Olivocochlear neurons in the chinchilla: a retrograde fluorescent labelling study

Abstract

Although the chinchilla is widely used as a model for auditory research, little is known about the distribution and morphology of its olivocochlear neurons. Here, we report on the olivocochlear neurons projecting to one cochlea, as determined by single and double retrograde fluorescent tracer techniques. 10 adult chinchillas were anesthetized and given either unilateral or bilateral injections of a fluorescent tracer (either Fluoro-Gold or Fast Blue) into scala tympani or as a control, a unilateral injection into the middle ear cavity. The results indicate that there are similarities as well as significant differences between the chinchilla and other species of rodents in the distributions of their olivocochlear neurons. Based on three well-labelled cases, there was a mean total of 1168 olivocochlear neurons in the chinchilla. Of these, the majority (mean 787) were small, lateral olivocochlear neurons found almost exclusively within the ipsilateral lateral superior olivary nucleus. The next largest group consisted of a mean of 280 medial olivocochlear neurons virtually all of which were located in the dorsomedial peri-olivary nucleus. Chinchilla medial olivocochlear neurons were more predominantly crossed in their projections (4:1) than in any known species. The smallest group of olivocochlear neurons (mean 101) consisted of larger lateral olivocochlear neurons (shell neurons) which were located on the margins of the superior olivary nucleus and which projected mainly (2.2:1) ipsilaterally. Double retrograde labelling was observed only in medial olivocochlear neurons and occurred in only 1–2% of these cells. The results confirm previous findings which indicated a relative paucity of fibers belonging to the uncrossed as compared to the crossed olivocochlear bundle. This, together with the strong apical bias of the uncrossed projection reported previously, offers possible explanations for the apparent absence of efferent-mediated suppressive effects of contralateral acoustic stimulation in this species. Regarding the lateral olivocochlear system, the chinchilla is shown to possess both intrinsic and shell neurons, as in the rat.

Introduction

Previous studies of the olivocochlear (OC) system in mammals indicate that the efferent projection to each cochlea arises from neurons located in two distinct zones of the superior olivary complex, lateral and medial. Moreover, across species, the OC neurons in these two zones exhibit clear similarities in cellular morphology, preponderant side of projection and synaptic targets in the organ of Corti (for review, see Warr, 1992). The majority of these efferents are small, lateral OC neurons (LOC) which project predominantly (70–100%) to the ipsilateral cochlea where they synapse with the radial afferent fibers beneath the inner hair cells. The second largest group consists of large, medial OC neurons (MOC), which project predominantly (53–75%) to the contralateral cochlea where they synapse with outer hair cells. Finally, in some rodents, there is a small number of generally larger LOC neurons which closely surround the lateral superior olivary nucleus, for which reason they were named ‘shell’ neurons (Vetter and Mugnaini, 1992). These neurons are of special interest because they form distinctive, bidirectional, longitudinally diffuse, inner spiral and tunnel spiral bundle fibers (Brown, 1987) that terminate beneath the inner hair cells, preponderantly on the side ipsilateral to their origin (Warr et al., 1997).

The chinchilla is a widely used species in studies of the auditory system but the origins and distribution of its OC neurons, as demonstrated by retrograde tracing techniques, has until now not fully been documented. Nevertheless, previous work has explored a number of features of both the LOC and MOC systems in this species, including; (1) the fine structure of efferent synapses beneath the inner and outer hair cells (Smith and Rasmussen, 1963, Gulley and Reese, 1977), (2) the differences in magnitude and distribution between crossed and uncrossed efferents to the outer hair cells (Iurato et al., 1978), (3) the distribution and morphology of acetylcholinesterase (AChE)-stained neurons in the superior olivary complex (Osen et al., 1984), (4) the role of the OC bundle (OCB) in modulating cochlear mechanics (Siegel and Kim, 1982), (5) the role of the OCB in protecting against acoustic trauma (Zheng et al., 1997a, Zheng et al., 1997b), (6) the facilitatory action of opioid peptides contained in OC fibers beneath the inner hair cells (Sahley et al., 1991) and (7) the vulnerability of OC fibers beneath the inner hair cells to cholinergic neurotoxins (Morley et al., 1991).

Prior to the preliminary reports of retrograde labelling of OC efferents in chinchilla (Weekly et al., 1992, Azeredo et al., 1997, Kliment et al., 1997), there was indirect evidence, based on histochemical staining for AChE, on the cells of origin of the OCB in this species (Osen et al., 1984). This study demonstrated small neurons (presumably LOC) within the lateral superior olivary nucleus (LSO) and large neurons (presumably MOC) in the dorsomedial periolivary (DMPO) area. Supplementing these findings were electron microscopic data on the proportion of crossed versus uncrossed efferent innervation in the chinchilla (Iurato et al., 1978). After severing the decussation of the OCB in the brainstem, Iurato and his co-workers found only a slight loss of innervation beneath the inner hair cells, but a large percentage of boutons (81%) on outer hair cells degenerated. Such a large loss of endings was somewhat unexpected since, in the cat, the ratio of the crossed to uncrossed MOCs was closer to 2:1 than the nearly 4:1 observed in the chinchilla (Warr, 1975).

In the present study, we have pooled two independent sets of data, one obtained by investigators in Syracuse (NY, USA)using a double retrograde fluorescent labelling protocol and the other obtained by investigators in Omaha (NE, USA) using a single retrograde fluorescent labelling protocol. The motivation for the two studies differed somewhat in the two locations. In Syracuse, an investigation of the organization of the MOC system in the chinchilla became of interest when efforts to demonstrate the phenomenon of efferent-mediated contralateral sound suppression, either as measured by changes in the compound action potential (CAP) or in distortion product otoacoustic emissions (DPOAE), met with failure (E. Relkin, unpublished observations; D. Henderson, personal communication). Such a contralateral suppression, which is mediated by MOC neurons whose axons do not cross the midline but instead innervate the ear ipsilateral to their side of origin, has been clearly demonstrated in both the cat (Buno, 1978, Murata et al., 1980, Warren and Liberman, 1989, Puria et al., 1996) and the guinea pig (Puel and Rebillard, 1990, da Costa et al., 1997), as well as in man (Mott et al., 1989, Veuillet et al., 1991, Chery-Croze et al., 1993). Since these effects are mediated by the ipsilateral contingent of MOC neurons (Liberman and Brown, 1986, Warren and Liberman, 1989), it was of key importance to confirm the experimental indications that this pathway supplied only some 20% of efferent terminals beneath the outer hair cells (Iurato et al., 1978).

In Omaha, the motivation was to determine how the chinchilla’s efferent system is organized and how it compares to that of other mammals, generally, and more specifically, whether dual populations of LOC neurons, intrinsic and shell, were present in this species, as observed in the rat (Vetter and Mugnaini, 1992, Warr et al., 1997).