Rationale:
Antiseizure medication (ASM) use during pregnancy is associated with drug and dose-dependent teratogenic risks across both anatomical and neurobehavioral domains. The latter category includes IQ reductions (verbal and nonverbal) that occur transdiagnostically, as well as a higher incidence of attention deficit and autism spectrum disorders. The underlying molecular mechanisms remain poorly understood. In this study, we apply home-cage monitoring to objectively appraise the neurobehavioral consequences of prenatal valproic acid (VPA) exposure in laboratory mice. In addition, we explore associated metabolomic changes in hippocampal samples from adult offspring.
Methods:
All study protocols were approved by the BCM IACUC. Seven to eight week old C57BL/6J mice were assembled as breeder pairs and administered VPA in minimally sweetened drinking water designed to achieve a target daily dose of ~500-600mg/kg/d. Control pairs received vehicle solution alone. At birth, all bottles were replaced with standard drinking water. ~Sixty day old offspring were examined in instrumented home-cage chambers) to measure various aspects of task-free spontaneous behavior (horizontal activity, feeding/drinking patterns, sleep) as well as responses to standardized in-cage provocative maneuvers. Flash-frozen whole hippocampi from a separate cohort of mice were provided to the BCM Metabolomics Core to measure a set of targeted metabolites through liquid chromatography-mass spectrometry.
Results:
We conducted a within-lab meta-analysis of CTRL (n=106) vs prenatal VPA exposed (n = 125) offspring studied since 2019. Across a range of homecage-derived scalar variables, we identified two main behavioral phenotypes. VPA exposure was associated with increased total sleep time (~30 mins/day), an effect driven exclusively by male offspring. Similarly, VPA exposure also resulted in increased wheel-running, which was restricted to female offspring. We did not identify any overt seizures, survival deficits or outwardly obvious structural deformities. In LC-MS analysis, 136 metabolites were successfully assayed. Averaging across both sexes (n=8/group) and after correcting for multiple comparisons, prenatal VPA exposed hippocampi displayed significantly elevated levels of carnitine, lactate, fumarate, succinate, glutamate and aspartate, suggesting increased energy consumption and ATP production. Acetylcholine and taurine were also significantly elevated. Fold changes were larger in female mice, which also uniquely displayed significantly elevated glutamate/GABA ratios.
Conclusions:
In a mouse model of pangestational VPA exposure, we find distinct neurodevelopmental trajectories in male and female offspring in conjunction with distinct hippocampal neurometabolomic profiles. These endpoints may serve as predictive preclinical safety biomarkers to assess new and existing ASMs. In ongoing studies, we apply these results as a starting point to decipher the long-term epigenomic signatures imparted by early VPA exposure.
Funding:
This work is supported by an AES Junior Investigator Award, NINDS K08NS110924, NINDS R01NS13199 and seed funding from the BCM Office of Research.