Abstracts

Rationale: Sudden unexplained death in epilepsy (SUDEP) is a devastating condition in which epilepsy patients die for no apparent reason with or without evidence of a recent seizure. Though this has been a long recognized syndrome, the pathophysiology of this disease is not well understood. Both cardiac and respiratory etiologies have been implicated to underlie SUDEP. Serotonin (5-HT) is a key regulator of breathing control, and has been implicated in the pathophysiology of SUDEP. Here we employ a mouse model in which nearly all 5-HT neurons have been genetically deleted to determine whether the absence of 5-HT neurons contributes to seizure-related respiratory dysfunction and death. Methods: EEG, EMG, EKG, body temperature, locomotor activity and breathing plethysmography were recorded in wildtype (WT) mice and mice lacking 5-HT neurons (Lmx1b^f/f/p) before, during and after seizure induction via either graded pilocarpine treatments (50 mg/kg i.p. every 20 min until category 5 seizures attained) or maximal electroshock (10-100 mA, 0.2-0.5 s, 60 Hz sine wave stimulation via ear electrodes). Results: Lmx1b^f/f/p mice experienced seizures after lower doses of pilocarpine compared to WT mice (50 mg/kg i.p. Lmx1b^f/f/p; 200 mg/kg i.p. WT). Lmx1b^f/f/p mice also experienced each Racine category of seizure at a shorter latency than WT. Invariably, Category 5 seizures progressed to motor status epilepticus (SE) in all animals of both genotypes. In the post-ictal period following prolonged seizures Lmx1b^f/f/p mice display profound reduction of respiratory rate and irregularity of respiratory rhythm. All Lmx1b^f/f/p mice and 50% of WT mice further progressed to death. Only WT mice in which SE resulted in death were included in this analysis. The latency to death from time of first injection was shorter for Lmx1b^f/f/p compared to WT. Seizures induced by MES were relatively short (15-30 s) in duration and were comprised of a characteristic tonic phase followed by a brief clonic phase and a subsequent period of reduced activity. Lmx1b^f/f/p mice experienced seizures at lower stimulus intensity (30 mA, 0.2 s, 60 Hz) than WT mice (50 mA, 0.5 s, 60 Hz). In both genotypes, the tonic phase was accompanied by respiratory arrest. Normal breathing spontaneously resumed at the onset of the clonic phase in WT mice. In the majority of Lmx1b^f/f/p mice, however, breathing did not spontaneously recover and the animals expired. Many Lmx1b^f/f/p mice did not exhibit clonic activity. Following respiratory arrest and attenuation of the EEG signal resulting from pharmacologically- or electrically-induced seizures, cardiac potentials could be recorded for up to 4 min. Conclusions: These results demonstrate that elimination of central 5-HT neurons renders mice more susceptible to seizure induction and more prone to seizure-related sudden death. These data further suggest that respiratory mechanisms may be more directly responsible for death than cardiac arrest and that a 5-HT neuron deficit causes post-ictal breathing abnormalities. These findings may have important implications for SUDEP.