Abstracts

Rationale: Multiple lines of clinical and experimental evidence suggest that seizures in early life can be associated with long lasting cognitive and behavioral deficits. Neurodevelopmental disorders that give rise to autism and intellectual disability often include early life seizures (ELS). There are no current treatments or preventions for the evolution of synaptic impairment. The developing postnatal brain undergoes a dynamic period of synaptogenesis during which we have shown ELS causes rapid changes to excitatory glutamate receptors in developing neuronal networks. Despite the evidence for ELS being associated with long lasting behavioral changes and impaired plasticity at a network level, far less is known about the evolution of changes at a single neuron level that occur following a seizure through the lifespan.

Methods: Using activity-dependent genetic labeling (TRAP, which relies upon the detection of Cre-induced permanent tdTomato expression, which is driven by the immediate early gene Fos, a widely used marker of activated neurons), we identified and characterized a specific subgroup of neurons activated by ELS in the hippocampus.

Results: The mice with tonic-clonic seizure induced by postnatal day 10 (P10) intraperitoneal injection of chemoconvulsant kainate (KA) showed a robust neuronal activation in CA1 region. The TRAPed cells in CA1 included mainly glutamatergic neurons (93.517± 8.9, p < 0.0001 vs. saline) with little astrocytes (6.9% ± 2.1%). Whole-cell patch clamp recording revealed that the ELS-activated CA1 neurons (tdtomato+, td+) during early development have persistent increases of postsynaptic AMPAR function (spontaneous EPSC amplitude at P30 38.6 ± 8.9, p < 0.05 vs. saline and Td- cells, Kruskal-Wallis test) and rapid diminishment of NMDAR-only silent synapses (fraction of silent synapse at P15 of td+ cells was 25.5%, much less than Td- 63.8% and saline 73.8%; the fraction of silent synapse were comparable among groups at P30, p > 0.05, one-way ANOVA), a process that involved synaptic insertion of calcium-permeable AMPARs (CP-AMPARs) (rectification index at P30 td+: 3.58 vs td-: 1.84 and saline: 1.64, p < 0.05, one-way ANOVA). In vivo pharmacological treatment following the ELS with a selective GluA2-lacking receptor blocker specifically reversed the long lasting deficits in the subsets of neurons initially recruited by the ELS (rectification index at P30 IEM: 1.31 vs saline 2.34, p < 0.01, unpaired t test).