Discharge properties of Purkinje cells in the oculomotor vermis during visually guided saccades in the macaque monkey.

1. We previously described discharge properties of cerebellar output cells in the fastigial nucleus during ipsilateral and contralateral saccades. Fastigial cells exhibited unique responses depending on the direction of saccades and were involved in execution of accurate targeting saccades. Purkinje cells in the oculomotor vermis (lobules VIc and VII) are thought to modulate these discharges of fastigial cells. In this study we reexamine discharge properties of Purkinje cells on the basis of this hypothesis. 2. Initially we physiologically identified the right and left sides of the oculomotor vermis. Saccade-related discharges of 79 Purkinje cells were recorded from both sides of the vermis during visually guided saccades toward the sides ipsilateral and contralateral to the recording side in two trained macaque monkeys. To clarify the correlation of Purkinje cell discharge with burst activities in the fastigial nucleus during saccadic eye movements, we analyzed our data by employing methods used in the study of fastigial neurons. 3. Among the 79 cells, 56 (71%) showed burst discharges during saccades (saccadic burst cells). Of the 56 cells, 29 exhibited a peak of burst discharges in both the contralateral and ipsilateral directions (bidirectional cells). The remaining 27 saccadic burst cells showed a peak of burst discharges during either contralateral or ipsilateral saccades (unidirectional cells). Among the 79 cells, 14 (18%) exhibited a pause of discharges during contralateral saccades (pause cells). Among the 79 cells, 9 (11%) showed burst discharge during contralateral saccades followed by tonic discharge that was correlated with eye position (burst tonic cells). 4. The timing of bursts in bidirectional cells with respect to saccade onset was dependent on the direction of saccade. During ipsilateral saccades, Purkinje cells exhibited a long lead burst that built up gradually, peaked near the onset of the saccade, and terminated sharply near midsaccade. The mean lead time relative to saccade onset was 29.3 +/- 24.5 (SD) ms. During contralateral saccades, Purkinje cells exhibited a short lead/late burst that built up sharply, peaked near midsaccade, and terminated gradually after the end of the saccade. The mean lead time relative to saccade onset was 10.7 +/- 20.8 ms. The burst onset time during contralateral saccades and the burst offset time during ipsilateral saccades preceded the saccade offset time by about the same interval regardless of the saccade amplitude. 5. In pause cells the pause preceded saccade onset by 17.5 +/- 10.6 ms. The duration of the pause was not correlated with the duration of saccades. There was little trial-to-trial variability in the onset time of the pause with respect to the onset of saccades, whereas there was large trial-to-trial variability in the offset time of the pause with respect to the offset of saccades. In addition, the mean onset time of the pause for each cell had a relatively narrow distribution. 6. The burst lead time of burst tonic cells relative to saccade onset was 9.5 +/- 3.9 ms. The tonic discharge rate of burst tonic cells was a nonlinear function of eye position. The regression of each cell was fit to two lines. The regression coefficient ranged from 0.95 to 0.99 (mean = 0.97). 7. Axons of Purkinje cells in the oculomotor vermis are thought to project exclusively to saccadic burst cells in the fastigial oculomotor region (FOR), which is located in the caudal portion of the fastigial nucleus. Our previous studies indicated that FOR cells provide temporal signals for controlling targeting saccades. The present results suggest that Purkinje cells in the oculomotor vermis modify the temporal signals of FOR cells for saccades in different directions and amplitudes. The modification of FOR cell activity by Purkinje cells is thought to be essential for the function of the cerebellum in the control of saccadic eye movements.