Place cells recorded in the parasubiculum of freely moving rats.

Abstract

Previous studies have identified neurons in the hippocampus, subiculum, and entorhinal cortex which discharge as a function of the animal's location in the environment. In contrast, neurons in the postsubiculum and anterior thalamic nucleus discharge as a function of the animal's head direction in the horizontal plane, independent of its behavior and location in the environment. Because the parasubiculum (PaS) has extensive connections, either directly or indirectly, with these structures, it is centrally located to influence the neuronal activity in these areas. This study was therefore designed to determine the types of behavioral and spatial correlates in neurons from the PaS. Single unit recordings were conducted in the PaS of freely moving rats trained to retrieve food pellets thrown randomly into a cylindrical apparatus. A total of 10.3% of the cells were classified as place cells because they discharged in relation to the animal's location in the cylinder. A large percentage of cells (41.4%) were classified as theta cells. The remaining cells had nondiscernable behavioral correlates. Quantitative analysis of the firing rate maps for the place cells showed they had higher levels of background activity and contained larger firing fields than values reported previously for hippocampal place cells. Directional analysis showed that only three out of 16 cells contained a secondary directional correlate; the firing rate for the remaining cells was not affected by the animal's directional heading within the firing field. A time shift analysis, which shifted the spike time series relative to the animal location series, was conducted to determine whether the quality of the location-specific firing could be improved. The time shifts for three different spatial parameters were optimal when cell discharge led the animal's position. Furthermore, the optimal time shifts for two of these parameters (firing area and information content) were less than the optimal shift reported for hippocampal place cells and suggested that PaS cell discharge lagged behind hippocampal place cell activity. Rotation of the cue card with the animal out of view led to near equal rotation of the firing field when the animal was returned to the apparatus. These results indicate that a small population of cells in the PaS encode the animal's location in its environment, although the representation of space encoded by these cells is different from the type of representation encoded by hippocampal place cells.