Abstracts

Rationale: A variety of brain insults have the potential to induce the development of epilepsy, particularly temporal lobe epilepsy (TLE) in humans and rodent models of TLE. The mechanisms of this process, named epileptogenesis, are still poorly understood. Accumulating evidence suggests that expression changes of the cation-chloride-cotransporters KCC2 and NKCC1 lead to a GABA-shift from an inhibitory to an excitatory action. Several studies indicate that NKCC1 is upregulated in models of TLE, and that the loop diuretic bumetanide seems to be a useful tool to counteract this upregulation. Furthermore, as our group described recently, co-administration with the GABA-agonist phenobarbital leads to disease-modifying effects in the pilocarpine model of TLE. The problem of bumetanide's rapid metabolization in rats and the fact that it only hardly crosses the blood-brain-barrier prompted us to investigate several strategies to enhance brain penetration, as well as to prolong the half-life. The aim of this study was to evaluate bumetanide's antiepileptogenic and anticonvulsive properties in rats after intracerebral or systemic administration, either alone or in combination with other agents in the kindling model of TLE. Methods: Rats were kindled by once daily electrical stimulation of the basolateral amygdala. In fully kindled rats the afterdischarge threshold (ADT) was determined and repeated until all rats exhibited a reproducible ADT. Intravenous (i.v.) administration of bumetanide and, in order to prevent rapid metabolization, intraperitoneal pretreatment with the monooxygenase-inhibitor piperonyl butoxide were chosen to investigate antiepileptogenic properties on kindling development. Furthermore, to determine anticonvulsant effects, fully kindled rats were treated by either intracerebroventricular or i.v. administration of bumetanide +/- the GABA-potentiating drug phenobarbital prior to ADT determination. Results: Bumetanide alone did not exert any antiepileptogenic or anticonvulsant effects, even when administered at high doses. However, when bumetanide was administered before a low dose of phenobarbital, which was ineffective alone, bumetanide significantly potentiated the anticonvulsant effect of phenobarbital in fully kindled rats. Conclusions: Although alterations in NKCC1 and GABA functions have been shown previously in the kindling model, neither systemic nor intracerebral injection of bumetanide exerted any significant effects on kindling development or fully kindled seizures. However, combined administration of bumetanide and phenobarbital exerted anticonvulsant effects in fully kindled rats. It may be suggested that NKCC1 inhibition was involved in the potentiation of phenobarbital by bumetanide. These results indicate that the use of bumetanide might be beneficial as add-on treatment together with GABAmimetic antiepileptic drugs in patients with TLE. Further kindling studies with prodrugs of bumetanide are currently under way to evaluate possibly stronger anticonvulsant properties of these derivatives.