCB Pharma, Braine-l’Alleud, Belgium Received 6 June 2013; received in revised form 10 February 2014; accepted 23 February 2014 Available online 5 March 2014
KEYWORDS Kindling; Seizures; Epilepsy; Levetiracetam; 6 Hz; Pharmacoresistance
∗
Summary Corneal kindling is a useful alternative to electrically induced amygdala or hippocampal kindling, which requires advanced surgical and EEG techniques that may not be easily available in many laboratories. Therefore the first aim of this study was to evaluate whether repeated 6 Hz corneal stimulation in mice would lead to an increased and persistent seizure response as described for higher frequency (50/60 Hz) corneal kindling. Male NMRI mice stimulated twice daily (except weekends) for 3 s with 6 Hz electrical current at 44 mA displayed robust kindling development, i.e. a progressive increase in seizure severity. The majority of the animals (about 90%) developed a fully kindled state, defined as at least 10 consecutive stage 3—5 seizures within 5 weeks of corneal stimulation. Afterwards, the fully kindled state was maintained for at least 8 weeks with only two days of stimulations per week. Next, the protective efficacy of four mechanistically different antiepileptic drugs (AEDs; clonazepam, valproate, carbamazepine and levetiracetam) was assessed and compared between 6 Hz and 50 Hz fully kindled mice. All tested AEDs showed a relatively lower potency in the 6 Hz kindling model and a limited efficacy against partial seizures was observed with carbamazepine and levetiracetam. We can conclude that 6 Hz kindling may be more advantageous than the previously described 50/60 Hz corneal kindling models due to its robustness and persistence of the fully kindled state. Furthermore, the observed low potency and limited efficacy of AEDs in 6 Hz fully kindled mice suggest that this model could be a useful tool in the discovery of novel AEDs targeting treatment resistant epilepsy. © 2014 Elsevier B.V. All rights reserved.
Abbreviations: Hz, hertz; EEG, electroencephalogram; mA, milliampere; ms, millisecond; i.p., intraperitoneal. Corresponding author at: UCB Pharma, Chemin du Foriest, R9, B-1420 Braine-l’Alleud, Belgium. Tel.: +32 2 386 6499; fax: +32 2 386 3339. E-mail addresses: [email protected], rafal [email protected] (R.M. Kaminski).
http://dx.doi.org/10.1016/j.eplepsyres.2014.02.013 0920-1211/© 2014 Elsevier B.V. All rights reserved.
676
Introduction Despite the introduction of several novel antiepileptic drugs (AEDs) there is still a sizable population of patients suffering from treatment resistant seizures (Potschka and Brodie, 2012). This creates a strong need for the development of highly differentiated AEDs, which would specifically target this patient population. Innovative approaches for target identification and preclinical testing of candidate molecules are required in order to achieve this goal (Löscher and Schmidt, 2011). In line with this, the epilepsy research community came forth with a consensus proposal and guidance concerning preclinical AEDs discovery. This guidance makes a particular emphasis on the need for implementation of novel models of treatment resistant seizures (Galanopoulou et al., 2012). Traditionally, preclinical AEDs discovery involved relatively simple models, where seizures were acutely induced in naïve animals, which did not reflect underlying pathophysiology of the disease. Therefore chronic epilepsy models, such as seizure kindling, have been increasingly used in drug testing (Löscher, 2011). In kindling, an initially nonconvulsive stimulus applied repeatedly produces a clear progressive change in seizure response leading to so-called fully kindled state, characterized by stable seizure response to each stimulation (Goddard et al., 1969). The kindling process is considered as a relevant model of brain plasticity that may lead to epileptogenesis and progression of epilepsy (Sutula, 2004; Morimoto et al., 2004). There are, in fact, several ways to induce seizure kindling; electrical, chemical or even acoustic stimuli could be used for this purpose (Fisher, 1989; Kupferberg, 2001). Among those paradigms the best characterized is amygdala kindling, where repeated electrical stimulation is applied through an electrode implanted in amygdala (Goddard, 1983; Löscher and Schmidt, 1988). The amygdala kindling model has been widely used for AEDs testing; however it requires invasive stereotactic surgery and EEG equipment, which may not be easily accessible in many labs. An alternative that does not require electrode implantation could be the corneal kindling model in mice (Matagne and Klitgaard, 1998; Potschka and Löscher, 1999). In this model, a low intensity current (2—3 mA) 50 Hz electrical stimulation is applied twice daily for 3 s via corneal electrodes in contrast to a higher current (25—50 mA) single stimulation typically used in the maximal electroshock model (MES). A progressive seizure response develops over a few weeks of stimulations and a large proportion of animals reach fully kindled state and can be used for AEDs testing. The 50 Hz corneal kindling is indeed a useful alternative, but it does not seem to reproduce some of the treatment resistance aspects that had been reported in the amygdala kindling model (Löscher et al., 1986; Matagne and Klitgaard, 1998; Potschka and Löscher, 1999; Rowley and White, 2010). The 6 Hz model of partial seizures, originally developed in the 1950s (Toman, 1951; Brown, 1953; Swinyard, 1972), has been re-introduced in 2001 and proposed as a model of treatment resistant epilepsy (Barton et al., 2001). Since that time the 6 Hz model has been increasingly used for testing of new AEDs in naïve animals (Bialer and White, 2010) and has become a standard screening model in the NINDS
K. Leclercq et al. anticonvulsant screening program (Löscher, 2011). The acute 6 Hz electrical stimulation at 32—44 mA current intensity lasting 3 s produces a partial seizure, stages 1—2 according to Racine’s scale (Racine, 1972) and secondarily generalized seizures (stages 3—5) are never observed (Barton et al., 2001). The first aim of this study was to evaluate whether repeated 6 Hz corneal stimulation in mice would also lead to increased seizure responses and a persistent fully kindled state as described for repeated 50 Hz corneal stimulation (Matagne and Klitgaard, 1998; Potschka and Löscher, 1999). Having confirmed robust and reproducible 6 Hz kindling development, we next compared the effect of four mechanistically diverse AEDs on fully kindled seizures in both the 50 Hz and 6 Hz kindling paradigms. In the course of these experiments we have demonstrated that the efficacy and potency of AEDs in the 6 Hz kindling model is lower than in the 50 Hz kindling paradigm, which is akin to their limited efficacy in the amygdala kindling model. Therefore we propose that 6 Hz corneal kindling may be a novel, convenient model for treatment resistant seizures.
Materials and methods Animals Male NMRI mice (Charles River, France) weighing 19—29 g were used in all experiments. They were maintained on a 12/12-h light/dark cycle with lights on at 06:00 h and had free access to food and drinking water. The temperature in the husbandry was maintained at 20—21 ◦ C and humidity about 40%. For each experiment, the mice were housed in groups of 10 per cage and habituated for at least one hour in the experimental room. All procedures were carried out according to the Helsinki declaration and conducted according to the guidelines of the European Community Council directive 86/609/EEC. A local Ethics Committee approved all performed experiments.
Corneal kindling 6 Hz stimulations The mice (N = 140) were stimulated through corneal electrodes connected to a stimulator (ECT Unit 57800, UgoBasile, Comerio, Italy) using a current intensity of 44 mA, 0.2 ms monopolar pulses at 6 Hz and 3 s duration. A drop of Unicaine (0.4% oxybuprocainum in saline) was placed on the eyes before stimulation to induce local anesthesia and ensure good conductivity. Stimulations were performed twice daily (Monday—Friday), with a 4 h minimum interval between each stimulation, for 3 consecutive weeks. After each stimulation, the animal was gently placed in a plastic bowl (diameter = 19 cm) and observed for 30 s. At the beginning of the kindling process, the mice displayed a range of behaviors including stereotypy, Straub tail, stunned posture and sometimes mild myoclonus. After few days of stimulations approximately 50—60% of the mice began to exhibit secondarily generalized convulsions (stages 3—5) according to Racine’s scale (Racine, 1972). The mice were considered fully kindled when they displayed at least 10 consecutive,
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model
677
Figure 1 Proportion of generalized seizures in mice developing fully kindled state (panel A) and mean seizure severity score with S.D. (panel B) as a function of kindling stimulation days. Male NMRI mice (N = 140, N = 120 in 6 Hz and 50 Hz kindling, respectively) were used. The mice were stimulated through corneal electrodes using current intensity of 44 mA, 0.2 ms monopolar pulses at 6 Hz and 3 s duration. Stimulations were performed twice daily (Monday—Friday), with a 4 h minimum interval between each stimulation, for 3 consecutive weeks. In the 50 Hz kindling mice were stimulated daily (including weekends) using 3 mA current intensity, at 50 Hz pulse frequency for 3 s duration (Matagne and Klitgaard, 1998). Seizures were scored according to Racine’s scale (Racine, 1972). The mice were considered as fully kindled when they displayed at least 10 consecutive, secondarily generalized convulsions (stages 3—5). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
secondarily generalized convulsions. At the end of the third week of stimulations, 60—70% of the mice reached the fully kindled state, which was maintained by twice daily stimulation (Monday, Thursday) every week. The kindling procedure was pursued for 1 or 2 more weeks with remaining nonkindled animals before including them for maintenance. Finally, about 90% of mice reached a fully kindled state within 5 weeks of chronic stimulations. 50 Hz stimulations The 50 Hz stimulations were conducted as described by Matagne and Klitgaard (1998). Briefly, male NMRI mice (N = 120) were stimulated twice daily through corneal electrodes connected to a stimulator (WITT Industrie Elektronik, Berlin, Germany) using 3 mA current intensity, at 50 Hz pulse frequency for 3 s duration. A drop of Unicaine (0.4% oxybuprocainum in saline) was placed on the eyes
Table 1
before stimulation to induce local anesthesia and ensure good conductivity. Stimulations were performed twice daily (including weekends), with a 4 h minimum interval between each stimulation, for 4 consecutive weeks. After each stimulation, the animal was gently placed in a plastic bowl (diameter = 19 cm) and observed for 30 s. At the beginning of the kindling process, the mice displayed jaw movements and facial clonus, head nodding (stages 1—2) according to Racine’s scale (Racine, 1972). After few days of stimulations approximately 20—30% of the mice began to exhibit secondarily generalized convulsions (stages 3—5) according to Racine’s scale (Racine, 1972). The mice were considered fully kindled when they displayed at least 10 consecutive, secondarily generalized convulsions. At the end of the fourth week of stimulations, 50—60% of the mice reached the fully kindled state, which was maintained by twice daily stimulation (Monday, Thursday) every week.
The effects of antiepileptic drugs (AEDs) against partial and generalized seizures in 50 Hz or 6 Hz fully kindled mice.
Drugs
Clonazepam Valproate Carbamazepine Levetiracetam
50 Hz kindling
6 Hz kindling
Generalized
Partial
Generalized
Partial
0.032 (0.030—0.034) 66.1 (65.6—66.5) 5.4 (2.9—10.2) 6.5 (4.8—8.7)
0.048 (0.015—0.15)# 85.0 (64.1—112.8)# MAD 7.6 10.7 (5.8—19.9)
0.054 (0.032—0.089) 92.4 (82.7—103.2)* 8.2 (6.0—11.4) 18.7 (10.4—33.6)**
0.195 (0.094—0.40)*,## 168.5 (89.4—317.4)* 23.6 (10.0—56.0)## 108.0 (35.7—326.1)**,##
The ED50 values, i.e. doses protecting 50% of the animals against partial seizures (stages 1—2) and secondarily generalized seizures (stages 3—5) together with their 95% confidence intervals were calculated and compared using non-linear regression analysis (GraphPad Prism 5, San Diego, CA). MAD, minimal active dose. * P < 0.05. ** P < 0.01 — comparison between 50 Hz and 6 Hz kindling models. # P < 0.05. ## P < 0.01 — comparison between generalized and partial seizures.
678
K. Leclercq et al.
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model
Drug testing Carbamazepine (Sigma), valproate (Sigma) and clonazepam (Hoffman-La Roche; Sequoia Res.) were suspended in 0.9% saline containing 0.1% Tween 80. Levetiracetam (UCB Pharma S.A., Belgium) was dissolved in 0.9% saline. All drugs were administered i.p. at the following pre-administration times (time to peak effect) consistent with their pharmacokinetic profiles in mice: carbamazepine (15 min), valproate (15 min), clonazepam (30 min) and levetiracetam (60 min) (Barton et al., 2001). The drugs were tested in separate groups of kindled animals (N = 9—11) allowing one week wash-out between each trial.
Data analysis and statistics Seizure severity scores were expressed as medians and comparison between pre- and post-drug treatment was performed for each drug and dose using a Wilcoxon signed rank test for paired replicates; comparison between seizure scores in 50 Hz and 6 Hz kindling models was performed for each drug using a Kruskal—Wallis test followed by Dunn’s comparison. The ED50 values (dose protecting 50% of the animals) for partial seizures (stages 1—2) and secondarily generalized seizures (stages 3—5) with 95% confidence intervals were calculated using a non-linear regression analysis (GraphPad Prism 5, San Diego, CA). The dose—response curves and the ED50 values obtained in both kindling paradigms were statistically compared with the same software.
Results 6 Hz kindling development Initially we explored different current intensities (ranging from 14 to 44 mA) with different durations of stimulation (3—5 s) applied once or twice daily for at least 12 consecutive days (data not shown). Based on these initial experiments we chose 44 mA, 3 s stimulation as optimal to achieve robust kindling development within three weeks in the majority (60—70%) of the animals. Fig. 1A illustrates the 6 Hz kindling acquisition (generalized seizures) in those mice which developed a fully kindled state within three weeks. The 50 Hz kindling acquisition data within the same time frame are shown for comparison (Fig. 1A). Furthermore, the 6 Hz kindling procedure was continued for additional 1 or 2 weeks with remaining non-kindled animals and about 90% of mice finally reached the fully kindled state (data not shown).
679
There was a clear progression in seizure severity score from a mean of 1.8 on the first day of 6 Hz stimulations to a mean of 4.3 after three weeks of daily stimulations (except on weekends) (Fig. 1B). In comparison, 50 Hz kindling stimulation resulted in an increase of seizure severity from the mean score of 0, on the first day, up to the mean score of 4 after three weeks of daily stimulations (Fig. 1B). The 6 Hz fully kindled state was maintained for at least 8 weeks, using twice daily maintenance stimulation for two days per week, as described in the methods section. During this time the mean seizure score remained stable at approx. 4.3—4.6 (data not shown).
Comparison of AEDs activity in the 6 Hz and 50 Hz corneal kindling models Even though all tested AEDs displayed protective effects against fully kindled seizures in both kindling models (Fig. 2A—H), the dose—response curves for protection against generalized (scores 3—5) and partial (scores 1—2) seizures revealed interesting differences between the two models (Fig. 2A—H). These also allowed calculation of ED50 values against both generalized and partial seizures as well as comparison between the models (Table 1). Clonazepam (Fig. 2A and B) and valproate (Fig. 2C and D) afforded nearly complete protection against both partial and generalized seizures in both kindling models. Carbamazepine also showed dose-dependent and nearly complete protection against generalized seizures in both models (Fig. 2E), but had only limited efficacy (max. 50—65% protection) against partial seizures (Fig. 2F). In comparison, levetiracetam displayed a different profile in the two kindling models (Fig. 2G and H). A significant difference in the slopes of dose—response curves was observed, slopes which were much steeper in the 50 Hz kindling compared to the 6 Hz kindling (Fig. 2G and H). Levetiracetam displayed only limited protection (75%) against partial seizures in the 6 Hz kindling model (Fig. 2H). In the 50 Hz kindling model, the potency of carbamazepine and levetiracetam against generalized or partial seizures was comparable. Clonazepam and valproate were significantly less potent against partial seizures (Table 1). In the 6 Hz kindling paradigm, we observed a striking decrease in potency of all tested AEDs against partial seizures when compared to generalized seizures (Table 1). Moreover, ED50s against partial seizures in the 6 Hz kindling model were significantly higher when compared to ED50s against partial seizures in the 50 Hz kindling model (Table 1). The median seizure score is often used in the amygdala kindling model to assess the efficacy of AEDs. Consistently, we compared the effects of all tested AEDs on the
Figure 2 Dose—response curves for clonazepam, valproate, carbamazepine and levetiracetam in 6 Hz or 50 Hz fully kindled mice. Left panel illustrates protective effects of antiepileptic drugs against generalized seizures (stages 3—5) according to Racine’s scale (Racine, 1972). Right panel shows protective effects of antiepileptic drugs against partial seizures (stages 1—2). Solid red lines illustrate sigmoidal fitting of the dose—response curves in the 6 Hz kindling model, while dotted blue lines pertain to the 50 Hz kindling model. Red square symbols show percent protection against 6 Hz seizures and blue square symbols illustrate percent protection against 50 Hz seizures. N = 9—11 mice per group. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
680
K. Leclercq et al.
Figure 3 Effects of clonazepam, valproate, carbamazepine and levetiracetam on median seizure severity score in the 50 Hz (left panel) and 6 Hz (right panel) fully kindled mice. Open bars represent median seizure score before drug treatment and solid bars indicate median seizure score post-drug treatment. N = 9—11 mice per group. *P < 0.05, **P < 0.01, ***P < 0.001 — Wilcoxon signed rank test for paired replicates.
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model median seizure score in both 50 Hz and 6 Hz kindling models (Fig. 3A—H). Clonazepam dose-dependently reduced the seizure severity in both kindling models (Fig. 3A and B). Valproate was also efficacious in reducing the median seizure score, but it produced ‘‘all or none’’ response in the 50 Hz kindling model from the dose of 93 mg/kg (Fig. 3C), while the effects in the 6 Hz kindling were dose-dependent (Fig. 3D). Carbamazepine displayed a very flat dose—response curve in the 6 Hz model (Fig. 3F) and also showed ‘‘all or none’’ response at the dose of 7.6 mg/kg in the 50 Hz kindling model (Fig. 3E). Levetiracetam dose-dependently reduced the median seizure scores in both 50 Hz and 6 Hz kindling models, but it was more potent in reducing seizure score in the 50 Hz kindling model (Fig. 3G and H).
Discussion The present study demonstrates for the first time the development of seizure kindling induced by repeated 6 Hz corneal stimulations; a fully kindled state was easily achieved and maintained for several weeks. The potency and efficacy of four mechanistically different AEDs was established in the 6 Hz kindling model and compared with their effects in previously described 50 Hz corneal kindling model (Matagne and Klitgaard, 1998) at optimal pre-administration time (Barton et al., 2001). The tested AEDs showed relatively lower potency and limited efficacy in protection against partial seizures induced with the 6 Hz kindling paradigm suggesting that this model could be a useful tool in the discovery of novel AEDs targeting treatment resistant epilepsy. Corneal kindling models have been described in the literature, but so far only 50 or 60 Hz stimulation frequencies were used (Sangdee et al., 1982; Wlaz et al., 1998; Matagne and Klitgaard, 1998; Potschka and Löscher, 1999). These models were proven to be useful alternatives to the more laborious amygdala kindling model, particularly for drug screening purposes. However, the corneal kindling models described thus far have several limitations. For example, variability in the proportion of mice developing kindling, lack of persistence of the fully kindled state and relatively high mortality rate (10—20%) were observed (Sangdee et al., 1982; Wlaz et al., 1998; Matagne and Klitgaard, 1998; Potschka and Löscher, 1999); the mortality rate could reach up to 40% of kindled mice during longer maintenance phase (Matagne and Klitgaard, 1998). In that context the 6 Hz kindling model described in the present study appears more advantageous, even though the rate of kindling acquisition appears comparable between the two paradigms and partial seizures (scores 1—2) are already present during the first 6 Hz kindling stimulation. The observed differences between the 50 Hz and 6 Hz kindling models may be due to activation of different brain structures following single stimulation with these frequencies as evidenced by differential expression of an immediate early gene (c-fos) (Barton et al., 2001). Expression of c-fos in brainstem areas was only observed after 50 Hz stimulation, which might explain higher lethality observed after 50 Hz kindling. Furthermore, there are obvious differences in seizure semiology between acute MES and 6 Hz models (Barton et al., 2001). Tonic hindlimb extension is the primary endpoint in the MES model and is never observed in the acute 6 Hz model, which is more akin to a
681
focal limbic seizure, despite similar current intensity used in both of these acute models. Consequently, tonic hindlimb extension is occasionally observed at the beginning of the 50 Hz kindling development (Matagne and Klitgaard, 1998; Potschka and Löscher, 1999), which could lead to higher lethality, while these seizures are never observed in the 6 Hz kindling (present study). The proportion of mice developing 6 Hz kindling is very consistent (80—90%) and the fully kindled state persists for at least eight weeks, even after stopping the stimulations during the weekends. In addition, no mortality is associated to low frequency chronic stimulations. Overall the behavioral seizure responses during the development and maintenance of 6 Hz kindling resemble the traditional amygdala kindling which is well described in the literature and claimed to be a suitable model of treatment resistance (Löscher et al., 1986; Löscher, 2002a). In the present study we used four mechanistically different AEDs, i.e. clonazepam, valproate, carbamazepine and levetiracetam (Rogawski and Löscher, 2004). Clonazepam acts as a positive allosteric modulator of GABAA receptor (Whiting, 2003), valproate is a multi-mechanistic drug (Löscher, 2002b), and carbamazepine inhibits voltage-gated sodium channels (Ambrosio et al., 2002). Finally, levetiracetam binds to the synaptic vesicle protein 2A (SV2A), which mediates its anticonvulsant activity (Lynch et al., 2004; Kaminski et al., 2009). The differences in their mechanisms of action could have been responsible for their differential efficacy and potency in the 6 Hz and 50 Hz kindling models. Indeed, the potency of these AEDs was generally lower in the 6 Hz versus 50 Hz kindling model. This was particularly evident when potencies in protection against partial seizures were compared between the two models. Furthermore, the potency loss against partial seizures was much more pronounced in the 6 Hz kindling model. Levetiracetam did not afford complete protection against partial seizures in the 6 Hz kindling, while carbamazepine had limited efficacy against partial seizures in both models. It is worth noting that in the present study the ED50 of levetiracetam for protection against partial seizures was 108 mg/kg, whereas Barton et al. (2001) found that it was 1089 mg/kg. Even though the same stimulation parameters were used, these researchers have performed only acute 6 Hz stimulations in CF1 mice, while we determined it in fully kindled NMRI mice after repeated 6 Hz stimulations. More recently, Rowley and White (2010) have compared the potency of several AEDs in 60 Hz corneal kindling mode (CF1 mice) and found a good correlation between this model and a range of acute seizure models in mice such as the maximal electroshock (MES), the acute 6 Hz model, and the subcutaneous pentylentetrazole (scPTZ) test, as well as with the hippocampal kindled rat model. The present study is therefore the first one to compare the pharmacology of AEDs in two corneal kindling models in mice, i.e. 50 Hz and 6 Hz kindling. It is also the first study to suggest the usefulness of 6 Hz kindling in the context of therapy resistant epilepsy. In the rat amygdala kindling model, the ED50s of clonazepam and carbamazepine against generalized seizures were comparable to those calculated in the 6 Hz kindling model (Löscher and Schmidt, 1988). For carbamazepine and levetiracetam, the ED50s were higher in the rat amygdala kindling model (Löscher and Schmidt, 1988; Matagne et al., 2008), but complete seizure suppression was not achieved in
682 the 6 Hz kindling model despite administration of high doses of these AEDs. Median seizure severity score reduction is another important endpoint in the amygdala kindling model (Racine, 1972). Consequently, we assessed the ability of all four tested AEDs to reduce the median seizure severity score. We observed that the highest tested doses, going beyond clinically relevant therapeutic dose level, were able to reduce the median severity score to 0, but in general the AEDs displayed lower potency in the 6 Hz kindling model and there were still few animals not protected against partial seizures. In amygdala kindling models, depending on the stimulation protocol, it is often difficult to reduce the median seizure score to 0 and to protect all the animals against partial seizures (Löscher et al., 1986). Based on these observations the amygdala kindling model had been proposed as a model of treatment resistant seizures (Löscher et al., 1986). As described above, since persistence of partial seizures despite treatment with high doses of AEDs is present in the 6 Hz corneal kindling model, we suggest the latter can also be considered as a model of treatment resistant seizures. In the present study we have established a robust protocol for 6 Hz kindling in male NMRI mice, but as a next step this should also be confirmed in female mice to assess a potential effect of gender (Potschka and Löscher, 1999). Some modifications of the 6 Hz kindling protocol may be required if other strains of mice are used since there is a well described variability in seizure susceptibility among different mice strains (Frankel et al., 2001; Ferraro et al., 2002). In fact, differences in 6 Hz seizure thresholds have been observed by various groups depending on the mouse strains used (Barton et al., 2001; Hartman et al., 2008; Thomas et al., 2012), but head to head comparisons under the same experimental conditions have not been reported so far. Kindling models may also be useful to study epileptogenesis since they are characterized by a progressive increase in seizure incidence or severity. Consequently, antiepileptogenic therapies have been tested in kindling models (Dudek et al., 2008); for example, levetiracetam showed a robust delay of kindling development in both amygdala and corneal kindling models (Löscher et al., 1998; Matagne et al., 2008; Dudek et al., 2008). It would be interesting to confirm this effect of levetiracetam and other AEDs on 6 Hz kindling development. The usefulness of kindled animals to study epilepsy associated comorbidities (depression, anxiety, cognition, etc.) or treatment adverse effects has also been reported. In fact, amygdala kindled rats display a higher sensitivity to the side effects induced by anticonvulsants, in particular NMDA receptor antagonists (Hönack and Löscher, 1995), which also extends to corneally kindled mice (Klitgaard et al., 2002). Therefore the 6 Hz kindling model may be also useful for the assessment of side effects profiles and determination of therapeutic index as demonstrated in the 50 Hz corneal kindling model (Klitgaard et al., 2002). The present study reports the development and pharmacological validation of the 6 Hz corneal kindling model. The obtained data indicate the robustness of this model and seizure resistance to treatment with several AEDs. Finally, the 6 Hz corneal kindling model may also be used in future
K. Leclercq et al. studies aiming at the development of antiepileptogenic therapies.
Acknowledgments We thank Colette Chaussée, Fabienne Coddens and MarieChristine Tordeur for helpful technical assistance and Prof. Ilse Smolders (Center for Neurosciences, C4N, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Belgium) for critical review of the manuscript.
References Ambrosio, A.F., Soares-da-Silva, P., Carvalho, C.M., Carvalho, A.P., 2002. Mechanisms of action of carbamazepine and its derivatives, oxcarbazepine, BIA 2-093, and BIA 2-024. Neurochem. Res. 27, 121—130. Barton, M.E., Klein, B.D., Wolf, H.H., White, H.S., 2001. Pharmacological characterization of the 6 Hz psychomotor seizure model of partial epilepsy. Epilepsy Res. 47, 217—227. Bialer, M., White, H.S., 2010. Key factors in the discovery and development of new antiepileptic drugs. Nat. Rev. Drug Discov. 9, 68—82. Brown, W.C., 1953. Properties and alterations of electricallyinduced seizures in mice. Epilepsia 2, 127—137. Dudek, F.E., Bertram, E.H., Staley, K.J., 2008. Antiepileptogenesis therapy with levetiracetam: data from kindling versus status epilepticus models. Epilepsy Curr. 8, 28—30. Ferraro, T.N., Golden, G.T., Smith, G.G., DeMuth, D., Buono, R.J., Berrettini, W.H., 2002. Mouse strain variation in maximal electroshock seizure threshold. Brain Res. 936, 82—86. Fisher, R.S., 1989. Animal models of the epilepsies. Brain Res. Rev. 14, 245—278. Frankel, W.N., Taylor, L., Beyer, B., Tempel, B.L., White, H.S., 2001. Electroconvulsive thresholds of inbred mouse strains. Genomics 74, 306—312. Galanopoulou, A.S., Buckmaster, P.S., Staley, K.J., Moshé, S.L., Perucca, E., Engel Jr., J., Löscher, W., Noebels, J.L., Pitkänen, A., Stables, J., White, H.S., O‘Brien, T.J., Simonato, M., 2012. Identification of new epilepsy treatments: issues in preclinical methodology. Epilepsia 53, 571—582. Goddard, G.V., McIntyre, D.C., Leech, C.K., 1969. A permanent change in brain function resulting from daily electrical stimulation. Exp. Neurol. 25, 295—330. Goddard, G.V., 1983. The kindling model of epilepsy. Trends Neurosci. 6, 275—279. Hartman, A.L., Lyle, M., Rogawski, M.A., Gasior, M., 2008. Efficacy of the ketogenic diet in the 6-Hz seizure test. Epilepsia 49, 334—339. Hönack, D., Löscher, W., 1995. Kindling increases the sensitivity of rats to adverse effects of certain antiepileptic drugs. Epilepsia 36, 763—771. Kaminski, R.M., Gillard, M., Leclercq, K., Hanon, E., Lorent, G., Dassesse, D., Matagne, A., Klitgaard, H., 2009. Proepileptic phenotype of SV2A-deficient mice is associated with reduced anticonvulsant efficacy of levetiracetam. Epilepsia 50, 1729—1740. Klitgaard, H., Matagne, A., Lamberty, Y., 2002. Use of epileptic animals for adverse effect testing. Epilepsy Res. 50, 55—65. Kupferberg, H., 2001. Animal models used in the screening of antiepileptic drugs. Epilepsia 42, S7—S12. Löscher, W., 2002a. Animal models of epilepsy for the development of antiepileptogenic and disease-modifying drugs. A comparison of the pharmacology of kindling and post-status epilepticus models of temporal lobe epilepsy. Epilepsy Res. 50, 105—123.
Low potency and limited efficacy of antiepileptic drugs in the mouse 6 Hz corneal kindling model Löscher, W., Hönack, D., Rundfeldt, C., 1998. Antiepileptogenic effects of the novel anticonvulsant levetiracetam (ucb L059) in the kindling model of temporal lobe epilepsy. J. Pharmacol. Exp. Ther. 284, 474—479. Löscher, W., Jäckel, R., Czuczwar, S.J., 1986. Is amygdala kindling in rats a model for drug-resistant partial epilepsy? Exp. Neurol. 93, 211—226. Löscher, 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. Löscher, W., Schmidt, D., 2011. Modern antiepileptic drug development has failed to deliver: ways out of the current dilemma. Epilepsia 52, 657—678. Löscher, W., 2002b. Basic pharmacology of valproate: a review after 35 years of clinical use for the treatment of epilepsy. CNS Drugs 16, 669—694. Löscher, W., 2011. Critical review of current animal models of seizures and epilepsy used in the discovery and development of new antiepileptic drugs. Seizure 20, 359—368. Lynch, B.A., Lambeng, N., Nocka, K., Kensel-Hemmes, P., Bajjalieh, S.M., Matagne, A., Fuks, B., 2004. The synaptic vesicle protein SV2A is the binding site for the antiepileptic drug levetiracetam. Proc. Natl. Acad. Sci. U.S.A. 101, 9861—9866. Matagne, A., Klitgaard, H., 1998. Validation of corneally kindled mice: a sensitive screening model for partial epilepsy in man. Epilepsy Res. 31, 59—71. Matagne, A., Margineanu, D.G., Kenda, B., Michel, P., Klitgaard, H., 2008. Anti-convulsive and anti-epileptic properties of brivaracetam (ucb 34714), a high-affinity ligand for the synaptic vesicle protein, SV2A. Br. J. Pharmacol. 154, 1662—1671. Morimoto, K., Fahnestock, M., Racine, R.J., 2004. Kindling and status epilepticus models of epilepsy: rewiring the brain. Prog. Neurobiol. 73, 1—60.
683
Potschka, H., Brodie, M.J., 2012. Pharmacoresistance. Handb. Clin. Neurol. 108, 741—757. Potschka, H., Löscher, W., 1999. Corneal kindling in mice: behavioral and pharmacological differences to conventional kindling. Epilepsy Res. 37, 109—120. Racine, R.J., 1972. Modification of seizure activity by electrical stimulation. II. Motor seizure. Electroencephalogr. Clin. Neurophysiol. 32, 281—294. Rogawski, M.A., Löscher, W., 2004. The neurobiology of antiepileptic drugs. Nat. Rev. Neurosci. 5, 553—564. Rowley, N.M., White, H.S., 2010. Comparative anticonvulsant efficacy in the corneal kindled mouse model of partial epilepsy: correlation with other seizure and epilepsy models. Epilepsy Res. 92, 163—169. Sangdee, P., Turkaris, S.A., Karler, R., Portera Sanchez, A., 1982. Kindling-like effect induced by repeated corneal electro shock in mice. Epilepsia 23, 471—479. Sutula, T.P., 2004. Mechanisms of epilepsy progression: current theories and perspectives from neuroplasticity in adulthood and development. Epilepsy Res. 60, 161—171. Swinyard, E.A., 1972. Electrically induced convulsions. In: Purpura, D.B., Penry, J.K., Tower, D., et al. (Eds.), Experimental Models of Epilepsy. Raven Press, New York, pp. 443—458. Thomas, N.K., Willis, S., Sweetman, L., Borges, K., 2012. Triheptanoin in acute mouse seizure models. Epilepsy Res. 99, 312—317. Toman, J.E.P., 1951. Neuropharmacologic considerations in psychic seizures. Neurology 1, 444—460. Whiting, P.J., 2003. The GABAA receptor gene family: new opportunities for drug development. Curr. Opin. Drug Discov. Dev. 6, 648—657. Wlaz, P., Potschka, H., Löscher, W., 1998. Frontal versus transcorneal stimulation to induce maximal electroshock seizures or kindling in mice and rats. Epilepsy Res. 30, 219—229.
