Rostrocaudal and lateromedial density distributions of superior colliculus neurons projecting in the predorsal bundle and to the spinal cord: a retrograde HRP study in the cat

Abstract

Efferent neurons of the cat superior colliculus (SC) which project in the predorsal bundle (PDB) and to the spinal cord (PDB neurons) form a major pathway by which the SC controls the changes of the direction of gaze in response to stimuli of visual and other modalities. Knowledge of rostrocaudal and lateromedial density distributions of different groups of PDB neurons within the SC is necessary to analyse their relationships with the topography of sensory and motor maps. Density gradients may also bear on the efficacy of connections originating from topographically different collicular regions. In the present study, large injections of HRP/WGA-HRP were made in the C1 segment of the spinal cord and in the pontobulbar tegmentum. Judged by several morphological criteria, axons of passage, including those not subjected to a direct mechanical damage, were participating in the uptake of tracers. Therefore, labeled SC neurons corresponded to the nearly total populations of contralaterally projecting tectospinal neurons (TSNs) and neurons projecting in the PDB, respectively. Subtraction of the TSN density map from that of the whole PDB population was used to infer the distribution of tectal neurons terminating in the rhombencephalic tegmentum (TRhN). This subtotal labeling method proved useful in resolving the contradictions between the earlier HRP studies on the TSN and TRhN topography. The following density distributions were obtained for different groups of PDB neurons: 1) The mean TSN density is more than two times higher in the lateral half of the SC, representing the lower visual field. In this region the density remains constant from rostral to caudal, i.e., from the representation of vertical meridian to large contralateral azimuths. In the medial half, the average density decreases from rostral to caudal. Consequently, TSNs do not show the caudalward increment predicted by the higher efficacy of caudal stimulation points in eliciting head movements. 2) The distribution of PDB neurons is symmetrical with respect to the representation of the horizontal meridian. It is close to homogeneous at all azimuths of the retinotopic map and within the zone limited by small (10-15 degrees) upward and downward elevations.(ABSTRACT TRUNCATED AT 400 WORDS)