Abstracts

Rationale: Allopregnanolone (3α-hydroxy-5α-pregnan-20-one; 5α,3α-P), an endogenous neuroactive steroid, is a positive allosteric modulator of synaptic and extrasynaptic GABA-A receptors. 5α,3α-P has antiseizure activity in various animal seizure models. However, 5α,3α-P is not orally bioavailable and has a short duration of action when administered by the usual parenteral routes (e.g., intravenous or intramuscular). These factors limit the usefulness 5α,3α-P in the treatment of seizures and epilepsy. The objective of the present study was to determine if administration of 5α,3α-P by the intranasal (IN) route could overcome these limitations. Methods: Antiseizure activity was assessed in the mouse 6 Hz limbic seizure test, the mouse maximal pentylenetetrazol (PTZ) seizure test, and the mouse timed intravenous PTZ threshold test. 5α,3α-P was dissolved at a concentration of 16 mg/ml in 0.9% saline with 40% sulfobutylether-β-cyclodextrin sodium salt. The 5α,3α-P solution was administered IN at various times prior to administration of the electrical stimulus in the 6 Hz test, the injection of PTZ (80 mg/kg, IP) in the maximal PTZ test, or the initiation of the PTZ infusion in the PTZ threshold test. The times to onset of myoclonic body twitches and clonic and tonic seizures were recorded. 5α,3α-P was considered to have antiseizure activity if it delayed the onset of seizure signs when compared with the time of their occurrence in vehicle-treated animals. Results: In the 6 Hz test, IN 5α,3α-P at the dose of 16 mg/kg conferred prolonged seizure protection (up to 8 h). In the maximal PTZ test 5α,3α-P administered IN delayed the time to onset of all seizure signs with a marked effect on tonic hindlimb extension and reduction in mortality. Significant protection in the maximal and threshold PTZ tests was observed at early time points and was markedly diminished at 1h post IN application. Conclusions: We hypothesize that the prolonged duration of 5α,3α-P antiseizure activity in the 6 Hz test may be due to direct nose-to-brain delivery, which conveys 5α,3α-P to temporal lobe structures relevant to seizure generation in this model. In contrast, protection in the maximal and threshold PTZ tests was less prolonged, possibly because direct nose-to-brain delivery is less relevant than hematogenous delivery to hindbrain structures in these models. It is becoming increasingly recognized that substances deposited in the nasal passages may track along the olfactory nerves into forebrain structures. Our results suggest that administration via this route may extend the duration of action because the nasal structures serve as a depot that prolongs delivery. We conclude that IN delivery of 5α,3α-P may not only provide a practical noninvasive approach to overcoming the lack of oral bioavailability, but unexpectedly confers extended seizure protection that would make chronic dosing feasible. Funding: NINDS grant #1U54NS079202: DZ, CW, TF, MAR