Abstracts

Rationale: Complement activation leading to inflammation appears to play a role in epilepsy and seizure development. Levels of various complement factors, including the membrane attack complex, are increased in both animal models of epilepsy and patients with human temporal epilepsy (Aronica et al., 2007; Neurobiol Dis 26:497ff), suggesting that the anaphylatoxin C5a is produced in the epileptic brain. We hypothesized that activation of receptors for C5a, specifically CD88, promote seizure generation. Methods: The anticonvulsant profile of the cyclic anti-inflammatory peptide PMX53, an inhibitor of CD88, was investigated in several murine seizure models. PMX53 levels were quantified in plasma and saline perfused brains at 15, 30, 60, and 90 min using liquid chromatography coupled to mass spectroscopy (LC-MS/MS). Results: PMX53 displayed reproducible anticonvulsant effects in both the 6Hz and chronic corneal kindling models, but not in the pentylenetetrazole (i.v.) or maximal electroshock threshold tests. Statistically significant elevations in CC50 value (mean +/- SEM: 8.1 +/-1.2 mA n= 5 independent experiments) were seen in CD1 mice when PMX53 was administered 30 minutes prior to 6 Hz seizure induction at a dose of 3mg/kg (s.c.). PMX53 (3 mg/kg s.c.) also protected fully corneally kindled CD1 mice against stage 3-5 seizures (p=0.04 and p=0.027 n=2 experiments). The anticonvulsant effects in the 6 Hz model disappeared by 60-90 minutes after administration. Our pharmacokinetic study using quantitative LC-MS analysis revealed peak levels in plasma and brain within 30 minutes of PMX53 (3mg/kg s.c.) administration. Peak plasma levels were 482+/-57.5nM (mean +/- SEM, n=6) at 15 minutes post administration, with a rapid decline in plasma levels following this timepoint. Brain levels were 11.5+/-1.9 ng/g and 15.6+/-4.1 ng/g (mean +/- SEM, n=6) at 15 and 30 minutes, respectively, post administration indicating a 40-60 nM level in the extracellular fluid. These levels in the extracellular fluid are close to the IC50 of PMX53 on human CD88 of approximately 20nM (Woodruff et al., 2005 J Pharmacol Exp Ther 314:811-7). We are currently assessing to which extent PMX53's effects in the 6 Hz model are dependent on CD88 using CD88-deficient and wild type mice. Conclusions: PMX53 is anticonvulsant in the murine 6 Hz and corneal kindling models. The anticonvulsant time window of PMX53 matches with its brain concentration peaks early after administration. In conclusion we show blood brain barrier permeability and reproducible anticonvulsant effects of a cyclic anti-inflammatory peptide against electrically induced seizures in mice. Elucidation of its CNS target and development of more stable analogues appear to be a promising approach to find new treatments for pharmacoresistant epilepsy.