Purpose: A number of previous ncurophysiological studies have indicated that the glutamatergic system is important in the induction of epileptiform activity and the dcvelopment of epileptogenesis. Clutamate transport is the primary mechanism of inactivation of syiiaptically released glutamate. GLAST is classified BS an astrocytic transporter and occurs in high concentrations in the ccrebcllum. The pathophysiologic rolc of GLAST in epilepsy is not known in detail. To investigate the role of thc astroglial glutamatc transporter GLAST in epileptogenesis, we compared amygdalu‐kindling and pentylenctetrazolc (PTZ) induced seizures in GLAST‐deficient mice (GLAST(‐/‐)) wild‐type mice (GLAST(+/+)), and maternal C57Black6/J mice (C57). Methods: Kindling: All mice were between 10–1 I weeks of age. Under diethyl ether anesthesia, a chronic tripolar electrode w taxically implanted into the left amygdala (CS7: n = 6, GLAST(+/+): n = 3, CLAST(‐I‐): n = 5). All rats underwent kindling stimulation once daily that coiisistcd of 2 second duration of 100 Hz monophasic square pulses at the intensity of thc afterdischarge (AD) threshold. The seizure responses of thc kindled mice were classified as follows: stage 0, no response; stage 1, mouth and facial movement; stage 2, forclimb clonus; stage 3, bilateral forclimb clonus and tail up. The numbers of stimulations and AD durations required eliciting stage I or stage 2 werc compared, and the mean AD duration, latency and acizurc duration or five consecutive stagc 3 scizures (generalized scizures) were compared in each type of mouse. PTZ: PTZ was injected i.p. at a dose of SO mg/kg (GLAST(+/+): n = 13, GLAST(‐/‐): n = I3), and each mousc was observed for 30 min after the injection. Seizure rcsponses to PTZ werc rated as follows: stage 0, no response; stage I, isolated twitches; stagc 2, tonic‐chic convulsion; stage 3, tonic extension or death. Results: Kindling: No significant differcnces wcre scen in the first responses of amygdala‐kindled GLAST knockout mice. Nor were any significant differences seen in kindling seizure development. The mean AD duration, latency, and scizure duration of 5 consccutive stage 3 seizures in GLAST(‐/‐), GLAST(+/+), and C57 were compared. No significant differences were seen in AD duration and latency, but the generalized seizure duration (stage 3) of the AM‐kindled scizures in the GLAST(‐/‐) was significantly (approximately 35%) longer than in CS7. PTZ: GLAST(‐/‐) showed more severe scizurc stages of PTZ induced generalized seizures than GLAST(+/+) (p < 0.02, MannWhitney rank sum test), and seizure latency was significantly shortcr in the mutant mice (p < 0.005, unpaired t‐test). Conclusions: The results of present study clearly demonstrate that GLAST has different roles according to the type of epilepsy. In amygdala‐kindling, a limbic partial epilepsy model with secondary generalization, GLAST seems to have little role in seizure induction or the development of epileptogenesis in the limbic focus. Secondarily generalized seizures, on the other hand, may be modulated by GLAST, since the duration of bilateral forelimb clonus was found to be significantly longer in GLAST knockout mice. Several previous experiments have suggested significant increases in GLAST mRNA expression after amygdala‐kindled seizures or systemic administration of kainic acid. In contrast to kindled seizures, GLAST is closely related to the neuronal mcchanism underling seizure induction and severity in PTZ‐induced seizures, a primary generalized epilepsy model. GLAST may participate in the modulation of generalized tonicchic seizure activity in the cerebral cortex, because GLAST is widely localized i n the cerebral cortex, and the basic mechanism underling PTZ‐induced seizures is rclated to thalamocortical circuitry. Although no spontaneous seizures or EEG paroxysmal discharges were observed in GLAST knockout mice, it is concluded that GLAST is involved in the molecular mechanism of genetic idiopathic generalized epilepsy.
Read full abstract