Vestibular Axons Research Articles

Glutamate and GABA in Vestibulo-Sympathetic Pathway Neurons.

The vestibulo-sympathetic reflex (VSR) actively modulates blood pressure during changes in posture. This reflex allows humans to stand up and quadrupeds to rear or climb without a precipitous decline in cerebral perfusion. The VSR pathway conveys signals from the vestibular end organs to the caudal vestibular nuclei. These cells, in turn, project to pre-sympathetic neurons in the rostral and caudal ventrolateral medulla (RVLM and CVLM, respectively). The present study assessed glutamate- and GABA-related immunofluorescence associated with central vestibular neurons of the VSR pathway in rats. Retrograde FluoroGold tract tracing was used to label vestibular neurons with projections to RVLM or CVLM, and sinusoidal galvanic vestibular stimulation (GVS) was employed to activate these pathways. Central vestibular neurons of the VSR were identified by co-localization of FluoroGold and cFos protein, which accumulates in some vestibular neurons following galvanic stimulation. Triple-label immunofluorescence was used to co-localize glutamate- or GABA- labeling in the identified VSR pathway neurons. Most activated projection neurons displayed intense glutamate immunofluorescence, suggestive of glutamatergic neurotransmission. To support this, anterograde tracer was injected into the caudal vestibular nuclei. Vestibular axons and terminals in RVLM and CVLM co-localized the anterograde tracer and vesicular glutamate transporter-2 signals. Other retrogradely-labeled cFos-positive neurons displayed intense GABA immunofluorescence. VSR pathway neurons of both phenotypes were present in the caudal medial and spinal vestibular nuclei, and projected to both RVLM and CVLM. As a group, however, triple-labeled vestibular cells with intense glutamate immunofluorescence were located more rostrally in the vestibular nuclei than the GABAergic neurons. Only the GABAergic VSR pathway neurons showed a target preference, projecting predominantly to CVLM. These data provide the first demonstration of two disparate chemoanatomic VSR pathways.

Differential Roles for EphA and EphB Signaling in Segregation and Patterning of Central Vestibulocochlear Nerve Projections

Auditory and vestibular afferents enter the brainstem through the VIIIth cranial nerve and find targets in distinct brain regions. We previously reported that the axon guidance molecules EphA4 and EphB2 have largely complementary expression patterns in the developing avian VIIIth nerve. Here, we tested whether inhibition of Eph signaling alters central targeting of VIIIth nerve axons. We first identified the central compartments through which auditory and vestibular axons travel. We then manipulated Eph-ephrin signaling using pharmacological inhibition of Eph receptors and in ovo electroporation to misexpress EphA4 and EphB2. Anterograde labeling of auditory afferents showed that inhibition of Eph signaling did not misroute axons to non-auditory target regions. Similarly, we did not find vestibular axons within auditory projection regions. However, we found that pharmacologic inhibition of Eph receptors reduced the volume of the vestibular projection compartment. Inhibition of EphB signaling alone did not affect auditory or vestibular central projection volumes, but it significantly increased the area of the auditory sensory epithelium. Misexpression of EphA4 and EphB2 in VIIIth nerve axons resulted in a significant shift of dorsoventral spacing between the axon tracts, suggesting a cell-autonomous role for the partitioning of projection areas along this axis. Cochlear ganglion volumes did not differ among treatment groups, indicating the changes seen were not due to a gain or loss of cochlear ganglion cells. These results suggest that Eph-ephrin signaling does not specify auditory versus vestibular targets but rather contributes to formation of boundaries for patterning of inner ear projections in the hindbrain.

Cross talk between vestibular neurons and Schwann cells mediates BDNF release and neuronal regeneration

It is now well-established that an active cross-talk occurs between neurons and glial cells, in the adult as well as in the developing and regenerating nervous systems. These functional interactions not only actively modulate synaptic transmission, but also support neuronal growth and differentiation. We have investigated the possible existence of a reciprocal interaction between inner ear vestibular neurons and Schwann cells maintained in primary cultures. We show that ATP released by the extending vestibular axons elevates intracellular calcium levels within Schwann cells. Purinergic activation of the Schwann P2X(7) receptor induces the release of neurotrophin BDNF, which occurs via a regulated, tetanus-toxin sensitive, vesicular pathway. BDNF, in turn, is required by the vestibular neuron to support its own survival and growth. Given the massive release of ATP during tissue damage, cross-talk between vestibular neurons and Schwann cells could play a primary role during regeneration.

Adaptations of the vestibular system to short and long-term exposures to altered gravity

Long-term space flight creates unique environmental conditions to which the vestibular system must adapt for optimal survival of a given organism. The development and maintenance of vestibular connections are controlled by environmental gravitational stimulation as well as genetically controlled molecular interactions. This paper describes the effects of hypergravity on axonal growth and dendritic morphology, respectively. Two aspects of this vestibular adaptation are examined: (1) How does long-term exposure to hypergravity affect the development of vestibular axons? (2) How does short-term exposure to extremely rapid changes in gravity, such as those that occur during shuttle launch and landing, affect dendrites of the vestibulocerebellar system? To study the effects of longterm exposures to altered gravity, embryonic rats that developed in hypergravity were compared to microgravity-exposed and control rats. Examination of the vestibular projections from epithelia devoted to linear and angular acceleration revealed that the terminal fields segregate differently in rat embryos that gestated in each of the gravitational environments.To study the effects of short-term exposures to altered gravity, mice were exposed briefly to strong vestibular stimuli and the vestibulocerebellum was examined for any resulting morphological changes. My data show that these stimuli cause intense vestibular excitation of cerebellar Purkinje cells, which induce up-regulation of clathrin-mediated endocytosis and other morphological changes that are comparable to those seen in long-term depression. This system provides a basis for studying how the vestibular environment can modify cerebellar function, allowing animals to adapt to new environments.

Pathfinding and growth termination of primary trigeminal sensory afferents in the embryonic rat hindbrain.

Axons of the trigeminal ganglion convey sensory information from mechanoreceptors, thermoreceptors, and nociceptors in the face and nasal mucosa, then terminate on several groups of neurons including the principal sensory nucleus and the nuclei of the spinal trigeminal tract. To understand guidance mechanisms during the development of trigeminal sensory axons (TA) in the embryonic brain, we first investigated the growth pattern of TA in relation to organization in the hindbrain using flat whole-mount preparation from rat. We found that the primary TA from the trigeminal ganglion entered the brainstem and grew longitudinally within the hindbrain. Whereas descending axons ran just medial to the primary vestibular axons to innervate the spinal nucleus, ascending axons stayed near the entry point. In flat whole-mount culture, the TA extended both ascending and descending branches as they do in vivo. Rostral hindbrain was found to be a less permissive substrate for the TA compared to caudal hindbrain. In addition, the nonpermissive property of the ventral hindbrain substrate restricted the invasion of TA along the entire length of the hindbrain. Thus, cooperation of absolute and relative permissiveness of the substrate plays important roles in the guidance of TA to their targets.

Demyelination of Vestibular Nerve Axons in Unilateral Ménière's Disease

We conducted a study to determine whether vestibular nerves in patients with unilateral Ménière's disease whose symptoms are refractory to medical management exhibit neuropathologic changes. We also endeavored to determine whether retrocochlear abnormalities are primary or secondary factors in the disease process. To these ends, we obtained vestibular nerve segments from five patients during retrosigmoid (posterior fossa) neurectomy, immediately fixed them, and processed them for light and electron microscopy. We found that all five segments exhibited moderate to severe demyelination with axonal sparing. Moreover, we noted that reactive astrocytes produced an extensive proliferation of fibrous processes and that the microglia assumed a phagocytic role. We conclude that the possible etiologies of demyelination include viral and/or immune-mediated factors similar to those seen in other demyelinating diseases, such as multiple sclerosis and Guillain-Barré syndrome. Our findings suggest that some forms of Ménière's disease that are refractory to traditional medical management might be the result of retrocochlear pathology that affects the neuroglial portion of the vestibular nerve.

Local nonpermissive and oriented permissive cues guide vestibular axons to the cerebellum.

Information that originates from peripheral sensory organs is conveyed by axons of cephalic sensory cranial ganglia connecting the sensory organs to appropriate central targets in the brain. Thus, the establishment of correct axonal projections by sensory afferents is one of the most important issues in neural development. Previously, we examined the development of the vestibular nerve that originates from the VIIIth ganglion using a flat whole-mount preparation of the rat hindbrain and developed an in vitro, culture preparation that can recapitulate vestibular nerve development (Tashiro, Y., Endo, T., Shirasaki, R., Miyahara, M., Heizmann, C. W. and Murakami, F. (2000) J. Comp. Neurol. 417, 491-500). Both in vivo and in vitro, the ascending branch of the VIIIth ganglion projecting to the cerebellum reaches the base of the cerebellar primordium and starts to splay out towards the rhombic lip, apparently avoiding the ventral metencephalon. We now examine the nature of cues that guide vestibulocerebellar axons by applying various manipulations to the flat whole-mount in vitro preparation. Our observations suggest that local nonpermissive cues and oriented cues play a pivotal role in the guidance of vestibular axons to their central target.

MuSC expression is correlated with fasciculation in mouse primary vestibular axons

MuSC expression is correlated with fasciculation in mouse primary vestibular axons

Update on the Pathology and Management of Benign Paroxysmal Positional Vertigo

Objective: To describe the otopathology in benign paroxysmal positional vertigo (BPV) and report our experience with singular neurectomy (SN) over a 24-year period. Materials: (I) The temporal bones (TB) of 3 patients who demonstrated BPV before death were acquired and prepared for examination by light microscopy. (II) A clinical series of patients who underwent SN for chronic disabling BPV has been updated to December 1998. Results: (I) In each of the 3 TB representing the downmost ear in the Hallpike maneuver, two pathologic changes were observed: (1) scattered degenerative neurons were found in the facial nerve meatal ganglion (MG); (2) focal axonal degeneration of vestibular axons (2 TB) or degenerated ganglion cells (1 TB) were observed in the inferior vestibular division. (II) From 1974 to 1998, SN was performed unilaterally in 177 patients and bilaterally (sequential) in 10 patients (n = 197). The age, sex and etiologic factors in the series were consistent with the literature. In 189 patients (96%), BPV was completely relieved by SN, while 3 patients (1.5%) experienced partial relief and 5 patients (2.5%) failed to benefit from SN. Sensorineural hearing loss followed SN in 5 patients (2.5%). Conclusions: The focal degeneration in the inferior vestibular nerve or ganglion may result from reactivation of a latent neurotropic virus in inferior vestibular ganglion cells as a result of various stressors (upper respiratory infection, trauma, surgery, general anesthesia, pregnancy). The portal of entry of the virus is likely from the nose and oral cavity to the MG of the facial nerve where latency is assumed and spread to the vestibular ganglion develops. The series of 197 SN over a 24-year period demonstrates that SN is a safe, effective surgical maneuver to selectively ablate function of the posterior semicircular canal crista.

Vestibular afferents to the dorsal vagal complex: substrate for vestibular-autonomic interactions in the rat

Vestibular afferents to the nucleus tractus solitarii (NTS) were identified for the first time in the male Sprague-Dawley rat. Cells of vestibular origin were labeled by deposits of cholera toxin B (CT-B) centered on the general viscerosensory division of NTS and dorsal motor nucleus (DMX). Vestibular-visceral afferents derive from neurons concentrated at caudal levels of medial and inferior vestibular nuclei as observed in other species. Vestibular afferent processes were labeled in the NTS and DMX by anterograde transport of the tracer, biotinylated dextran-amine from injection deposits confined to the inferior and/or medial vestibular nuclei. Vestibular axons terminate in the NTS, predominantly at intermediate levels of the dorsal vagal complex. Projections overlapped sites in NTS that receive terminal input from first-order alimentary and cardiorespiratory afferents. The somato-visceral reflex circuit corroborates recent evidence in the rat of increases in functional activity in the vestibular nuclear complex and NTS in response to changes in gravito-inertial force [Kaufman, G.D., Anderson, J.H. and Beitz, A.J., J. Neurosci., 12 (1992) 4489–4500]. Vestibular input to the NTS and DMX may assist in compensating for the effects imposed by movements and gravity on breathing, alimentary reflex function and the systemic circulation.

Distribution of efferent cholinergic terminals and alpha-bungarotoxin binding to putative nicotinic acetylcholine receptors in the human vestibular end-organs.

Although acetylcholine (ACh) has been identified as the primary neurotransmitter of the efferent vestibular system in most animals studied, no direct evidence exists that ACh is the efferent neurotransmitter of the human vestibular system. Choline acetyltransferase immunohistochemistry (ChATi), acetylcholinesterase (AChE) histochemistry, and alpha-bungarotoxin binding were used in human vestibular end-organs to address this question. ChATi and AChE activity was found in numerous bouton-type terminals contacting the basal area of type II vestibular hair cells and the afferent chalices surrounding type I hair cells; alpha-bungarotoxin binding suggested the presence of nicotinic acetylcholine receptors on type II vestibular hair cells and on the afferent chalices surrounding type I hair cells. This study provides evidence that the human efferent vestibular axons and terminals are cholinergic and that the receptors receiving this innervation may be nicotinic.

Developmental plasticity of reticulospinal and vestibulospinal axons in the north American opossum, Didelphis virginiana.

We have shown previously that rubral axons grow around a lesion of their spinal pathway in the North American opossum if it is made at early stages of development. In the present experiments, we have asked whether reticular and vestibular axons have the same ability. The spinal cord was hemisected at postnatal day 20, 12, or 5, well within the critical period for rubrospinal plasticity, and, approximately 30 days later, bilateral injections of fast blue were made about four segments caudal to the lesion. The pups were killed 4 or 5 days after the injections. In most of the animals lesioned on postnatal day 20, labeled neurons were not found in the medial part of the pontine reticular nucleus or the dorsal part of the lateral vestibular nucleus ipsilateral to the lesion. The spinal projections from both areas are exclusively ipsilateral. When the lesions were made at postnatal day 12 or 5, however, labeled neurons were present in both areas, suggesting that they supported axons that had grown caudal to the lesion. As was expected from previous studies, rubral neurons were labeled contralateral to the lesion in all three groups. In the opossum, as in other species, the red nucleus projects contralaterally. We conclude that reticular and vestibular axons, like axons from the red nucleus, grow around a lesion of their pathway during development and that the critical period for their plasticity ends earlier than that for rubrospinal axons.

Choline acetyltransferase immunoreactivity in the human vestibular end-organs

Acetylcholine (ACh) is believed to play a major role in the efferent vestibular system in several animal models, however no information regarding the role of ACh in the human efferent vestibular system has been published. Post-embedding immunohistochemistry in a hydrophilic resin was used to investigate the choline acetyltransferase immunoreactivity (ChATi) and acetylcholinesterase (ACHE) histochemistry in human vestibular end-organs. ChATi and AChE activity was found in numerous bouton-type terminals at the basal area of the vestibular hair cells. These terminals were found to contact type II vestibular hair cells and the afferent chalices surrounding type I hair cells. This study provides the first evidence that the human efferent vestibular axons and terminals are cholinergic.

Mice lacking brain-derived neurotrophic factor develop with sensory deficits.

During vertebrate development, neuronal survival depends on target-derived neurotrophic factors. Brain-derived neurotrophic factor (BDNF), a member of the neurotrophin family, can prevent the death of particular peripheral sensory neurons in vitro, and of central motor neurons as well as dopaminergic and cholinergic neurons of the basal forebrain during development. It also prevents the death of motor neurons and midbrain dopaminergic neurons induced by lesions. Here we show that mutant mice lacking BDNF have severe deficiencies in coordination and balance, associated with excessive degeneration in several sensory ganglia including the vestibular ganglion. The few remaining vestibular axons fail to contact the vestibular sensory epithelia, and terminate in the adjacent connective tissue. Survival of sympathetic, midbrain dopaminergic and motor neurons is not affected. These results indicate that BDNF is required for the survival and target innervation of particular neuronal populations.

Effect of precocious and delayed afferent arrival on synapse localization on the amphibian mauthner cell

Afferents often form synapses on restricted regions of their target cells. The connections between vestibular axons and the Mauthner cell are an example of this sort of specificity. The Mauthner cells are a pair of identifiable central neurons in certain fish and amphibians. In the axolotl (Ambystoma mexicanum), the large vestibular axons form club endings on only one portion of one dendrite of the Mauthner cell. During development, this specific distribution might result from nothing more than when and where the growing axons and dendrite meet (spatiotemporal model). Alternatively, the distribution might reflect constraints on where the developing axons may grow (axon segregation model) or form synapses (active recognition model). As a test of the spatiotemporal model, ears and vestibular ganglia from older embryos (Harrison stage 33/34-) were unilaterally grafted in place of prospective ear/ganglion placodes of younger embryos (stage 23). Light and electron microscopic examination revealed that the axons precociously entered the brain and formed synapses on the Mauthner cell. Similarly, grafts from younger to older embryos resulted in a delay. In both situations, mapping the club endings on the mature Mauthner cells of 21-mm larvae revealed a distribution similar to that of the contralateral control cells. Thus precocious and delayed afferent arrival did not alter the eventual localization of the synapses. These results conflict with the spatiotemporal model and imply that the vestibular axons are either restricted to a certain region of the neuropil or capable of recognizing a specific region of the Mauthner cell surface.

A light and electron microscopic study of the development of the mauthner cell and vestibular nerve in the axolotl

Vestibular axons form synapses on a restricted area of the lateral dendrite of the Mauthner cell, a large, identified brainstem neuron found in fish and amphibians. The differentiation of the vestibular nerve, medullary neuropil, and Mauthner cell of the axolotl (Ambystoma mexicanum) was studied to understand better the means by which this synaptic specificity arises. The Mauthner cell first extends a medial process and then a lateral dendrite. The latter initially elongates as a simple process and later sends out branches. As the lateral dendrite grows, vestibular axons enter the brainstem to form one of the earliest of several discrete axon fascicles that course longitudinally through the neuropil. The fascicles, many of which are identifiable on the basis of their location and axonal morphology, are the precursors of the longitudinal tracts of the mature salamander. The lateral dendrite grows dorsally over the orthogonally oriented fascicles, making contact with each at a characteristic time and place. The first afferents to form synapses do so on the soma and proximal lateral dendrite; subsequent afferent groups terminate more distally. Axons within a given fascicle form synapses with the Mauthner cell in a discrete and initially homogeneous domain. As dendritic branches form and the organization of the longitudinal fascicles becomes more complex, the homogeneity of axons terminating on a given region of the Mauthner cell surface is lost, but no major rearrangement or migration of terminals is apparent. These observations are consistent with both active recognition and passive spatiotemporal models of synaptic site specificity.

Dual projections of secondary vestibular axons in the medial longitudinal fasciculus to extraocular motor nuclei and the spinal cord of the squirrel monkey

Recordings were made from secondary vestibular axons in the medial longitudinal fasciculus (MLF) of barbiturate-anesthetized squirrel monkeys. Antidromic stimulation techniques were used to identify the axons as belonging to one of three classes of neurons: vestibulo-oculo-collic (VOC) neurons project both to the extraocular motor nuclei and to the spinal cord; vestibulo-ocular (VO) neurons do not have a spinal projection; and vestibulocollic (VC) neurons do not have an oculo-motor projection. Galvanic stimulation was used to show that axons of all three classes received excitatory inputs from one labyrinth and inhibitory inputs from the other. VOC axons were confined to the MLF contralateral to the labyrinth from which they were excited. They made up more than half of the vestibular axons descending in the contralateral medial vestibulospinal tract (MVST), but less than one-quarter of those ascending in the contralateral MLF to the level of the oculomotor nucleus. Spinal projections were restricted to cervical segments with about half of the axons reaching segment C6. Conduction velocities, measured for C6-projecting axons, were similar for VOC and VC axons and were typically 25-50 m/s. Unlike the situation in the rabbit (Akaike et al. 1973) and cat (Akaike 1983), none of the MVST axons had conduction velocities greater than 75 m/s. The morphology of VOC neurons was studied by injection of horseradish peroxidase (HRP) into 60 physiologically identified axons in the MLF. Since individual axons were only stained for short distances, it was not possible to ascertain their complete branching patterns. Labeled fibers could be traced to an origin in and around the ventral lateral vestibular nucleus. This localization was confirmed by comparing the distributions within the vestibular nuclei of neurons retrogradely labeled from the upper cervical spinal cord (this study) and from the oculomotor nucleus (McCrea et al. 1987a; Highstein and McCrea 1988). VOC axons reached the contralateral MLF at the level of the abducens nucleus and immediately divided into an ascending and a descending, usually thicker, branch. Seven VOC axons could be traced to the extraocular motor nuclei; three terminated in the medial aspect of the oculomotor nucleus bilaterally and four terminated in the medial aspect of the contralateral abducens nucleus. The former axons may be part of a crossed, excitatory anterior-canal pathway; the latter, part of a similar horizontal-canal pathway. There were no terminations in the trochlear nucleus even though 12 labeled fibers passed close to it.(ABSTRACT TRUNCATED AT 400 WORDS)

Vertical eye movement-related secondary vestibular neurons ascending in medial longitudinal fasciculus in cat. II. Direct connections with extraocular motoneurons

1. The preceding study in the alert cat has shown that many secondary vestibular axons that ascend in the medial longitudinal fasciculus (MLF) increase their firing rate in proportion to downward eye position. In the present study, projection and termination of these downward-position-vestibular (DPV) neurons within extraocular motoneuron pools were studied electrophysiologically by spike-triggered averaging techniques and morphologically be reconstructing their axonal trajectory after intra-axonal injection of horseradish peroxidase (HRP). 2. Extracellular field potentials recorded within the trochlear nucleus and/or the inferior rectus subdivision of the oculomotor nucleus were averaged by the use of spike potentials of single DPV neurons as triggers. All the crossed-DPV axons tested induced negative unitary field potentials in the trochlear nucleus (n = 9) and in the inferior rectus subdivision of the oculomotor nucleus (n = 5), suggesting that they made monosynaptic excitatory connection with motoneurons in these nuclei. The four crossed-DPV axons tested in the two motoneuron pools induced unitary field potentials in both. The majority of crossed-DPV axons terminated in these nuclei were directly activated from the caudal MLF, indicating that they had descending collaterals projecting to the spinal cord as well. The uncrossed-DPV axons did not induce such unitary field potentials either in the trochlear nucleus (n = 4) or in the inferior rectus subdivision (n = 3). 3. All the uncrossed-DPV axons examined (n = 14) induced positive unitary field potentials in the superior rectus subdivision of the oculomotor nucleus, suggesting that they made monosynaptic inhibitory connections with motoneurons innervating the superior rectus muscle. These uncrossed-DPV axons displayed regular firing patterns and were not activated from the caudal MLF. None of the crossed-DPV axons tested (n = 4) induced unitary field potentials in the superior rectus subdivision. 4. Five crossed-DPV axons were injected with HRP. All these axons ascended in the MLF contralateral to their soma, gave off many collaterals to the trochlear nucleus, and projected more rostrally. For three well-stained axons, numerous terminal branches were also found in the rostroventral part of the contralateral oculomotor nucleus, the area corresponding to the inferior rectus subdivision. Some collaterals in the oculomotor nucleus recrossed the midline to terminate in the medial part of the ipsilateral oculomotor nucleus. Other terminations were observed in the interstitial nucleus of Cajal and in the periaqueductal gray adjacent to the oculomotor nucleus. The crossed axons injected included both regular and irregular types, and three of the four examined were activated from the caudal MLF.(ABSTRACT TRUNCATED AT 400 WORDS)

Two distinct phases and mechanisms of axonal growth shown by primary vestibular fibres in the brain, demonstrated by parvalbumin immunohistochemistry

Antibodies to parvalbumin label intensely a small number of non-overlapping fibre systems in embryonic rat brain. All are in hindbrain—the oculomotor and trochlear motor fibres, the acoustic and vestibular fibres of the VIIIth nerve, and an unidentified group of fibres which ascend under the dorsal surface in caudal medulla. Of these, the vestibular fibres are the first to acquire parvalbumin immunore-activity, and we have used this property to follow the growth of their axons in the brain. This occurs in two phases. In the first, occurring at embryonic days 12–14, the axons grow in small groups or fascicles under the pial surface to their most distant terminal zones rostrally in the cerebellum and caudally in the descending vestibular nuclei. This growth is directed towards the two sites where germinal neuroepithelium is expanding over the medullary velum in forming the cerebellum and lateral recess of the IVth ventricle. In a second stage, commencing at E15, individual collaterals branch from these fascicles to arborize amongst their presumptive synaptic targets (cells of the vestibular nuclei and vestibulocerebellum) located in the sub-ventricular and ventricular layers. In this phase the axons follow a radial route, at right angles to their original subpial course, possibly by growing along radial glial processes. The target cells then migrate to their final position with the vestibular axons maintaining contact with them. The vestibular fibres are the first axons to enter the cerebellum, but from E15 onwards their fascicles are joined by increasing numbers of non-vestibular fibres following the same course. These other axons, and the movement of cells to form the deep cerebellar nuclei, separate the fascicles of vestibular fibres so that their course into the cerebellum becomes very diffuse. Thus this single set of axons grow, not only in two distinct phases, but also follow distinctly different substrates for growth in each. Furthermore, they then appear to act as pioneer fibres guiding the entry or egress of later-developing axons to or from the cerebellum.

Anatomical and physiological characteristics of vestibular neurons mediating the vertical vestibulo-ocular reflexes of the squirrel monkey.

The morphology of 35 vestibular neurons whose firing rate was related to vertical eye movements was studied by injection of horseradish peroxidase intracellularly into physiologically identified vestibular axons in alert squirrel monkeys. The intracellularly injected cells were readily classified into four main groups. One group of cells, down position-vestibular-pause neurons (down PVPs; N = 12), increased their firing rate during downward eye positions, paused during saccades, and were located in the medial vestibular nucleus (MV) and the adjacent ventrolateral vestibular nucleus (VLV). They had axons that crossed the midline and ascended in the medial longitudinal fasciculus (MLF) to terminate in the trochlear nucleus, the lateral aspect of the caudal oculomotor nucleus, and the dorsal aspect of the rostral oculomotor nucleus. A second group of cells (N = 15) were also located in the MV and VLV, but increased their firing rate during upward eye positions, and paused during saccades. These cells had axons that crossed the midline and ascended in the contralateral MLF to terminate in the medial aspect of the oculomotor nucleus. A third group of cells (N = 4) were located in the superior vestibular nucleus, generated bursts of spikes during upward saccades, and increased their tonic firing rate during upward eye positions. These cells had axons that ascended laterally to the ipsilateral MLF to terminate in regions of the trochlear and oculomotor nuclei similar to those in which down PVPs terminated. A fourth group of cells (N = 4), located in the VLV, had axons that projected to the spinal cord, although they had firing rates that were significantly correlated with vertical eye position. Electrical stimulation of the vestibular nerve evoked spikes at monosynaptic latencies in each of the above classes of cells, six of which were injected with horseradish peroxidase. Each group of cells had collateral projections to other areas of the brainstem. Some of the neurons that projected to the contralateral trochlear and oculomotor nuclei had collaterals that crossed the midline to terminate in the oculomotor nucleus ipsilateral to the soma, and some gave rise to small collaterals that terminated in the abducens nucleus. Other areas of the brainstem that received collateral inputs from neurons projecting to oculomotor and trochlear nuclei included the interstitial nucleus of Cajal, the caudal part of the dorsal raphe nucleus, the nucleus raphe obscurus, Roller's nucleus, the intermediate and caudal interstitial nuclei of the MLF, and the nucleus prepositus.