Abstracts

Rationale: Hypoxia induced neonatal seizures are often associated with later development of long-term epilepsy, the mechanisms for which are unknown. These seizures are often refractory to conventional GABAergic drugs, thought to be in part due to a developmental overexpression of the chloride (Cl-) co-transporter NKCC1. This overexpression leads to intracellular Cl- accumulation, resulting in a paradoxical Cl- efflux and depolarization in response to GABA channel activation. As in humans, rodent neonatal brain also expresses higher than adult levels of NKCC1 (Dzhala, et al, Nat Med 2005). Bumetanide (BMX), a NKCC1 inhibitor, has been shown to suppress neonatal seizures in the rodent (Dzhala, et al, Ann Neural 2008). We previously reported that 0.15 mg/kg BMX in combination with 15 mg/kg phenobarbital (PB) effectively suppresses hypoxia-induced seizures in postnatal day (P)10 rats. In the present study, we assessed serum and brain concentrations of BMX at the anticonvulsant doses in this seizure model as well as in na ve control litter mates. Methods: Postnatal day (P)10 rats were pretreated with PB (15 mg/kg) and BMX (0.15 or 0.30 mg/kg), then killed 30 min after graded global hypoxia. Serum and homogenized brain tissue were treated with acetonitrile to precipitate protein. After centrifugation, supernatant was analyzed by LC-MS/MS for BMX quantification (deuterated BMX as the internal standard). In addition, we calculated the ratio of BMX in brain tissue to that in serum. Results: There was a dose dependent increase in both serum and brain BMX concentrations between the 0.15 and the 0.30 doses (ng/g brain mean: 0.15 dose, 3.50 2.29; 0.30 dose, 4.95 2.24; n=8, p=0.07. ng/ml serum mean: 0.15 dose, 149.4 52.4; 0.30 dose, 294.6 109; n=8, p=0.003). Overall, the mean ratio of brain:serum levels was 0.024 0.015 for the 0.15 dose and 0.018 0.002 for the 0.30 dose (n=8). Interestingly, we found a trend for hypoxic animals having higher concentrations compared to normoxic BMX treated littermate controls (0.15 dose mean: hypoxic, 5.00 2.24; control, 2.00 0.60; n=4, p=0.05. 0.30 dose mean: hypoxic, 6.35 2.49; control, 3.55 0.52; n=4, p=0.07). Conclusions: Taken together, these data show that anticonvulsant doses of BMX do result in elevated brain and serum levels, suggesting that BMX crosses the blood brain barrier. The trend for higher brain concentrations in animals with seizures suggests that BMX permeability is affected by ictal activity, perhaps due to a break down in the BBB. Finally, these preclinical data help validate the LC-MS/MS for clinical use, as this assay will be used in a clinical pilot study of BMX for neonatal seizures (http://www.clinicaltrials.gov). Supported by P30 HD 18655, 1RC1NS068938-01, the Charles H. Hood Foundation, and the Manton Center.