The goal of this study was to identify GABAergic input to the cat superior colliculus from neurons located in the caudal diencephalon, mesencephalon, pons and medulla. Cells efferent to the superior colliculus were labeled retrogradely with the tracer horseradish peroxidase, and an antibody to gamma-aminobutyric acid was used to label GABAergic neurons in the same sections. The results indicate that neurons in several distinct areas of the caudal diencephalon and brainstem are both immunocytochemically labeled for GABA and retrogradely labeled with horseradish peroxidase. These areas include zona incerta, nucleus of the posterior commissure, anterior and posterior pretectal nuclei, nucleus of the optic tract, superior colliculus, cuneiform nucleus, subcuneiform area, substantia nigra pars reticulata and pars lateralis, periparabigeminal area, external nucleus of the inferior colliculus, the area ventral to the external nucleus of the inferior colliculus, mesencephalic reticular formation, dorsal and ventral nuclei of the lateral lemniscus, and the perihypoglossal nucleus. The role that such diverse inhibitory input to the superior colliculus might play, particularly in influencing eye movements, is discussed.

The neuroanatomical tracer Phaseolus vulgaris leucoagglutinin (PHA-L) was used to label the local projections of neurons whose cell bodies are located in the superficial layers of the cat superior colliculus. Small, localized groups of neurons in the superficial layers project to all regions of the ipsilateral colliculus, including the deep layers. Comparable distributions of labeled terminals are seen throughout the colliculus when the PHA-L injection site is located in the rostral, middle, or caudal one-third of the colliculus. In addition, there is some evidence of a topographic projection from superficial to deep layers. These results suggest that a complex anatomical substrate exists for communication between the superficial and deep layers of the colliculus. Connections such as these may underlie the transfer of visual information from neurons in the superficial layers to populations of neurons in the deep layers that respond prior to saccadic eye movements.

Immunocytochemical techniques have been used to examine the distribution of substance-P (SP)-labeled neurons in the superior colliculus of rats from birth to adulthood. At birth, there are almost no SP-immunopositive neurons in the tectum. A small number of SP neurons appear over the next several days. However, the vast majority of SP neurons appear between P9 and P10, and by P12 have attained adult-like numbers and distribution. Neurons are confined to the superficial layers of the colliculus, specifically the upper two-thirds of the stratum griseum superficiale (SGS). There is no indication of a differential developmental sequence along rostrocaudal or mediolateral axes. Neuronal types can be distinguished as early as P6 and include horizontal, vertical, and multipolar cells.Substance-P-immunoreactive axons and boutons are also present in the superior colliculus at birth, and are for the most part confined to the deep layers. Boutons are generally of the en-passant type. The density of labeled axons and boutons increases progressively, and by P10 there is an almost adult-like lamination and patchiness. In the adult, labeled axons and boutons are most dense in the stratum opticum and stratum griseum intermedium. Bridges of dorsoventrally oriented labeled axons span the relatively label-poor stratum album intermedium. SP label in the stratum griseum profundum is dense and patchy, and there is also dense label in the stratum album profundum bordering the periaqueductal grey.The role of substance-P-labeled neurons in the superior colliculus is still a matter of speculation. The findings of this study indicate that SP neurons may play a role in intrinsic collicular circuitry.

The ultrastructure of large neurons in the stratum griseum intermedium of the cat superior colliculus was examined following injections of horseradish peroxidase (HRP) into the dorsal tegmental decussation. Four HRP-labeled cells were selected, and the synaptology of their cell bodies and selected regions of proximal and distal dendrites was examined. The four neurons represent four morphologically distinct cell types: multipolar radiating, tufted, large vertical, and medium-sized trapezoid radiating. These four neurons correspond with cell types X1, X2, X3, and T1 respectively, according to the recent classification of neurons in the superior colliculus of the cat by Moschovakis and Karabelas (J. Comp Neurol. 239:276-308, '85). The three X type neurons are similar in having 83% of their somata and over 74% of their proximal dendrites contacted by synaptic profiles. Distal dendrites of the X type neurons, however, receive fewer synaptic contacts. In contrast, in the T1 cell, only 69% of the soma membrane is contacted by synaptic profiles, and the synaptic coverage on proximal and distal dendrites does not vary much from this. Of the eight types of synaptic terminals described in the stratum griseum intermedium of the cat superior colliculus by Norita (J. Comp. Neurol. 190:29-48, '80), only five are found in contact with the X and T type efferent neurons described here. There are some regional differences in terminal distribution, although each terminal is represented on each cell. Type III terminals (small, contain mostly pleomorphic vesicles, and make symmetrical contacts) are the most abundant on cell bodies and dendrites of all four cell types. Terminal types II (medium-sized, containing round and flattened vesicles, and making asymmetrical contacts), and IV (medium to large in size, containing flattened vesicles, and making symmetrical contacts) are well represented. In general, terminal types I (small, containing densely packed round vesicles, and making asymmetrical contacts) and VI (small and irregular in shape, containing flattened vesicles and making symmetrical contacts) are found infrequently. The identity of different types of synaptic terminal is discussed.