Edaravone is an anti-stroke medication that may have nitric oxide (NO) modulating properties. This study evaluated the role of NO in the acute and sub-chronic anticonvulsant effects of edaravone in murine models of seizures induced by intraperitoneal (IP) or intravenous (IV) injections of pentylenetetrazole (PTZ) or electroshock (maximal electroshock seizure [MES]). 132 male albino mice were randomly divided into 22 groups (n=6) and given IP injections of vehicle or edaravone either acutely or for eight days (sub-chronically). The seizure was induced by electroshock or PTZ (IP or IV). The following edaravone doses were used: 7.5, 10, 12.5 (acute); 5, 7.5, 10 (sub-chronic) in IP PTZ model; 5, 7.5, 10 in IV PTZ model; and 5, 10 mg/Kg in the MES. To evaluate NO involvement, 216 mice were randomly divided into 36 groups (n=6) and pretreated with vehicle, edaravone, a non-specific nitric oxide synthase (NOS) inhibitor: N(ω)-nitro-L-arginine methyl ester (L-NAME) (5 mg/Kg), a specific nNOS inhibitor: 7-nitroindazole (7-NI) (60 mg/Kg), or a combination of edaravone plus L-NAME or 7-NI, either acutely or for eight days before seizure induction. Doses of edaravone were as follows: in IP PTZ model: 12.5 (acute) and 10 (sub-chronic); in IV PTZ model: 10; and in the MES: 5 mg/Kg. Data were analyzed using the one-way analysis of variance (ANOVA) followed by Tukey's test (SPSS 18). P≤0.05 was considered statistically significant. In the IP PTZ model, edaravone increased time latencies to seizures (P<0.001), prevented tonic seizures, and death. Edaravone increased the seizure threshold (P<0.001) in the IV PTZ model and shortened the duration of tonic hind-limb extension (THE) in the MES model (P<0.001). In comparison to mice treated with edaravone alone, adding L-NAME or 7-NI reduced seizure time latencies (P<0.001), reduced seizure threshold (P<0.001), and increased THE duration (P<0.001). Edaravone (acute or sub-chronic) could prevent seizures by modulating NO signaling pathways.