Abstracts

Rationale: The ketogenic diet (KD) is a highly effective anticonvulsant treatment for pediatric refractory epilepsies. Efforts to identify the mechanism of action of the KD have focused on biochemical constituents such as ketone bodies [acetone, acetoacetate (ACA), beta-hydroxybutyrate (BHB)] and long chain omega-3 and omega-6 polyunsaturated fatty acids (LcPUFAs) [docosahexaenoic acid (DHA), eicosapentaenoic acid (EPA) and arachidonic acid (AA)]. Although, acute effects of ketone bodies on KATP channels, vesicular glutamate transporters and mitochondria have been elucidated, it is unclear whether ketone bodies have a significant effect on pathologic oscillatory activity of a hyperexcitable network. Similarly, LcPUFAs are known to inhibit sodium channels and L-type calcium channels, but whether this effect translates into dampening network activity is not known. We have found that hippocampal slices dissected from epileptic Kcna1-null mice display sharp wave-high frequency oscillation (SPW-HFO) complexes that are significantly different from wild-type mice. Kcna1-null SPW-HFOs occur more frequently, have longer duration and have spectral disorganization with the emergence of pathologic fast ripples. KD-treatment reduces seizures in Kcna1-null mice; thus, we hypothesized that KD-treatment would reduce the pathologic abnormalities of in vitro Kcna1-null SPW-HFOs. We further hypothesized that acute application of ketone bodies and/or LcPUFAs would have effects on Kcna1-null network oscillations. Methods: Kcna1-null mice were weaned (P21) onto either a standard diet or KD for 10-14 days. Recordings of spontaneous and evoked extracellular field potentials were acquired using a 64 multi-electrode array. Acutely isolated hippocampal slices from wild-type (WT) and Kcna1-null mice were positioned with electrodes in the fields of CA1, CA3 and dentate gyrus. Electrodes underlying axon fibers of the perforant path, mossy fibers and Schaffer-collaterals were used for stimulation. Results: The Kcna1-null mossy fiber-CA3 synapse exhibited decreased paired pulse ratios (PPRs) compared to WT suggesting an increase in neurotransmitter release. Acute treatment of ketone bodies or LcPUFAs did not have effects on evoked field potentials or PPRs. Slices isolated from KD-treated Kcna1-null mice had significantly increased PPRs suggesting reduced neurotransmitter release. Application of either ketone bodies or LcPUFAs reduced the incidence of Kcna1-null SPW-HFOs by 20-40%, but did not affect other parameters or fast ripples. KD-treatment normalized all Kcna1-null SPW-HFO parameters to near WT values and abolished fast ripple emergence. Conclusions: These data indicate that the KD alleviates pathologic network oscillations associated with epilepsy. This finding supports the notion that in vitro Kcna1-null SPW-HFOs are a useful measurement of anticonvulsant activity. In contrast, ketone bodies and LcPUFAs had limited effects on pathologic network activity suggesting that determining their importance in the mechanism of the KD may require methods of chronic exposure.