This study aims to investigate whether valproic acid (VPA) anticonvulsant activity varied according to circadian dosing-time in mice.

VPA was administered to mice at four circadian stages (1, 7, 13 and 19h after light onset, (HALO)). The controls received a saline injection followed by a s.c. injection of pentylenetetrazol (PTZ) 30min later. In pretreated animals, VPA was administered 30min before PTZ administration.

The results obtained showed that VPA-induced anticonvulsant activity depends on circadian dosing-time in mice. VPA has significantly increased the latency of the first clonic seizure at all circadian stages. This increase varied depending on the time of VPA pre-treatment, the highest and the lowest increases were recorded respectively at 7 and 19 HALO. The Cosinor analysis has also validated a circadian rhythm of this increase. The intensity of seizures in pretreated mice varied significantly according to circadian stage. The highest seizure intensity was recorded at 19 HALO. A circadian rhythm of the seizure intensity in VPA pretreated mice was detected, with an acrophase located at the middle of the dark span (Ø=18.09 HALO±2.27h).

The present findings provide evidence for a pronounced anticonvulsant effect of VPA when administered in the 2nd middle of the light-rest span in mice. These results might probably be due to the circadian variation of VPA pharmacokinetic since our previous studies showed that the optimal tolerance corresponded to the middle of the rest span, the time which induces the highest total plasma clearance.