The projection from 11 auditory cortical areas onto the subdivisions of the inferior colliculus was studied in adult cats by using two different anterograde tracers to label cortico-collicular (CC) axon terminals. The main results were that: 1) a significant CC projection arose from every field; 2) the principal inferior collicular targets were the dorsal cortex, lateral nucleus, caudal cortex, and intercollicular tegmentum, with only a sparse projection to the central nucleus; 3) the input was usually bilateral, with the ipsilateral side by far the most heavily labeled, and the contralateral projection was a symmetrical subset of the ipsilateral input; 4) the CC system is both divergent and convergent, with single cortical areas projecting to six or more collicular subdivisions, and each auditory midbrain subdivision receiving a convergent projection from two to ten cortical areas; 5) cortical areas devoid of tonotopic organization have topographic projections to collicular target nuclei; 6) the heaviest CC projection terminated in the caudal half of the inferior colliculus; and finally, 7) the relative strength of the cortico-collicular labeling was far less than that of the corresponding corticothalamic projection in the same experiments. The CC system is strategically placed to influence both descending and ascending pathways arising in the inferior colliculus. Nuclei that participate in the premotor system, like the inferior collicular subdivisions that project to the pons, receive substantial corticofugal input. Both the dorsal (pericentral) and the lateral (external) nuclei of the inferior colliculus project to parts of the medial geniculate body whose closest auditory affiliations are with non-tonotopic cortical regions involved in higher order auditory perception. The cortico-collicular system may link brainstem and colliculo-thalamic circuits to coordinate premotor and perceptual aspects of hearing.

The distribution and morphology of neurons and axonal endings (puncta) immunostained with antibodies to gamma-aminobutyric acid (GABA) and glycine (Gly) were analyzed in auditory brainstem, thalamic, and cortical centers in the mustache bat. The goals of the study were (1) to compare and contrast the location of GABAergic and glycinergic neurons and puncta, (2) to determine whether nuclei containing immunoreactive neurons likewise have a similar concentration of puncta, (3) to assess the uniformity of immunostaining within a nucleus and to consider regional differences that were related to or independent of cytoarchitecture, and (4) to compare the patterns recognized in this bat with those in other mammals. There are nine major conclusions. (1) Glycinergic immunostaining is most pronounced in the hindbrain. (2) In the forebrain, GABA alone is present. (3) Some nuclei have GABAergic or glycinergic neurons exclusively; a few have neither. (4) Although there is sometimes a close relationship between the relative number of immunopositive neurons and the density of the puncta, just as often there is no particular correlation between them; this reflects the fact that many GABAergic and glycinergic neurons project beyond their nucleus of origin. (5) Even nuclei devoid of or with few GABAergic or glycinergic neurons contain relatively abundant numbers of puncta; some neurons receive axosomatic terminals of each type. (6) In a few nuclei there are physiological subregions with specific local patterns of immunostaining. (7) The patterns of immunostaining resemble those in other mammals; the principal exceptions are in nuclei that, in the bat, are hypertrophied (such as those of the lateral lemniscus) and in the medial geniculate body. (8) Cellular colocalization of GABA and Gly is specific to only a few nuclei. (9) GABA and glutamic acid decarboxylase (GAD) immunostaining have virtually identical distributions in each nucleus. Several implications follow. First, the arrangements of GABA and Gly in the central auditory system represent all possible patterns, ranging from mutually exclusive to overlapping within a nucleus to convergence of both types of synaptic endings on single neurons. Second, although both transmitters are present in the hindbrain, glycine appears to be dominant, and it is often associated with circuitry in which precise temporal control of aspects of neuronal discharge is critical. Third, the auditory system, especially at or below the level of the midbrain, contains significant numbers of GABAergic or glycinergic projection neurons. The latter feature distinguishes it from the central visual and somatic sensory pathways.

A novel and robust projection from y-aminobutyric acid-containing (GABAergic) inferior colliculus neurons to the medial geniculate body (MGB) was discovered in the cat using axoplasmic transport methods combined with immunocytochemistry. This input travels with the classical inferior colliculus projection to the MGB, and it is a direct ascending GABAergic pathway to the sensory thalamus that may be inhibitory. This bilateral projection constitutes 10-30% of the neurons in the auditory tectothalamic system. Studies by others have shown that comparable input to the corresponding thalamic visual or somesthetic nuclei is absent. This suggests that monosynaptic inhibition or disinhibition is a prominent feature in the MGB and that differences in neural circuitry distinguish it from its thalamic visual and somesthetic counterparts.The thalamus is the gateway controlling the flow of sensory information reaching the cerebral cortex (1). The consensus is that input ascending to the primary thalamic nuclei for hearing (2), vision (3), and somesthesis (4) is entirely excitatory. We describe here a prominent and unexpected projection from y-aminobutyric acid-containing (GABAergic) cells in the inferior colliculus (IC) to the medial geniculate body (MGB) in the auditory tectothalamic (TT) pathway that constitutes an exception to this principle. This suggests that there is a convergence of classical excitatory and putatively inhibitory projections within the MGB and that an ascending pathway considered as neurochemically unitary in fact contains more than one channel. This striking difference in feedforward input between the auditory thalamus and other thalamic sensory nuclei is surprising since each has similar structural and intrinsic features (5-7), shared physiological properties (8-10), analogous cortical projection patterns (11-13), and a closely related neurochemical organization (14, 15). The present results suggest that, despite these parallels, there must be differences in circuitry and information processing regimes within the sensory thalamus.with isoflurane at a level that suppressed all nociceptive reflexes. All experimental procedures followed the applicable and approved institutional animal care and use protocols. Four unilateral penetrations were made using stereotaxic coordinates to target specific nuclei in various subdivisions. A total of seven, 50-nl deposits were made at 500-,um intervals along a track with a glass micropipet; the net volume was -1.5 ,lI.The animal was reanesthetized 3 days later and perfused with PBS followed by 2% paraformaldehyde and 3% glutaraldehyde in 0.12 M PB. Vibratome sections, 50-,m-thick, were collected in 0.1 M PB and placed in blocking serum (5% normal goat serum) for 60 min. Tissue was incubated overnight at 4°C in rabbit-anti-GABA (Incstar, Stillwater, MN) diluted 1:5000 and then reacted using avidin-biotin immunoperoxidase (Vectastain, Vector Laboratories, Burlingame, CA) with diaminobenzidine as the chromagen. WAHG labeling was intensified with silver (IntenSE-M...

This study examined the output of the central nucleus of the inferior colliculus to the medial geniculate body and other parts of the nervous system in the mustached bat (Pteronotus parnellii). Small deposits of anterograde tracers (horseradish peroxidase, [3H]leucine, Phaseolus vulgaris leucoagglutinin, wheat germ agglutinin conjugated to horseradish peroxidase, or biocytin) were made at physiologically defined sites in the central nucleus representing major components of the bat's echolocation signal. The topography, frequency specificity, and axonal morphology of these outputs were studied. The medial geniculate body was a major target of inferior collicular neurons, with three distinct input patterns. The projection to the ventral division was tonotopically organized, but had a relatively sparse contribution from neurons representing frequency modulated components of the biosonar pulse. The second input was to the rostral medial geniculate body, in which projections from inferior collicular neurons representing constant frequency sonar components were separated from those representing frequency modulated components. A third input was to the suprageniculate nucleus, which received strong, topographically arranged projections. Inputs to the dorsal nucleus and medial division were also observed. Extrathalamic regions receiving input included the pontine gray, external nucleus of the inferior colliculus, pericollicular tegmentum, nucleus of the brachium of the inferior colliculus, and pretectum. These central nucleus projections differed in organization and the structure of axon terminals, suggesting different physiological influences on their target nuclei. These results demonstrate that the central nucleus has divergent projections to various sensory and premotor nuclei, besides its well-established projection to the medial geniculate body.

No abstract