Chemically distinct rat olivocochlear neurons.

We have produced a neurochemical map of the cell bodies of origin of the cochlear efferent terminals in rat by combining glutamic acid decarboxylase (GAD), choline acetyltransferase (ChAT), or calcitonin gene-related peptide (CGRP) immunocytochemistry with retrograde transport of horseradish peroxidase. The locations of cochlear efferent cell bodies are in general agreement with the medial and lateral systems described by White and Warr (J. Comp. Neurol. 219:203-214, 1983) with some minor modifications. The lateral system consists of at least two pools of chemically distinct neurons located within the lateral superior olive (LSO) ipsilateral to the injected cochlea. One pool immunostains with an antibody to GAD while the other immunostains with antibodies to ChAT and to CGRP. The medial efferent system consists of periolivary neurons that are almost exclusively large and ChAT-positive but CGRP-negative. They are located both ipsilateral and contralateral to the cochlea they project to. There are a few GAD-positive small neurons in the medioventral and rostral periolivary regions that project ipsilaterally, but these may prove tobe ectopic neurons. The ipsilateral lateroventral periolivary region (LVPO) contains some efferent neurons, all of which are ChAT-positive but CGRP-negative. Additional cochlear efferent neurons, some of which are ChAT-positive and others GAD-positive, are present within and immediately dorsal to the fiber capsule surrounding the medial limb, and to a lesser extent the lateral limb, of the ipsilateral LSO. Not all GAD-positive or ChAT-positive olivary cells project to the cochlea. We have complemented the results in the brainstem by demonstrating two immunocytochemically distinct populations of efferent terminals in the cochlea simultaneously, one CGRP-positive and the other GAD-positive. Approximately equal numbers of boutons immunoreactive for both markers are present beneath inner hair cells throughout the entire length of the cochlea. Surprisingly high numbers of GAD-positive and CGRP-positive boutons are also present on outer hair cells, with each class having its spatially and morphologically distinct features. The lack of CGRP-positive periolivary cells that are retrogradely labeled by cochlear injections of HRP suggests that the lateral olivocochlear system sends projections to outer hair cells. Our results raise questions about species differences in the organization of targets of the lateral and medial olivocochlear systems.