Abstracts

Rationale: Epileptic patients are subject to an elevated risk of premature mortality, and sudden unexpected death in epilepsy (SUDEP) is the most common cause of death in these patients. Clinical data indicate that generalized seizures and respiratory failure precede death in most witnessed cases of SUDEP. DBA/1 mice are a well-established animal model of SUDEP. DBA/1 mice exhibit generalized seizures, resulting in seizure-induced respiratory arrest (S-IRA), which leads to death unless resuscitation is rapidly instituted. Specific brainstem and limbic structures that modulate cardio-respiratory function are implicated in human SUDEP. These observations led us to use manganese enhanced magnetic resonance imaging (MEMRI) to investigate the neuronal activity changes in the brain that occur due to S-IRA in DBA/1 mice. Methods: DBA/1 and C57 (seizure non-susceptible) mice were exposed to an acoustic stimulus (bell stimulus, 96 dB SPL, for a mean of ~14 sec). DBA/1 mice exhibit audiogenic seizures (AGSz) followed by S-IRA, while C57 mice did not exhibit AGSz or S-IRA when exposed to the same stimulus. Ten sec after seizure cessation the DBA/1 mice were rapidly anesthetized and the brains were removed. The brains of the C57 mice were removed after an equivalent time after the stimulus. Comparative T1 weighted MEMRI images were evaluated with a 14T MRI and quantified using ImageJ software. Results: We observed significant (p < 0.05) increases in activity in DBA/1 mice as compared to C57 mice in several brain structures. These structures include the periaqueductal gray (PAG) (103.39 0.96 (SEM) vs. 99.1 0.89), amygdala (104.96 0.74 vs. 93.57 0.87), brainstem reticular formation (100.22 1.18 vs. 95 1.17), brainstem auditory structures (109.64 1.07 vs. 102.67 1.23), substantia nigra (SN) (104.24 1.13 vs. 97.07 1.08), raphe magnus (RM) (137.77 3.26 vs. 116.82 3.43) and medullary respiratory regions (MRR) (113.06 3.40 vs. 101.78 1.37). Conclusions: The increased activity in PAG and amygdala, resulting from seizures, in DBA/1 mice may perturb respiratory function, since these structures are known to modulate respiration. Peri-seizural declines in respiration are observed in many patients, and terminal apnea is the most common finding in observed cases of human SUDEP. Significant increases in the activity of the RM in DBA/1 mice suggest potential activation of serotonin-mediated compensatory respiratory mechanisms during seizures in a failed attempt to prevent S-IRA. Previous research has shown decreases in specific serotonin receptor expression in DBA/1 mice, which may explain the failure of such compensatory mechanisms. Thus, the current data suggest a crucial role of a brainstem-limbic network in DBA/1 mice, which have also been implicated in human SUDEP. Improved understanding of the causes and potential compensatory approaches for the altered function of these brain networks in animal models of SUDEP is crucial in development of treatments to prevent human SUDEP. Funding: Support: Epilepsy Foundation