Abstracts

Rationale: The ketogenic diet is a high-fat, low-carbohydrate regimen that reduces seizures through as yet ill-defined mechanisms. This metabolic therapy can be effective in medically refractory epilepsy, suggesting that a ketogenic diet achieves its anticonvulsant effects via mechanisms other than those targeted by traditional antiepileptic drugs. Adenosine is an endogenous anticonvulsant and a key link between metabolism and neuronal activity. Similar to a ketogenic diet, adenosine can stop pharmacoresistant seizures. We hypothesized that adenosine acting via inhibitory adenosine A1 receptors (A1Rs) plays a key role in the anticonvulsant effects of a ketogenic diet. Methods: To test this hypothesis, we used three types of transgenic mice, all with spontaneous electrographic seizures that are due to decreased A1R signaling: (1) mice with a complete absence of A1Rs (A1R-/-); (2) mice with a significant reduction in A1Rs (A1R /-, a heterozygote with 50% of normal A1R levels); and (3) mice with a transgenic overexpression of adenosine kinase (Adk-tg), an intracellular enzyme that serves as a major negative regulator of extracellular levels of adenosine. Intrahippocampal electrodes recorded seizure frequency and duration after 3 weeks of maintenance on either a control or a ketogenic diet. To determine the extent to which changes in the frequency and duration of spontaneous seizures were due specifically to the low carbohydrate nature of the KD, we administered intraperitoneal injections of a 30% glucose solution. Results: In transgenic mice with spontaneous recurrent seizures due to an adenosine deficiency but with intact A1Rs (Adk-tg), a ketogenic diet nearly abolished spontaneous seizures and reduced their duration significantly. In contrast, spontaneous recurrent seizures triggered by reduction or deletion of A1Rs were partly or completely resistant to ketogenic diet therapy, respectively. In mice which displayed reduced seizures after KD treatment we found that an injection of glucose restored seizure frequency to normal levels within 30 to 90 minutes. Conclusions: These data suggest that a ketogenic diet reduces seizures by increasing A1R-mediated inhibition, and that this effect depends on the low carbohydrate nature of the diet. Revealing an A1R-dependent component provides new insight into the anticonvulsant mechanisms underlying the anticonvulsant effects of a ketogenic diet. Furthermore, these results suggest that ketogenic metabolism increases the activity of adenosine at the A1 receptor subtype, and could offer insight into therapies for other clinical conditions where adenosine is known to have clinical benefits.