Experimental febrile seizures increase dendritic complexity of newborn dentate granule cells

Objective: Febrile seizures (FS) are fever-associated convulsions, being the most common seizure disorder in early childhood. A subgroup of these children later develops epilepsy characterized by a hyperexcitable neuronal network in the hippocampus. Hippocampal excitability is regulated by the hippocampal dentate gyrus (DG) where postnatal neurogenesis occurs. Experimental FS increase the survival of newborn hippocampal dentate granule cells (DGCs), yet the significance of this neuronal subpopulation to the hippocampal network remains unclear. In the current study, we characterized the temporal maturation and structural integration of these post-FS born DGCs in the DG. Methods: Experimental FS were induced in 10-day-old rat pups. The next day, retroviral particles coding for enhanced green fluorescent protein (eGFP) were stereotactically injected in the DG to label newborn cells. Histochemical analyses of eGFP expressing DGCs were performed one, 4, and 8 weeks later and consisted of the following: (1) colocalization with neurodevelopmental markers doublecortin, calretinin, and the mature neuronal marker NeuN; (2) quantification of dendritic complexity; and (3) quantification of spine density and morphology. Results: At neither time point were neurodevelopmental markers differently expressed between FS animals and normothermia (NT) controls. One week after treatment, DGCs from FS animals showed dendrites that were 66% longer than those from NT controls. At 4 and 8 weeks, Sholl analysis of the outer 83% of the molecular layer showed 20-25% more intersections in FS animals than in NT controls (p <0.01). Although overall spine density was not affected, an increase in mushroom-type spines was observed after 8 weeks. Significance: Experimental FS increase dendritic complexity and the number of mushroom- type spines in post-FS born DGCs, demonstrating a more mature phenotype and suggesting increased incoming excitatory information. The consequences of this hyperconnectivity to signal processing in the DG and the output of the hippocampus remain to be studied.