Cellular activity of intracerebrally transplanted fetal hippocampus during behavior

Hippocampal tissue derived from 12-, 20-, 25- and 34-mm rat fetuses was placed in a cavity formed by unilateral aspiration of the fimbria-fornix and the overlying neocortical tissue in adult rats. From 4 to 6 months after transplantation the rats were equipped with chronic recording and stimulating electrodes. Single cell activity of the transplant was monitored during running in a wheel, drinking, and sleeping. Both complex-spike cells (n = 151) and single-spike cells (n = 80) were recorded from the graft. A portion of the neurons changed their firing rates and discharge patterns as a function of ongoing behavior. About half of the single-spike cells increased their firing frequency during running. Fifteen per cent of the single-spike cells fired rhythmically at about 8 Hz during running, and the paradoxical phase of sleep and the discharge pattern correlated with rhythmic slow activity (theta) recorded concurrently from the contralateral (intact) hippocampus. These patterns were most frequently obtained from grafts of 20- and 25-mm (16 to 18 embryonic days) fetuses. Graft neurons could be activated by stimulating the ipsilateral hippocampus or the ipsilateral perforant path, with latencies of 8-30 ms. The most common electrical pattern in grafts of all groups was the synchronous bursts of several neighboring cells and concurrent electroencephalogram sharp-waves. Sharp-waves occurred during all behaviors. Large amplitude, high-frequency electroencephalogram spindles (14-18 Hz and 30-50 Hz) and associated neuronal bursts were recorded in grafts of 12-, 20-, and 25-mm fetuses. Based on these findings we suggest that both subcortical afferents and host hippocampal afferents send axons to hippocampal grafts and form viable synaptic connections with a portion of the neurons in the graft. The frequently encountered population bursts are explained by assuming that excitatory collaterals in the graft are more potent in the graft than in the normal hippocampus, and/or GABAergic inhibition is less efficient in the graft.