Abstracts

Rationale: Neuronal activity is known to play an important role in dendrite development. But the impact abnormal seizure activity has on dendrite growth and maturation has rarely been studied. Previous experiments in our laboratory have shown that when hippocampal slice cultures, taken from mice 4-6 days-old, are grown in media containing bicuculline (100 M) for 1 week, dendritic growth is markedly suppressed. We have also shown that recurrent seizures suppress the molecular maturation of hippocampal glutamatergic synapses in vivo using the flurothyl model. These results motivated us to study how recurrent seizures affect dendritic growth in developmental CA1 hippocampal pyramidal cells in vivo. Methods: Experiments were performed using Thy1-GFP-M mice obtained from Jackson Laboratory. On postnatal days (P) 7-11, brief (3 min) flurothyl-induced seizures were produced in infant mice. Three seizures were induced daily for 5 days. Sham littermate control mice were handled identically but without exposure of flurothyl. Following treatment, mice were allowed to survive for varying periods of time (1-20 days) before brains were removed for analysis. On P12, 15, 20, 25, 30, brains were fixed and 300 m slices were made. CA1 hippocampal pyramidal cells were confocally imaged and the basilar dendrites reconstructed using Neurolucida. Results: Under control conditions, the basilar dendrites of CA1 pyramidal neurons grow rapidly during the first 2 postnatal weeks. For instance, between P6 and P15 the average length of the basilar dendrite nearly doubles from 1354 85 m to 2548 133 m (SEM). Thereafter, the dendrites continue to grow but at a reduced rate until P25-30. Sholl analysis indicates that as the hippocampal pyramidal cells mature they add branches both near (< 50 m radial distance) and at a distance (between 50 and 250 m) from their somas. In mice that experienced recurrent seizures between P7 and 11, the total length of basilar dendrites is reduced 15.8 % (2494 147 versus 2100 124 m, P<0.05) at P25 and the growth is dramatically arrested from P12 (1958 155 m) to P30 (2058 120 m). Comparison of Sholl analyses of seizure-treated samples and their controls indicate that shorter dendrites predominate early-in-life (P12- P20) in experimental samples but with further maturation fewer dendrites of all lengths are observed. Conclusions: These results suggest that the interruption of synaptic maturation and decreases in molecular markers for glutamatergic synapses seen previously in the flurothyl model may be at least in part be due to a reduction in dendritic arborization. Since dendrites undergo rapid growth during early life, recurrent seizure in vivo may retard or even arrest dendritic growth. Our results suggest that developing neural networks employ unique compensatory mechanisms to control chronic network hyperexcitability. Neuronal network abnormalities produced by the morphological changes reported here could be an explanation for the learning and memory deficits observed in numerous models of early-onset epilepsy.