Abstracts

RATIONALE: Furosemide, a blocker of chloride co-transporters, exerts a powerful inhibitory effect on epileptiform activity. Furosemide has been shown to abolish bursting in the 4-AP and zero-magnesium hippocampal slice models in rats, mice and human tissue; kainic acid induced and audiogenic seizure models in vivo. There is preliminary evidence from population studies that the use of furosemide reduces the incidence of epilepsy in patients. The exact mechanism of action of furosemide has not been elucidated, but is presumed to be through non-synaptic mechanisms, which could include effects on pH, control of extracellular space volume, or regulation of extracellular potassium. Here we studied the ability of furosemide to eliminate epileptiform bursting in the presence of agents that block the inwardly rectifying potassium channels (K[sub]IR[/sub]) responsible for extracellular potassium buffering.
METHODS: Horizontal 400[mu]m thick brain sections were made from adult mice and stored in artificial CSF, then transferred to a gas-interface type recording chamber and continuously perfused with aCSF at 33-35[degree]C. Input-output curves and paired-pulse stimulation plots were obtained by stimulating Schaffer collaterals and recording in CA1. Comparisons were made before and after the addition of furosemide (2.5 mM). In separate experiments, slices were perfused with aCSF containing 4-aminopyridine (4-AP, 100 [mu]M) until epileptiform bursts occurred at regular intervals. In some experiments, the KIR channel blockers barium chloride (BaCl[sub]2[/sub], 100 [mu]M) or cesium chloride (CsCl[sub]2[/sub]; 1mM) were added to aCSF containing 4-AP, and then furosemide was added. In other experiments, bursting was initiated with 4-AP, then furosemide was added until bursting ceased (typically 30-45 minutes) and either barium or cesium were subsequently added. Recordings were continued for at least 60 min following addition of the final solution.
RESULTS: The addition of furosemide to the bathing medium had no effect on input-output function or on paired-pulse facilitation (5 slices). Analysis of a single-stimulus evoked CA1 population spike during baseline (10 min) and prolonged perfusion with furosemide (40 min) was also performed (3 slices). Furosemide exposure resulted in the development of a second population spike response (consistent with its blockade of GABA-A channels) but no significant change in spike amplitude. In seizure studies, 2.5 mM furosemide reliably abolished 4-AP induced bursting within 45 min, as expected. When perfused for 1 hr following slice pre-exposure to aCSF containing barium or cesium (30-40 min), furosemide consistently failed to block epileptiform bursting (8 slices). Likewise, the addition of barium/cesium to a combination of furosemide + 4-AP aCSF resulted in a return of epileptiform bursts within 30 min (4 slices).
CONCLUSIONS: Our results demonstrate that furosemide does not produce an antiepileptic action when K[sub]IR[/sub] channels are blocked. Furosemide does not inhibit input/output function or paired-pulse facilitation, suggesting that it does not produce a significant reduction of synaptic activity. Therefore, we propose that further investigation into the role of K[sub]IR[/sub] channel modulation as an anti-epileptic mechanism for furosemide is warranted, and may ultimately lead to the development of novel therapeutic treatments.
[Supported by: University of California at San Francisco Academic Senate Funds]