Antioxidants and free radical scavengers do not consistently delay seizure onset in animal models of acute seizures

Introduction

A number of plant products, which have been shown to have direct antioxidant activity, are now being shown to have anticonvulsant activity [1–4]. Specifically, the antioxidant herbal products appear to delay seizure onset in naïve animals after administration of chemical convulsants. These studies are presented in such a way that the antioxidant activity is assumed to underlie this seizure-delaying activity of the plant products. This implies the hypothesis that generation of reactive oxygen species, such as H₂O₂, superoxide anion (.O2^{) and the hydroxyl radical (.OH), is critical to the initiation of seizure activity in these models.}

There is considerable evidence that neuronal damage after generalized status epilepticus is due to generation of reactive oxygen species [5]. Oxidative stress occurs as a consequence of prolonged seizures and may contribute to seizure-induced brain damage and to the generation of the epileptic state, in which there are spontaneous seizures [6]. The generation of free radicals has also been demonstrated after a single generalized seizure induced with pentylenetetrazol using a molecular trap [7]. However, the idea that generation of reactive oxygen species is involved in the onset of an acute seizure has not been investigated. Therefore, the conclusion that the antioxidant activity of herbal preparation underlies the seizure-delaying activity of these preparations is unsubstantiated. This study tested the ability of a series of established antioxidants, which directly inactivate free radicals and have been shown to have activity in the central nervous system, to slow, or block, the onset of seizures. The compounds were tested in 3 animal models of acute seizures that are characterized by a relatively gradual build-up and spread of the epileptic activity in order to focus on the earliest phase of seizure onset and synchronization. The models chosen are also those in which herbal products have been shown to delay seizure onset.

Methods

All animal experiments were carried out in accordance with the NIH guide for the care and use of laboratory animals (NIH publication No. 8023, revised 1996) and with the approval of the local Animal Use Committee. Adult Sprague-Dawley rats (male and female, Harlan Sprague-Dawley, Indianapolis IN) weighing 150–250 g (6–8 weeks old) were used for this study. These animals were housed in a room with controlled temperature (22 ± 1 ºC) and humidity (50 ± 5%). All chemicals were obtained from Sigma Chemical Co (St. Louis, MO), except where indicated.

Pilocarpine, pentylenetetrazol (PTZ) and kainic acid (Ascent Scientific, Weston-Super-Mare, United Kingdom) were dissolved in normal saline and the pH adjusted to 7.4 for intraperitoneal (pilocarpine and kainic acid) or subcutaneous (PTZ) administration. The subcutaneous route of administration was used for PTZ to rule out a possible physical interaction between the convulsant and the antioxidant and because the latency to the first seizure is longer after subcutaneous than after intraperitoneal administration. Methylscopolamine was administrated to animals (1mg/kg, subcutaneously) 10 min prior to the pilocarpine. Previous experiments have shown that 10 mg/kg kainic acid, 75 mg/kg PTZ or 300 mg/kg pilocarpine are the minimal doses that result in behavioral seizure activity in all naive animals [3]. Therefore, these doses were used in the current study. The time from injection of each chemical convulsant to the first appearance of seizure activity was measured for each animal and is referred to as the seizure latency. After administration of kainic acid, this was the time to the first wet dog shake. After pilocarpine, this was defined as the time to the first appearance of forelimb clonus. For the PTZ-induced seizures, the total duration of the behavioral seizure activity was also measured for each animal.

For pilocarpine-induced seizures, the behavioral seizures were scored according to the scale of Setkowicz et al [8]. A score of 1 was given for immobility, piloerection, salivation, face and vibrissae twitching and rubbing the ears with the forepaws. A score of 2 was given for head nodding and chewing. Clonic movement of the forelimbs was scored 3, while clonic movement of fore- and hindlimbs was scored 4. Rearing and falling was scored 5 and loss of postural tone with general rigidity was scored 6. For kainic acid-induced seizures, the behavioral seizures were scored according to the scale of Giusti et al [9]. Head nodding was scored 1, a wet dog shake was scored 2, piano playing movements of the forelimbs was scored 3, a clonic convulsion with rearing and falling was scored 4, and a generalized convulsion with tonic extension was scored 5. The behavioral seizures induced by pentylenetetrazol were scored according to the scale of Ilhan et al [10]. Ear and facial twitching was scored 1, a convulsive wave was scored 2, myoclonic jerks were scored 3, a generalized convulsion turning onto the side was scored 4, and a generalized convulsion with tonic extension was scored 5. For each convulsant, each animal was assigned the score of the most severe seizure observed.

Trolox^{, a vitamin E analog (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid, Calbiochem/VWR (USA)), has been shown to increase the latency to first seizure after pilocarpine at a dose of 200 mg/kg [}11] and slow the onset of seizures in the ferrous chloride model (250 mg/kg), but has no effect in the PTZ-threshold test, the maximal electroshock model, the kainic acid model or the kindling model (100 to 400 mg/kg) [12,13]. The dominant action of vitamin E is to quench lipid peroxyl radicals [14]. For the present experiments, Trolox ^{was dissolved in normal saline at a concentration of 10 or 100 mg/ml and administered intraperitonally at doses between 3 and 200 mg/kg 10 min before the convulsant. Ten minutes was chosen to allow time for absorption of the antioxidant before absorption of the convulsant.}

Vitamin C (ascorbic acid) has been reported to significantly increase the latency to first seizure, reduce seizure severity and improve survival in the pilocarpine model at a dose of 250 mg/kg given 30 minutes before pilocarpine [15]. Vitamin C has also been shown to have efficacy against penicillin-induced seizures at intermediate doses (100–200 mg/kg ip) [16], but is reported to be ineffective in the kainic acid model at 50 mg/kg [17]. In the present study, vitamin C was dissolved in normal saline and administered intraperitoneally at a dose of 250 mg/kg 10 min before the convulsant.

Melatonin (N-acetyl-5-methoxytryptamine) is synthesized by the pineal gland and secreted into the cerebrospinal fluid in a circadian pattern. Melatonin has a number of effects on physiological function, including scavenging of hydroxyl and peroxyl radicals [18–21] and protecting neurons from potentially lethal insults [22,23]. Although it’s anticonvulsant activity has been debated, it has been shown, at a dose of ~23 mg/kg (0.1–0.2 mmol/kg), to reduce the seizures in a rat model of post-traumatic epilepsy [24]. It has also been reported to increase the latency to penicillin-induced epileptiform activity in the cortex of rats with administration 10 min before the penicillin [25]. Melatonin, at 20 mg/kg ip, can also block the induction of seizures induced with kainic acid in mice when administered simultaneously with the kainic acid and reduce the severity of the seizures when administered 30 min before the kainic acid [26]. For the present experiments, melatonin was dissolved in ethanol [9] and further diluted in normal saline for a final ethanol concentration of 5%. It was administered intraperitoneally at a dose of 20 mg/kg 10 min before the convulsant.

D,L-α-lipoic acid (1,2-dithiolane-3-pentanoic acid, ALA), a normal cellular constituent, is readily absorbed from the diet and can cross the blood-brain barrier [27]. It, and its reduced form, dihydrolipoic acid, have been found to scavenge hydroxyl radicals, hypochlorous acid and singlet oxygen, but probably do not scavenge peroxyl radicals [14,28,29]. The dithiol group can be used for regeneration of vitamin E, vitamin C and glutathione. Adult rats pretreated with α-lipoic acid (100 mg/kg, ip) had 55% less seizure activity in a model of traumatic brain injury [30]. In the present study, α-lipoic acid was dissolved in sodium bicarbonate (7.5% w/v) and administered intraperitoneally at a dose of 25 mg/kg 10 min before the convulsant [2].

Reduced glutathione (GSH) was assayed in the cortex from animals in the pilocarpine group using the enzymatic recycling method according to Mastocola et al [31] and Rahman et al [32]. The pilocarpine model was chosen because it was the model in which there were significant effects with some of the antioxidants. At the appropriate time, animals were anesthetized with urethane and perfused through the heart with ice cold phosphate buffered saline. The brain was removed and placed on ice while the cortex was dissected from the remainder of the brain. The seizures induced by pilocarpine begin in the cortex [33]. Therefore, by excluding the rest of the brain, it was hoped that there would be an increase in the sensitivity of the assay. The cortex was weighed and then homogenized in 0.1% Triton X-100 and 0.6% sulfosalicylic acid (w/v in potassium phosphate buffer with EDTA) to give a 10% homogenate. The homogenate was centrifuged at 8,000 rpm for 20 min at 4°C. The supernatant was saved for GSH assay. An aliquot of the sample was added to a mixture of 0.1M potassium phosphate buffer with 5 mM EDTA, pH 7.5 and 10 mM 5,5′-dithio-bis(2-nitrobenzoic acid) (DTNB) directly in the cuvette. The GSH concentration was evaluated after 2 min by measuring the absorbance at 412 nm and comparing to a standard curve. The results are expressed as μmol/g tissue.

The latency to seizure onset, seizure duration, seizure score and GSH levels were averaged across animals in the same treatment group. Comparisons between groups were done with an ANOVA with post-hoc test (Dunnett’s) with comparison to the appropriate control. The mortality was analyzed with a contingency table using Fisher’s Exact test. A p <0.05 was considered significant.

Results

Effect of antioxidants in the pilocarpine model of acute seizures

Animals treated with vehicle (normal saline, n=13, 5% ethanol, n=2, 7.5% bicarbonate, n=2 for a total of 17) followed by pilocarpine all had severe seizures (2 with score 4, 5 with score 5 and 9 with score 6) and 4 died (Table I, Figure 1). The mean latency to the first episode of forelimb clonus was 16.9 ± 1.2 min and the mean latency to the first episode of rearing was 29.5 ± 3.4 min for the animals that progressed that far. The mean seizure score was 5.4 ± 0.2. Of the 10 animals treated with 20 mg/kg Trolox ^{(n=10), only 1 animal progressed to rearing and falling (score 5). In the remaining 9 animals, 2 had a seizure score of 3 and 7 were scored 4. There was a significant increase in the latency to rearing and a significant decrease in seizure score.}

Figure 1

Effect of antioxidants in the pilocarpine model of seizures. Mean ± SEM for the latency to forelimb clonus (A), latency to first episode of rearing (B) and seizure score (C) are presented for each of the antioxidants. Number of animals in each group is presented on the graph in C. Abbreviations and doses: CON = vehicle control, Trolox ^{= 20 mg/kg, Vit C = vitamin C, 250 mg/kg, Mel = melatonin, 20 mg/kg, ALA = α-lipoic acid, 25 mg/kg. * = p<0.05, 1-way ANOVA with Dunnett’s multiple comparisons to control.}

Table I

Mortality

	Control	Trolox®	Vit. C	Melatonin	α-LA
Pilocarpine	4/17	0/10	0/8	3/5	0/5
PTZ	6/9	1/6	0/5*	1/6	2/6
Kainic acid	0/5	0/5	0/5	0/5	0/5

^{p< 0.05 compared to vehicle control using Fisher’s Exact test.}

Because of the range of doses of Trolox ^{that have been reported to be effective, additional doses were tested in the pilocarpine model of acute seizures (}Figure 2). No clear dose response curve for the effects of Trolox ^{on seizure parameters was evident. The dose of 3 mg/kg was not different from the control animals. The highest dose of 200 mg/kg was also not significantly different from control animals for measurements of latency to forelimb clonus or seizure score. There was a significant increase in the latency to rearing.}

Figure 2

Dose response curve for Trolox ^{in the pilocarpine model of seizures. Mean ± SEM for the latency to forelimb clonus (A), latency to first episode of rearing (B) and seizure score (C) are presented for 4 doses of Trolox. Number of animals for each dose is presented on the graph in C. CON represents vehicle control. * = p<0.05, 1-way ANOVA with Dunnett’s multiple comparisons to control.}

For the animals treated with 250 mg/kg vitamin C (n=14), 3 animals did not progress beyond head bobbing seizures (score 2). There was a significant increase in the latency to first forelimb clonus and first episode of rearing after treatment with vitamin C. There was also a significant decrease in the mean seizure score. Melatonin, at 20 mg/kg (n=5), caused an increase in the latency to the first episode of rearing, but no significant change in latency to forelimb clonus or seizure score. α-Lipoic acid, at 25 mg/kg (n=5), significantly increased the latency to the first forelimb clonus and first episode of rearing. It also significantly reduced the seizure score and completely prevented severe seizures in one animal (score 2). None of these antioxidants statistically altered the mortality after pilocarpine-induced seizures. However, no animal treated with vitamin C, Trolox^{, or α-lipoic acid died in the 24 hours after the status epilepticus.}

Effect of antioxidants in the pentylenetetrazol model of acute seizures

Because there was a significant effect with the antioxidants in the pilocarpine model, the antioxidants were tested in 2 additional models of acute seizures to determine whether the effect was specific for the pilocarpine model. In the pentylenetetrazol model, animals treated with vehicle followed by subcutaneous pentylenetetrazol (PTZ, n=9) all had severe seizures (2 with score 4, 7 with score 5, 6 animals died; Table I, Figure 3). The mean latency to generalized seizure was 5.7 ± 0.7 min and the mean seizure duration was 11.6 ± 1.0 sec. Of the 6 animals treated with 20 mg/kg Trolox ^{(n=10), only 1 animal died. Of the remaining animals, 1 had a seizure score of 3 and the remaining were scored 4 or 5. However, none of the seizure parameters were statistically different from the vehicle control. All of the animals treated with 250 mg/kg vitamin C (n=5) had a seizure score of 4 and none of the animals died. The reduction in mortality was statistically significant. Melatonin, at 20 mg/kg (n=5), completely blocked the seizures in one animal and a second one had a significant increase in the latency to generalized seizure (40 min). However, on average there was not a statistically significant effect of melatonin in the PTZ model. Treatment with α-lipoic acid, at 25 mg/kg (n=5), caused a trend towards a decrease in seizure duration, which did not quite reach statistical significance. None of the antioxidants statistically altered the mortality after pentylenetetrazol-induced seizures.}

Figure 3

Effect of antioxidants in the pentylenetetrazol model of seizures. Mean ± SEM for the latency to generalized seizure (A), total seizure duration (B) and seizure score (C) are presented for each of the antioxidants. Number of animals in each group is presented on the graph in C. Abbreviations and doses: CON = vehicle control, Trolox ^{= 20 mg/kg, Vit C = vitamin C, 250 mg/kg, Mel = melatonin, 20 mg/kg, ALA = α-lipoic acid, 25 mg/kg.}

Effect of antioxidants in the kainic acid model of acute seizures

Animals treated with vehicle followed by kainic acid (n=5) all had severe seizures (2 with score 4 and 3 with score 5, Figure 4), but none died. The mean latency to the first wet dog shake was 47.6 ± 3.3 min and the mean seizure score was 4.6 ± 0.2. Of the 5 animals treated with 20 mg/kg Trolox^{, 1 animal had a seizure score of 3 and the remaining were scored 4. Overall there was a trend towards a decrease in seizure score, but not in the latency to first wet dog shake. Pretreatment with vitamin C, at 250 mg/kg (n=5), melatonin, at 20 mg/kg (n=5), α-lipoic acid, at 25 mg/kg (n=5) resulted in no significant change in latency to wet dog shake or seizure score.}

Figure 4

Effect of antioxidants in the kainic acid model of seizures. Mean ± SEM for the latency to first wet dog shake (A) and seizure score (B) are presented for each of the antioxidants. There were 5 animals in each treatment group. Abbreviations and doses: CON = vehicle control, Trolox ^{= 20 mg/kg, Vit C = vitamin C, 250 mg/kg, Mel = melatonin, 20 mg/kg, ALA = α-lipoic acid, 25 mg/kg. * = p<0.05, 1-way ANOVA with Dunnett’s multiple comparisons to control.}

Effect of seizures, and vitamins E and C on glutathione levels

The reduced form of glutathione (GSH) is an important free radical scavenger in the mammalian nervous system [34,35] and an endogenous anticonvulsant [36]. If free radicals and reactive oxygen species are generated by the increasing neuronal activity associated with seizure initiation and reduced glutathione interacts with these reactive species, then one might predict an acute decrease in the available reduced glutathione early in the seizures induced by pilocarpine. Therefore, GSH was measured in cortex from animals given pilocarpine - the model in which some of the antioxidants were effective.

Reduced glutathione (GSH) levels were determined in tissue from the cortex of naïve animals (n=13) and compared to levels in the cortex from animals after administration of pilocarpine (Figure 5). Animals (n=4) were sacrificed 15 minutes after administration of pilocarpine, which is just as the behavioral seizures begin. Additional animals (n=6) were sacrificed 2 hours after administration of pilocarpine when the generalized status epilepticus had been ongoing for at least 1.5 hours. Finally, another group of animals (n=5) were sacrificed 24 hours after administration of pilocarpine to determine the longer term effects of the prolonged status epilepticus. There was no change in levels of reduced glutathione at 15 min or 2 hours after administration of pilocarpine. However, there was a statistically significant decrease in GSH at 24 hours.

Figure 5

Effect of Trolox ^{and vitamin C on the changes in glutathione in the pilocarpine model of status epilepticus. Mean (± SEM) glutathione levels in the cortex are presented in naïve control animals and in animals at 3 times points after administration of pilocarpine. Pilocarpine alone had a significant effect at 24 hours, but not 15 min or 2 hours, compared to the naïve control (p<0.05, 1-way ANOVA with post-hoc Dunnett’s comparison). Within each time point, the administration of either Trolox (20 mg/kg) or vitamin C (250 mg/kg) had no significant effect on the glutathione levels. Number of animals in each group is indicated above the bar.}

Treatment with Trolox ^{or vitamin C had been previously shown to significantly slow seizure onset and reduce seizure score in this model of acute seizures. Therefore, glutathione levels from the cortex of animals pretreated with Trolox (20 mg/kg) or vitamin C (250 mg/kg) before pilocarpine were compared to animals receiving only pilocarpine. There was no effect of either antioxidant on the glutathione levels at any of the time points tested.}

Effect of antioxidants in the pilocarpine model of acute seizures

Animals treated with vehicle (normal saline, n=13, 5% ethanol, n=2, 7.5% bicarbonate, n=2 for a total of 17) followed by pilocarpine all had severe seizures (2 with score 4, 5 with score 5 and 9 with score 6) and 4 died (Table I, Figure 1). The mean latency to the first episode of forelimb clonus was 16.9 ± 1.2 min and the mean latency to the first episode of rearing was 29.5 ± 3.4 min for the animals that progressed that far. The mean seizure score was 5.4 ± 0.2. Of the 10 animals treated with 20 mg/kg Trolox ^{(n=10), only 1 animal progressed to rearing and falling (score 5). In the remaining 9 animals, 2 had a seizure score of 3 and 7 were scored 4. There was a significant increase in the latency to rearing and a significant decrease in seizure score.}

Figure 1

Effect of antioxidants in the pilocarpine model of seizures. Mean ± SEM for the latency to forelimb clonus (A), latency to first episode of rearing (B) and seizure score (C) are presented for each of the antioxidants. Number of animals in each group is presented on the graph in C. Abbreviations and doses: CON = vehicle control, Trolox ^{= 20 mg/kg, Vit C = vitamin C, 250 mg/kg, Mel = melatonin, 20 mg/kg, ALA = α-lipoic acid, 25 mg/kg. * = p<0.05, 1-way ANOVA with Dunnett’s multiple comparisons to control.}

Because of the range of doses of Trolox ^{that have been reported to be effective, additional doses were tested in the pilocarpine model of acute seizures (}Figure 2). No clear dose response curve for the effects of Trolox ^{on seizure parameters was evident. The dose of 3 mg/kg was not different from the control animals. The highest dose of 200 mg/kg was also not significantly different from control animals for measurements of latency to forelimb clonus or seizure score. There was a significant increase in the latency to rearing.}

Figure 2

Dose response curve for Trolox ^{in the pilocarpine model of seizures. Mean ± SEM for the latency to forelimb clonus (A), latency to first episode of rearing (B) and seizure score (C) are presented for 4 doses of Trolox. Number of animals for each dose is presented on the graph in C. CON represents vehicle control. * = p<0.05, 1-way ANOVA with Dunnett’s multiple comparisons to control.}

For the animals treated with 250 mg/kg vitamin C (n=14), 3 animals did not progress beyond head bobbing seizures (score 2). There was a significant increase in the latency to first forelimb clonus and first episode of rearing after treatment with vitamin C. There was also a significant decrease in the mean seizure score. Melatonin, at 20 mg/kg (n=5), caused an increase in the latency to the first episode of rearing, but no significant change in latency to forelimb clonus or seizure score. α-Lipoic acid, at 25 mg/kg (n=5), significantly increased the latency to the first forelimb clonus and first episode of rearing. It also significantly reduced the seizure score and completely prevented severe seizures in one animal (score 2). None of these antioxidants statistically altered the mortality after pilocarpine-induced seizures. However, no animal treated with vitamin C, Trolox^{, or α-lipoic acid died in the 24 hours after the status epilepticus.}

Effect of antioxidants in the pentylenetetrazol model of acute seizures

Because there was a significant effect with the antioxidants in the pilocarpine model, the antioxidants were tested in 2 additional models of acute seizures to determine whether the effect was specific for the pilocarpine model. In the pentylenetetrazol model, animals treated with vehicle followed by subcutaneous pentylenetetrazol (PTZ, n=9) all had severe seizures (2 with score 4, 7 with score 5, 6 animals died; Table I, Figure 3). The mean latency to generalized seizure was 5.7 ± 0.7 min and the mean seizure duration was 11.6 ± 1.0 sec. Of the 6 animals treated with 20 mg/kg Trolox ^{(n=10), only 1 animal died. Of the remaining animals, 1 had a seizure score of 3 and the remaining were scored 4 or 5. However, none of the seizure parameters were statistically different from the vehicle control. All of the animals treated with 250 mg/kg vitamin C (n=5) had a seizure score of 4 and none of the animals died. The reduction in mortality was statistically significant. Melatonin, at 20 mg/kg (n=5), completely blocked the seizures in one animal and a second one had a significant increase in the latency to generalized seizure (40 min). However, on average there was not a statistically significant effect of melatonin in the PTZ model. Treatment with α-lipoic acid, at 25 mg/kg (n=5), caused a trend towards a decrease in seizure duration, which did not quite reach statistical significance. None of the antioxidants statistically altered the mortality after pentylenetetrazol-induced seizures.}

Figure 3

Effect of antioxidants in the pentylenetetrazol model of seizures. Mean ± SEM for the latency to generalized seizure (A), total seizure duration (B) and seizure score (C) are presented for each of the antioxidants. Number of animals in each group is presented on the graph in C. Abbreviations and doses: CON = vehicle control, Trolox ^{= 20 mg/kg, Vit C = vitamin C, 250 mg/kg, Mel = melatonin, 20 mg/kg, ALA = α-lipoic acid, 25 mg/kg.}

Effect of antioxidants in the kainic acid model of acute seizures

Animals treated with vehicle followed by kainic acid (n=5) all had severe seizures (2 with score 4 and 3 with score 5, Figure 4), but none died. The mean latency to the first wet dog shake was 47.6 ± 3.3 min and the mean seizure score was 4.6 ± 0.2. Of the 5 animals treated with 20 mg/kg Trolox^{, 1 animal had a seizure score of 3 and the remaining were scored 4. Overall there was a trend towards a decrease in seizure score, but not in the latency to first wet dog shake. Pretreatment with vitamin C, at 250 mg/kg (n=5), melatonin, at 20 mg/kg (n=5), α-lipoic acid, at 25 mg/kg (n=5) resulted in no significant change in latency to wet dog shake or seizure score.}

Figure 4

Effect of antioxidants in the kainic acid model of seizures. Mean ± SEM for the latency to first wet dog shake (A) and seizure score (B) are presented for each of the antioxidants. There were 5 animals in each treatment group. Abbreviations and doses: CON = vehicle control, Trolox ^{= 20 mg/kg, Vit C = vitamin C, 250 mg/kg, Mel = melatonin, 20 mg/kg, ALA = α-lipoic acid, 25 mg/kg. * = p<0.05, 1-way ANOVA with Dunnett’s multiple comparisons to control.}

Effect of seizures, and vitamins E and C on glutathione levels

The reduced form of glutathione (GSH) is an important free radical scavenger in the mammalian nervous system [34,35] and an endogenous anticonvulsant [36]. If free radicals and reactive oxygen species are generated by the increasing neuronal activity associated with seizure initiation and reduced glutathione interacts with these reactive species, then one might predict an acute decrease in the available reduced glutathione early in the seizures induced by pilocarpine. Therefore, GSH was measured in cortex from animals given pilocarpine - the model in which some of the antioxidants were effective.

Reduced glutathione (GSH) levels were determined in tissue from the cortex of naïve animals (n=13) and compared to levels in the cortex from animals after administration of pilocarpine (Figure 5). Animals (n=4) were sacrificed 15 minutes after administration of pilocarpine, which is just as the behavioral seizures begin. Additional animals (n=6) were sacrificed 2 hours after administration of pilocarpine when the generalized status epilepticus had been ongoing for at least 1.5 hours. Finally, another group of animals (n=5) were sacrificed 24 hours after administration of pilocarpine to determine the longer term effects of the prolonged status epilepticus. There was no change in levels of reduced glutathione at 15 min or 2 hours after administration of pilocarpine. However, there was a statistically significant decrease in GSH at 24 hours.

Figure 5

Effect of Trolox ^{and vitamin C on the changes in glutathione in the pilocarpine model of status epilepticus. Mean (± SEM) glutathione levels in the cortex are presented in naïve control animals and in animals at 3 times points after administration of pilocarpine. Pilocarpine alone had a significant effect at 24 hours, but not 15 min or 2 hours, compared to the naïve control (p<0.05, 1-way ANOVA with post-hoc Dunnett’s comparison). Within each time point, the administration of either Trolox (20 mg/kg) or vitamin C (250 mg/kg) had no significant effect on the glutathione levels. Number of animals in each group is indicated above the bar.}

Treatment with Trolox ^{or vitamin C had been previously shown to significantly slow seizure onset and reduce seizure score in this model of acute seizures. Therefore, glutathione levels from the cortex of animals pretreated with Trolox (20 mg/kg) or vitamin C (250 mg/kg) before pilocarpine were compared to animals receiving only pilocarpine. There was no effect of either antioxidant on the glutathione levels at any of the time points tested.}

Discussion

A common feature of herbal products with reported free radical scavenging activity is the ability to delay the onset of behavioral seizures after administration of chemical convulsants, such as kainic acid, pilocarpine or PTZ, in naïve animals [1–3]. Because the antioxidant activity has been assumed to underlie the delay in seizure onset, these observations suggested the hypothesis that, early in the onset of chemically-induced seizures, the increase in neuronal activity generates free radicals, which, in turn, play a role in synchronization and spread of the epileptiform activity in these models. If this hypothesis is correct, then antioxidants, or free radical scavengers would be predicted to interact with the free radicals resulting in an increase in latency to seizure onset. The results of this study do not support this hypothesis. Most importantly, the antioxidant activity of the compounds in the herbal preparations cannot account for the anticonvulsant activity of these preparations. Thus, there is anticonvulsant activity in the herbal products for which the mechanism remains to be determined.

While it is possible that the correct dose of a particular antioxidant was not given at the perfect time relative to seizure onset, the best result in this set of experiments was a modest increase in latency and decrease in seizure score after administration of pilocarpine. The results did not generalize to the PTZ or kainic acid models, using the same dose, route of administration and timing of administration that were effective in the pilocarpine model. The results of the assay for glutathione levels also do not support the general hypothesis. Thus, the results of these experiments indicate that antioxidants cannot, in general, delay the onset of acute seizures induced with chemical convulsants. Taking this conclusion a step further, the results suggest that free radicals, or reactive oxygen species, do not play a significant role in the early events in seizure onset (synchronization and spread) in these same models. This does not rule out a role of free radicals in the process that leads to neuronal death and spontaneous seizures after sustained status epilepticus induced with pilocarpine or kainic acid. Nor is a role of reactive oxygen species in the maintenance of status epilepticus ruled out. Finally, a role of free radicals in the initiation of seizures in epileptic brains has not been determined. The timing of antioxidant administration, 10 minutes, would only demonstrate a direct scavenging of free radicals by the antioxidants. If a secondary action is needed, such as gene activation, then this effect would be missed by the dosing schedule chosen in this study. Antioxidants known to activate defensive genes would be better suited to test an indirect effect.

In the brain, glutathione is thought to play a central role in defense against reactive oxygen species. It is synthesized in the brain, where it is maintained at millimolar levels. Glutathione can directly detoxify reactive oxygen species and can act as a substrate for several peroxidases. During scavenging of free radicals, glutathione disulfide is produced and GSH is reduced. In conditions of overproduction of free radicals or a deficiency of antioxidant systems, GSH is consumed and levels will fall. Decreases in GSH levels have been previously measured 24 hours after pilocarpine-induced seizures [37] and these findings are confirmed in the present study. Glutathione is found in the cytoplasm and in mitochondria and has been found to be released into the extracellular space by astrocytes [38]. Thus, glutathione should be able to neutralize free radicals in both the intra- and extracellular spaces. In this study, levels of the antioxidant form of glutathione (GSH) in the brain were determined and found to not change until long after the seizures have been initiated. This suggests that seizure onset is not related to depletion of antioxidant substances, such as glutathione, in the brain.

Initial testing in the pilocarpine model produced a significant reduction in seizure severity and an increase in seizure latency with vitamin E, vitamin C, melatonin and α-lipoic acid. However, these same antioxidants had no effect in the PTZ and kainic acid models suggesting that the effects do not generalize to other models. As reviewed in the information about the antioxidants in the methods sections, this inconsistency across experimental models has been noted before. The lack of effectiveness in the PTZ model could be due to the relatively rapid onset of the seizures in this model. However, the seizures induced by kainic acid develop more slowly than those induced by pilocarpine and there was no significant effect of any of the antioxidants in the kainic acid model. Thus, the antioxidant effect on seizure onset could be specific for the mechanism behind the seizures induced by pilocarpine. Pilocarpine is a potent muscarinic agonist [39], but the exact mechanism(s) underlying seizure onset in this model is not known. The present study would suggest that early after administration of pilocarpine there is generation of free radicals which contribute somewhat to the onset of the seizures.

Acknowledgments