Effects of neurotrophin and neurotrophin receptor disruption on the afferent inner ear innervation

Two neurotrophins and their two receptors appear to regulate the survival of vestibular and cochlear neurons in the developing ear. Mice lacking either brain derived neurotrophic factor (BDNF) or its associated receptor, Trk B, show a severe reduction in the number of vestibular neurons and a loss of all innervation to the semicircular canals. Mice lacking NT-3 or its receptor, Trk C, show a severe reduction of spiral neurons in the basal turn of the cochlea. Mice lacking both BDNF and NT-3 or Trk B and Trk C, reportedly lose all innervation to the inner ear. These two neurotrophins and their associated receptors are necessary for the normal afferent innervation of the inner ear.

Vestibuloocular reflex of the adult flatfish. III. A species-specific reciprocal pattern of excitation and inhibition

In juvenile flatfish the vestibuloocular reflex (VOR) circuitry that underlies compensatory eye movements adapts to a 90 degrees relative displacement of vestibular and oculomotor reference frames during metamorphosis. VOR pathways are rearranged to allow horizontal canal-activated second-order vestibular neurons in adult flatfish to control extraocular motoneurons innervating vertical eye muscles. This study describes the anatomy and physiology of identified flatfish-specific excitatory and inhibitory vestibular pathways. In antidromically identified oculomotor and trochlear motoneurons, excitatory postsynaptic potentials (EPSPs) were elicited after electrical stimulation of the horizontal canal nerve expected to provide excitatory input. Electrotonic depolarizations (0.8-0.9 ms) preceded small amplitude (<0.5 mV) chemical EPSPs at 1.2-1.6 ms with much larger EPSPs (>1 mV) recorded around 2.5 ms. Stimulation of the opposite horizontal canal nerve produced inhibitory postsynaptic potentials (IPSPs) at a disynaptic latency of 1.6-1.8 ms that were depolarizing at membrane resting potentials around -60 mV. Injection of chloride ions increased IPSP amplitude, and current-clamp analysis showed the IPSP equilibrium potential to be near the membrane resting potential. Repeated electrical stimulation of either the excitatory or inhibitory horizontal canal vestibular nerve greatly increased the amplitude of the respective synaptic responses. These observations suggest that the large terminal arborizations of each VOR neuron imposes an electrotonic load requiring multiple action potentials to maximize synaptic efficacy. GABA antibodies labeled axons in the medial longitudinal fasciculus (MLF) some of which were hypothesized to originate from horizontal canal-activated inhibitory vestibular neurons. GABAergic terminal arborizations were distributed largely on the somata and proximal dendrites of oculomotor and trochlear motoneurons. These findings suggest that the species-specific horizontal canal inhibitory pathway exhibits similar electrophysiological and synaptic transmitter profiles as the anterior and posterior canal inhibitory projections to oculomotor and trochlear motoneurons. Electron microscopy showed axosomatic and axodendritic synaptic endings containing spheroidal synaptic vesicles to establish chemical excitatory synaptic contacts characterized by asymmetrical pre/postsynaptic membrane specializations as well as gap junctional contacts consistent with electrotonic coupling. Another type of axosomatic synaptic ending contained pleiomorphic synaptic vesicles forming chemical, presumed inhibitory, synaptic contacts on motoneurons that never included gap junctions. Altogether these data provide electrophysiological, immunohistochemical, and ultrastructural evidence for reciprocal excitatory/inhibitory organization of the novel vestibulooculomotor projections in adult flatfish. The appearance of unique second-order vestibular neurons linking the horizontal canal to vertical oculomotor neurons suggests that reciprocal excitation and inhibition are a fundamental, developmentally linked trait of compensatory eye movement circuits in vertebrates.

Postlesional vestibular reorganization in frogs: evidence for a basic reaction pattern after nerve injury

Nerve injury induces a reorganization of subcortical and cortical sensory or motor maps in mammals. A similar process, vestibular plasticity 2 mo after unilateral section of the ramus anterior of N. VIII was examined in this study in adult frogs. The brain was isolated with the branches of both N. VIII attached. Monosynaptic afferent responses were recorded in the vestibular nuclei on the operated side following ipsilateral electric stimulation either of the sectioned ramus anterior of N. VIII or of the intact posterior vertical canal nerve. Excitatory and inhibitory commissural responses were evoked by separate stimulation of each of the contralateral canal nerves in second-order vestibular neurons. The afferent and commissural responses of posterior vertical canal neurons recorded on the operated side were not altered. However, posterior canal-related afferent inputs had expanded onto part of the deprived ramus anterior neurons. Inhibitory commissural responses evoked from canal nerves on the intact side were detected in significantly fewer deprived ramus anterior neurons than in controls, but excitatory commissural inputs from the three contralateral canal nerves had expanded. This reactivation might facilitate the survival of deprived neurons and reduce the asymmetry in bilateral resting activities but implies a deterioration of the original spatial response tuning. Extensive similarities at the synaptic and network level were noted between this vestibular reorganization and the postlesional cortical and subcortical reorganization of sensory representations in mammals. We therefore suggest that nerve injury activates a fundamental neural reaction pattern that is common between sensory modalities and vertebrate species.

State-space receptive fields of semicircular canal afferent neurons in the bullfrog

Receptive fields are commonly used to describe spatial characteristics of sensory neuron responses. They can be extended to characterize temporal or dynamical aspects by mapping neural responses in dynamical state spaces. The state-space receptive field of a neuron is the probability distribution of the dynamical state of the stimulus-generating system conditioned upon the occurrence of a spike. We have computed state-space receptive fields for semicircular canal afferent neurons in the bullfrog (Rana catesbeiana). We recorded spike times during broad-band Gaussian noise rotational velocity stimuli, computed the frequency distribution of head states at spike times, and normalized these to obtain conditional pdfs for the state. These state-space receptive fields quantify what the brain can deduce about the dynamical state of the head when a single spike arrives from the periphery.

Regional distribution of calcium currents in frog semicircular canal hair cells

In the present work we studied the regional expression of voltage-dependent Ca channels in hair cells from the frog semicircular canals, employing whole-cell patch-clamp on isolated and in situ hair cells. Although Ca channels are thought to play a major role in afferent transmission, up to now no data were available regarding their distribution in vestibular organs. The problem appears of interest, especially in the light of recent results showing the presence of multiple Ca current components in semicircular canal hair cells. Our data suggest the presence, in all regions of the crista ampullaris, of two classes of cells, one displaying an inactivating Ca current (R1) and one lacking it. In the former cells, Ca current amplitude decreased from the central to the peripheral zone (the maximal currents being observed in the intermediate zone). Only L-type and R2 current components displayed regional differences in expression, whereas the size and properties of R1, although variable among cells, were not regionalized. However, in cells lacking R1, Ca current amplitudes were similar regardless of cell shape and location. The possible contributions of this Ca current distribution to afferent discharge properties are discussed.

Central versus peripheral origin of vestibuloocular reflex recovery following semicircular canal plugging in rhesus monkeys

We have previously shown that there is a slowly progressing, frequency-specific recovery of the gain and phase of the horizontal vestibuloocular reflex (VOR) in rhesus monkeys following plugging of the lateral semicircular canals. The adapted VOR response exhibited both dynamic and spatial characteristics that were distinctly different from responses in intact animals. To discriminate between adaptation or recovery of central versus peripheral origin, we have tested the recovered vestibuloocular responses in three rhesus monkeys in which either one or both coplanar pairs of vertical semicircular canals had been plugged previously by occluding the remaining semicircular canals in a second plugging operation. We measured the spatial tuning of the VOR in two or three different mutually orthogonal planes in response to sinusoidal oscillations (1.1 Hz, +/-5 degrees, +/-35 degrees /s) over a period of 2-3 and 12-14 mo after each operation. Apart from a significant recovery of the torsional/vertical VOR following the first operation we found that these recovered responses were preserved following the second operation, whereas the responses from the newly operated semicircular canals disappeared acutely as expected. In the follow-up period of up to 3 mo after the second operation, responses from the last operated canals showed recovery in two of three animals, whereas the previously recovered responses persisted. The results suggest that VOR recovery following plugging may depend on a regained residual sensitivity of the plugged semicircular canals to angular head acceleration.

Convergence pattern of uncrossed excitatory and inhibitory semicircular canal-specific inputs onto second-order vestibular neurons of frogs. Organization of vestibular side loops

Second-order vestibular neurons of frogs receive converging monosynaptic excitatory and disynaptic excitatory and inhibitory inputs following electrical pulse stimulation of an individual semicircular canal nerve on the ipsilateral side. Here we revealed, in the in vitro frog brain, disynaptic inhibitory postsynaptic potentials (IPSPs) by bath application of antagonists specific for glycine or gamma-aminobutyric acid-A (GABA(A)) receptors. Differences in the response parameters between disynaptic IPSPs and excitatory postsynaptic potentials (EPSPs) suggested that disynaptic IPSPs originated from a more homogeneous subpopulation of thicker vestibular nerve afferent fibers than mono- or disynaptic EPSPs. To investigate a possible size-related organization of these canal-specific, parallel pathways, we combined long-lasting anodal currents of variable intensities with strong cathodal test pulses, to block pulse-evoked responses reversibly in a graded manner according to the size-related sensitivity of vestibular nerve afferent fibers. The anodal current intensity required to block a particular response component was about 15 times lower than the strength of the cathodal test pulse that activated this response component. These large threshold differences were exploited for a selective anodal suppression of the responses from thick vestibular nerve afferent fibers. In fact, response components known to originate exclusively from thick-caliber afferent fibers such as the electrically transmitted monosynaptic EPSP component exhibited the lowest thresholds for cathodal test pulses and were the first to disappear in the presence of small anodal polarization steps. Thresholds for the activation/inactivation of responses and current intensities required for response saturation/blockade were used to assess the fiber spectrum that evoked the different response components. Mono- and disynaptic EPSPs appeared to originate from a broad spectrum of thick and thin vestibular nerve afferent fibers. The spectrum of afferent fibers that activated disynaptic IPSPs on the other hand was more homogeneous and consisted of thick and intermediate fibers. Such a canal-specific and fiber type-related organization of converging inputs of second-order vestibular neurons via feedforward projections was shown for the first time by this study in frogs, but might also prevail in mammals. Similar differences in these feedforward pathways have been proposed earlier in a vestibular side-loop model. Our results are consistent with the basic assumptions of this model and relate to the processing and tuning of dynamic vestibular signals.

Spatial distribution of semicircular canal nerve evoked monosynaptic response components in frog vestibular nuclei

Most second-order vestibular neurons receive a canal-specific monosynaptic excitation, although the central projections of semicircular canal afferents overlap extensively. This remarkable canal specificity prompted us to study the spatial organization of evoked field potentials following selective stimulation of individual canal nerves. Electrically evoked responses in the vestibular nuclei were mapped systematically in vitro. Constructed activation maps were superimposed on a cytoarchitectonically defined anatomical map. The spatial activation maps for pre- and postsynaptic response components evoked by stimulation of a given canal nerve were similar. Activation maps for monosynaptic inputs from different canals tended to show a differential distribution of their peak amplitudes, although the overlap was considerable. Anterior vertical canal signals peaked in the superior vestibular nucleus, posterior vertical canal signals peaked in the descending and in the dorsal part of the lateral vestibular nucleus, whereas horizontal canal signals peaked in the descending and in the ventral part of the lateral vestibular nucleus. A similar, differential but overlapping, spatial organization of the canal inputs was described also for other vertebrates, suggesting a crude but rather conservative topographical organization of semicircular canal nerve projections within the vestibular nuclei. Differences in the precision of topological representations between vestibular and other sensory modalities are discussed.

Convergence patterns of the posterior semicircular canal and utricular inputs in single vestibular neurons in cats

The convergence of the posterior semicircular canal (PC) and utricular (UT) inputs in single vestibular nuclei neurons was studied intracellularly in decerebrate cats. A total of 160 vestibular neurons were orthodromically activated by selective stimulation of the PC and the UT nerve and classified according to whether or not they were antidromically activated from the spinal cord and oculomotor nuclei into vestibulospinal (VS), vestibulooculospinal (VOS), vestibuloocular (VO), and unidentified vestibular neurons. Fifty-three (33%) of 160 vestibular neurons received convergent inputs from both the PC and UT nerves. Seventy-nine (49%) vestibular neurons responded to PC inputs alone, and 28 (18%) neurons received inputs only from the UT nerve. Of 53 convergent neurons, 8 (15%) were monosynaptically excited from both nerves. Thirty-five (66%) received monosynaptic excitatory inputs from the PC nerve and polysynaptic excitatory or inhibitory inputs from the UT nerve, or vice versa. Approximately one-third of VS and VOS neurons received convergent inputs. A majority of the VS neurons descended to the spinal cord through the lateral vestibulospinal tract, while almost all the VOS neurons descended to the spinal cord through the medial vestibulospinal tract. The convergent neurons were found in all vestibular nuclei but more in the lateral nucleus and descending nucleus. The VS neurons were more numerous than VO neurons or VOS neurons.

Input of anterior and posterior semicircular canal interneurons encoding head-velocity to the dorsal Y group of the vestibular nuclei

Neurons in the Y group of the vestibular nuclei are activated disynaptically from the ipsilateral VIIIth nerve and polysynaptically from the contralateral nerve. The ipsilateral anterior and posterior semicircular canals project to the Y group via interneurons in the vestibular nuclei. Candidate interneurons located in the rostrolateral corner of the superior (SVN) and in the caudal medial (MVN) vestibular nuclei were retrogradely labeled by the iontophoretic injection of biocytin into the Y group. The physiology of these interneurons named Y-group projecting neurons (YPNs) was studied in the SVN. SVN-YPNs were activated antidromically by electric pulse stimulation in the Y group. The properties of SVN-YPNs are distinct from those of SVN flocculus projecting neurons (FPNs). Namely, YPNs have a lower resting rate than FPNs, have more irregular interspike intervals, show a different phase and gain during the vestibuloocular reflex, and are located differentially within the SVN. After the injection of biocytin into the Y group, the locations of Purkinje cells that project to the Y group were confined to the vertical zones of the flocculus and ventral paraflocculus. However, mossy fibers originating in the Y group terminate in both the vertical and horizontal zones of the flocculus and ventral paraflocculus as well as in the ipsilateral nodulus.

Calcium channels functional roles in the frog semicircular canal

Different types of voltage-operated calcium channels have been described in hair cells; however, no clear functional role has been assigned to them. As a first functional characterization of vestibular calcium channels, we studied the effect of several calcium channel agonists and antagonists on whole nerve firing rate in an isolated frog semicircular canal preparation. Resting activity was affected by all dihydropyridines tested and by omegaconotoxin GVIA, whereas only nimodipine was able to reduce the mechanically evoked activity. These results indicate that nimodipine-sensitive channels play a major role in afferent transmitter release, and omega-conotoxin GVIA sensitive channels regulate the afferent firing (possibly on the postsynaptic side) but with a less important role.

Canal and otolith inputs to single vestibular neurons in cats

Convergence of both afferents from the PC and saccular macula, and those from the PC and utricular macula on single vestibular neurons was noted by use of intercellular recording from vestibular neurons. Vestibular neurons were classified VO neurons (vestibulo-ocular proper neurons), VOS (Vestibulo-oculo-spinal neurons sending axon collaterals both to the extraocular motoneuron pools and to the spinal cord), VS neurons (vestibulospinal proper neurons) and V neurons (vestibular neurons without axons to the oculomotor nuclei or the spinal cord) on the basis of whether or not they responded antidromically to stimulation of the oculomotor nuclei and the spinal cord. Of the total 143 vestibular neurons recorded in the series of experiments on convergence of the PC and saccular afferents, 47 neurons (33%) were received inputs from both the PC and saccular nerves. Twenty-six of the 47 convergent neurons were identified as having the nature of VS neurons. Half (13/26) of those were activated monosynaptically from both the PC and saccular nerves. Only one saccular-activated neuron without PC inputs sent an axon to the oculomotor nuclei. In the other series of experiments on the convergence of the PC and utricular afferents, 41 (37%) of 111 vestibular neurons were proved to converge on inputs from both nerves. The majority (35/41) of the neurons received monosynaptic inputs from the PC nerve and polysynaptic EPSP-IPSP sequences from the utricular nerve, or vice versa. The ratio of PC-otolith convergent neurons among utricular-activated neurons (41/54, 76%) was higher than that among saccular activated neurons (47/88, 53%). The percentage of utricular alone neurons without PC inputs (13/111, 12%) was less than that of the saccular alone without PC inputs (41/145, 28%). In conclusion, the convergence of canal and otolith inputs likely contribute mainly to vestibulospinal reflexes including the vestibulocollic reflex, by sending inputs to the neck and other muscles during head inclination which creates the combined stimuli of angular and linear acceleration.

The effect of otolith and semicircular canal convergence on the VOR during eccentric rotation

VOR gain modulation was systematically investigated in the Rhesus monkey (M. mulatta) during centric and variable eccentric (up to 50 cm) sinusoidal rotation (4 Hz, 0.75 degree) with the nose facing in- or outward to test convergence of otolith and semicircular canal afferences. Earth-stationary lit LED-targets were placed at different distances (12-180 cm) from the monkey. Results were compared to biological demands. During centric rotation at 4 Hz when smooth pursuit mechanisms do not play a role, VOR gain--as expected--was approximately 1 without dependence on target distance. Phase of VOR and centrifuge were shifted by about 180 degrees as was predicted. If the monkey was rotated eccentrically with the nose facing outward the expected gain enhancement for close targets was obtained. Maximal experimental VOR gain during 4 Hz rotation was 4.4 which was close to demand at 50 cm eccentricity and 15 cm target distance (predicted gain: 4.6). If the nose points inward three situations have to be distinguished from simulation: (1) target behind the axis of rotation--VOR gain decrement should occur; (2) target on the axis of rotation--"inverse VOR suppression"; (3) target between monkey and axis of rotation--phase reversal. Experimentally, VOR gain decrement was obtained (situation 1). VOR gain was minimal (but not zero) for targets around the axis of rotation (situation 2). Situation 3 has not been investigated in detail so far.

Hyperventilation-induced nystagmus in patients with vestibular schwannoma

OBJECTIVE

To analyze the nystagmus evoked by hyperventilation in patients with unilateral vestibular schwannoma and to use this information to predict the effects of hyperventilation on individual ampullary nerves.

METHODS

Three-dimensional scleral search coil eye movement recording techniques were used to record the magnitude and time course of eye movements in six patients with unilateral vestibular schwannoma and hyperventilation-induced nystagmus. The presenting complaints in five of these patients were vertigo or dysequilibrium.

RESULTS

The eye movement response to hyperventilation was a "recovery" nystagmus with slow-phase components corresponding to excitation of the affected vestibular nerve. Projection of the eye velocity vector into the plane of the semicircular canals revealed that fibers arising from the ampulla of the horizontal canal were most affected by hyperventilation with lesser activation of fibers to the superior canal and smaller, more variable responses from posterior canal fibers.

CONCLUSIONS

The three-dimensional characteristics of the nystagmus evoked by hyperventilation in patients with vestibular schwannoma provide insight into the vestibular end organs affected by the tumor and the mechanism responsible for the nystagmus. This finding indicates that hyperventilation resulted in a transient increase in activity from these partially demyelinated axons.

Regional distribution of ionic currents and membrane voltage responses of type II hair cells in the vestibular neuroepithelium

Basolateral ionic currents and membrane voltage responses were studied in pigeon vestibular type II hair cells using a thin slice through either the semicircular canal (SCC) crista or utricular macular epithelium. Whole cell tight-seal patch-clamp recording techniques were used. Current-clamp and voltage-clamp studies were carried out on the same cell. One hundred ten cells were studied in the peripheral (Zone I) and central (Zone III) zones of the SCC crista, and 162 cells were studied in the striolar (S Zone) and extrastriolar (ES Zone) zones of the utricular macula. One of the major findings of this paper is that hair cells with fast activation kinetics of their outward currents are found primarily in one region of the SCC crista and utricular macula, whereas hair cells with slow activation kinetics are found in a different region. In Zone I of the crista, 95% of the cells have fast activation kinetics ("fast" cells) and in Zone III of the crista, 86% of the cells have slow activation kinetics ("slow" cells). In the utricular macula slice, 100% of the cells from the S Zone are slow cells, whereas 86% of the cells from the ES Zones are fast cells. Oscillation frequency (f) and quality factor (Q) of the damped oscillations of the membrane potential during extrinsic current injections were studied in hair cells in the different regions. The slow cells in Zone III and in the S Zone have a statistically significantly lower f, as a function of the amplitude of injected current, when compared with the fast cells in Zone I and the ES Zone. Although Q varied over a small range and was <2.6 for all cells tested, there was a statistically significant difference between Q for the membrane oscillations of the slow cells and fast cells in response to a range of current injections.

Sonic motor nucleus and its connections with octaval and lateral line nuclei of the medulla in a rockfish, Sebastiscus marmoratus

The sonic motor nucleus and its fiber connections were examined in a rockfish, Sebastiscus marmoratus by means of tracer methods using horseradish peroxidase (HRP), biocytin, and carbocyanine dye (DiI). Sebastiscus has a swimbladder and a pair of extrinsic sonic/drumming muscles. The sonic muscle is ipsilaterally innervated by the occipital nerve which is composed of two ventral roots arising from the sonic motor nucleus. The sonic motor neurons are distributed in the most ventral part of the ventral column from the caudal medulla to the rostral spinal cord, and form a ventrally located columnar nucleus. Each neuron in this nucleus possesses a long thick dendrite and several short dendrites. The long dendrite extends dorsolaterally and branches in the lateral funiculus, whereas the short dendrites branch around their cell bodies. After biocytin injections into the sonic motor nucleus, two groups of premotor neurons were retrogradely labeled bilaterally, one in the dorsomedial portion of the descending octaval nucleus (DO) and the other in the medial zone of the reticular formation (RF) in the medulla. The DO premotor neurons were multipolar with several dendrites branching near the cell bodies, and the RF premotor neurons were bipolar. One of the two dendrites of the RF premotor neurons extends laterally into the ventral portion of the DO, and the other dendrite extends into the ventromedial area in the medulla. In the ventromedial dendritic field of the RF premotor neurons, descending fibers arising from the optic tectum (TO) and torus semicircularis (TS) traverse in the tractus tectobulbaris and terminate bilaterally. After DiI insertion into the ventromedial dendritic field, retrogradely labeled neurons were found bilaterally in the TS and TO. The majority of tectal neurons were located in the stratum griseum centrale. These neurons had two short basal dendrites branching in the cell layer and a long apical dendrite extending to the stratum fibrosum et griseum superficiale and stratum opticum. The toral neurons were bipolar and were distributed throughout the TS. Furthermore, biocytin injections into the medial nucleus of the lateral line system revealed that the nucleus projects bilaterally to the RF premotor neurons. These results show that premotor neurons for the sonic motor nucleus are located in the dorsomedial portion of the DO and the medial zone of the RF in the medulla. It is suggested that the sonic motor nucleus receives auditory input via the DO premotor neurons and input from RF premotor neurons which receive lateral line input via the medial nucleus, vestibular input through the lateral dendrite extending into the ventral portion of the DO, and information from the TO and TS via the tractus tectobulbaris.

Characteristics of regenerating horizontal semicircular canal afferent and efferent fibers in the toadfish, Opsanus tau

The horizontal semicircular canal nerve of the toadfish, Opsanus tau, was transected and allowed to regenerate. The time course, morphometrics, and projection patterns of regenerating afferent and efferent vestibular fibers were determined. Nerve transections were performed both pre- and postganglionically, and regeneration was assessed in afferent and efferent fibers by bulk labeling the peripheral axons of the horizontal semicircular canal nerve with biocytin after nerve regrowth. Afferent fibers regrew through the transection site within 14 days and projected to all vestibular nuclei within 3 weeks. Bouton and branch number, axon length, surface area, volume, fiber diameter, and internodal distance were quantified for afferent fibers from eight sites within the vestibular nuclei, and axon number and soma size was quantified for the efferent fibers. Extensive regeneration was seen within 5 weeks of transection in all nuclei, and most morphometric parameters approached or exceeded control levels within 10 weeks. Regeneration appeared to recapitulate morphogenesis with an initial overproduction of boutons and branch points followed by elimination of presumably superfluous structures. Internodal distance remained significantly shorter in regenerating afferent axons than in control fish throughout the 15-week observation period. Efferent fibers also were observed to regenerate. Efferent axon number, diameter, and soma size were indistinguishable from those in controls from 3 weeks posttransection through week 15. Electrophysiological recordings from the horizontal canal nerve during mechanical stimuli of the canal confirmed that the regenerated axons transmitted normal signals. The return of normal equilibrium and behavior coincided with the projection of afferent fibers into the central vestibular nuclei, indicating that functional connections had been reestablished.

Modelling the firing pattern of bullfrog vestibular neurons responding to naturalistic stimuli

We have developed a neural system identification method for fitting models to stimulus-response data, where the response is a spike train. The method involves using a general nonlinear optimisation procedure to fit models in the time domain. We have applied the method to model bullfrog semicircular canal afferent neuron responses during naturalistic, broad-band head rotations. These neurons respond in diverse ways, but a simple four parameter class of models elegantly accounts for the various types of responses observed.

Canal and otolith afferent activity underlying eye velocity responses to pitching while rotating

Pitching the head while rotating (PWR) combines periodic activation of the semicircular canals and the otoliths to generate pitch and roll eye deviations and continuous horizontal nystagmus. Monkeys were tested after individual pairs of semicircular canals were plugged and single units were recorded in the vestibular nerve while the animals were sinusoidally pitched 20-40 deg about a spatial horizontal axis with 5- and 16-s periods and simultaneously rotated about a spatial vertical axis at 30-120 deg/s. As previously shown, the steady-state horizontal response disappeared after plugging the vertical semicircular canals, but was maintained when the lateral canals were plugged. When the left anterior and right posterior canal (LARP) pair was left intact, the steady-state response depended on the axis about which the pitching took place. When the axis was normal to the LARP plane, there was no steady-state response. When the pitching axis was perpendicular to the LARP normal, the response was maximal. Firing rates of otolith units were approximately in phase with pitch position, and the addition of rotation about a vertical axis did not change the response. Lateral canal units did not have a steady-state modulation during pitch or constant velocity rotation. During PWR, they oscillated at twice the pitch frequency. This corresponded to the frequency at which the canal was maximally activated as it aligned with the plane of rotation. The amplitude of modulation increased proportionally to rotational velocity, but the phase remained the same. These characteristics were unchanged during roll while rotating (RWR), which induces little continuous nystagmus. Anterior and posterior canal units were maximally excited near pitch-velocity maxima and minima, respectively, during pure pitching. During PWR, however, the phases of both components simultaneously shifted toward each other and toward being in phase with otolith units. The peak excitation tended toward a forward-pitch position when the rotation was to the ipsilateral side, and toward a backward pitch position when the rotation was to the contralateral side. With 120-deg/s rotation during a 16-s pitch period, the phase difference between anterior and posterior canal units was as small as 17 deg. These data support the postulate that the correlation between vertical canal and otolith units is the critical factor in generating continuous unidirectional horizontal nystagmus during PWR.

Recovery of semicircular canal primary afferent activity in the pigeon after streptomycin ototoxicity

Recovery of semicircular canal primary afferent activity in the pigeon after streptomycin ototoxicity. J. Neurophysiol. 80: 3297-3311, 1998. The electrophysiological activity of horizontal semicircular canal primary afferents (HSCPA) was investigated in vivo in the barbiturate-anesthetized pigeon by means of extracellular single-fiber vestibular nerve action potential recordings. The spontaneous and driven discharges to pulse (step/trapezoid waveform, peak velocity = 120 degrees/s) and sum-of-sines (0.03, 0.09, 0.21, 0.39, 0.93, 1.83 Hz, peak velocity = 30 degrees/s for each frequency) rotations were measured both in normal control animals and a group of animals at 30, 40, 50, 60, 71, and 150 days postinjection sequence (PIS) of streptomycin sulfate. Prior to 30 days PIS, the activity in the nerve was not appropriately modulated during and after rotation. At 30 days PIS and thereafter, the responses resembled those observed in control animals but with systematic changes in parameters of fitted pulse responses and fitted Bode plots as days PIS increased. The return of parameters characterizing the neural dynamics of the semicircular canals were monotonic and could be best described by either linear or exponential functions. After 30 days PIS, the mechanical cupula-endolymph system, the function of which can be inferred from the cupula long time constant (tauL) following step velocity, did not change systematically (tauL = 6.92 +/- 3.96, 8.64 +/- 5.52, 8.35 +/- 4.21, 10.00 +/- 2.79, 9.05 +/- 3.67, 7.05 +/- 2.72; means +/- SD). However, the mean gain (G) of the HSCPA response to pulse rotation nearly doubled between 30 and 150 days PIS (from 1.31 +/- 0. 39 to 2.40 +/- 1.04) and returned linearly to control values (G = 2. 39 +/- 0.77) over this time period [G = 1.33 + 0.009(PIS-30), R2 = 0. 92, P < 0.05]. Meanwhile, neural adaptation as quantitated using a fractional operator, k, decayed exponentially (single exponential) to an asymptote. The time constant of this exponential was approximately 55 days [k = 0.034 + 0.33e-(PIS-30)/55.4, R2 = 0.99, P < 0.01]. Features of the spontaneous discharge previously shown to be correlated with k changed appropriately. That is, the coefficient of variation (CV) and frequency of firing (FF) decayed and grew asymptotically, respectively. These parameters also exhibited an exponential time course of return to control values from 30 to 150 days PIS [CV = 0.44 + 0.65e-(PIS-30)/21.5, R2 = 0.96, P < 0.01, and FF = 39.97 + 101.42(1 - e-(PIS-30)/32.6), R2 = 0.97, P < 0.01]. The trends of recovery for G, k, and tauL derived from analysis of the pulse response were confirmed by strong positive correlations with best fitted parameters obtained from analysis of the sum-of-sines frequency domain response of HSCPAs. There were statistically significant correlations (r = 0.90, P < 0.05 and r = 0.93, P < 0.05) between parameters (G, k) derived from pulse responses and those (G', k') from sum-of-sines responses, respectively. The cupula time constant based on sum-of-sines' data (tau'L) showed no statistically significant change between 30 and 150 days PIS (P > 0.05, analysis of variance). Thus the results in present study indicate that both the spontaneous discharge and the driven response to rotation of pigeon HSCPAs recovered their normal physiological status between 30 and 150 days PIS after hair cell death due to aminoglycoside ototoxicity. The recovery was systematic for the parameters chosen to be tested with the exception of the cupula long time constant, tauL. The mechanisms (changes in ciliary dynamics, changes in hair cell ionic currents, changes in bouton terminals, etc.) underlying these changes await further morphophysiological studies.

The metabotropic glutamate receptors of the vestibular organs

This research sought to test the presence and function of metabotropic excitatory amino acid receptors (mGluR) in the frog semicircular canal (SCC). The mGluR agonist +/- 1-aminocyclopentane-trans-1,3-dicarboxylate (ACPD) produced an increase in afferent firing rates of the ampullar nerve of the intact posterior canal. This increase was not due to a stimulation of cholinergic efferent terminals or the acetylcholine (ACh) receptor, since atropine, in concentrations which blocked the response to exogenous acetylcholine, did not affect the response to ACPD. Likewise, ACPD effects were not due to stimulation of postsynaptic NMDA receptors, since the NMDA antagonist D(-)-2-amino-5-phosphonopentanoate (AP-5) did not affect the response to ACPD, reinforcing the reported selectivity of ACPD for mGluRs. When the SCC was superfused with artificial perilymph known to inhibit hair cell transmitter release (i.e. low Ca-high Mg), ACPD failed to increase afferent firing. This suggests that the receptor activated by ACPD is located on the hair cell. Pharmacological evidence suggested that the mGluRs involved in afferent facilitation belong to Group I (i.e. subtypes 1 and 5). In fact, the Group III agonist AP-4 had no effect, and the ACPD facilitatory effect was blocked by the Group I mGluR antagonists (S)-4-carboxyphenylglycine (CPG) and (RS)-1-aminoindan-1,5-dicarboxylic acid (AIDA). Additional pharmacological evidence supported the presence of Group I mGluRs. Interestingly, the mGluR antagonists, AIDA and 4CPG, by themselves did not affect the resting firing rates of ampullar afferents. This may suggest that the mGluRs are not involved in resting activity but perhaps only in evoked activity (as suggested in Guth et al. (1991) Hear. Res. 56, 69-78). In addition, the mRNA for the mGluR1 has been detected in hair cells of both SCC, utricle, and saccule. In summary, the evidence points to an mGluR localized to the hair cell (i.e. an autoreceptor) which may be activated to produce a positive feedback augmentation of evoked but not resting transmitter release and thus affect afferent activity.

Sympathetic nerve activity during natural stimulation of horizontal semicircular canals in humans

We have shown that static head-down neck flexion elicits increases in muscle (MSNA) but not skin sympathetic nerve activity (SSNA) in humans. These findings suggest that stimulation of the otolith organs causes differential sympathetic outflow to vascular beds. The purpose of the present study was to determine whether yaw head rotation (YHR), which stimulates the horizontal semicircular canals, elicits sympathetic nerve responses. To test this question, we recorded MSNA (n = 33) and SSNA (n = 25) before and during 3 min of sinusoidal YHR performed at 0.1, 0.6, and 1.0 Hz. At all frequencies, YHR elicited no significant changes in heart rate and mean arterial pressure. Likewise, YHR did not significantly change either MSNA or SSNA at all frequencies. Our results indicate that stimulation of the horizontal semicircular canals by YHR does not alter SNA to either muscle or skin. Moreover, these results provide evidence to support the concept that the otolith organs but not the horizontal semicircular canals participate in the regulation of SNA in humans.

Evolutionary trends in the organization of the vertebrate crista ampullaris

Intraaxonal labeling studies in the toadfish, frog, turtle, and chinchilla suggest broad evolutionary trends in the vertebrate crista ampullaris. The crista of anamniotes (fish, amphibians) contains type II hair cells innervated by bouton afferents and is longitudinally organized. Type I hair cells are first seen in reptiles and birds, where they are confined to a central zone and are innervated by calyx and dimorphic afferents. The central zone is surrounded by a peripheral zone containing only type II hair cells innervated by bouton afferents. Results in the turtle suggest that the peripheral zone in reptiles and birds is organized similarly to the entire anamniote crista. The turtle central zone finds no parallel in anamniotes but resembles the mammalian central zone in its structure and afferent physiology. With the advent of a central zone in reptiles, a concentric organization is superimposed on a linearly organized peripheral zone. The mammalian crista, in contrast, has an entirely concentric organization. This may be related to the extension of the neuroepithelium further down the slopes of the crista in mammals than in other vertebrates and to the distribution of type I hair cells throughout the mammalian neuroepithelium.

Anatomical evidence for binaural processing in the descending octaval nucleus of the toadfish (Opsanus tau)

The connections of a potential auditory circuit were determined in the medulla of the toadfish (Opsanus tau). Fluorescent dextran amines placed in the medial torus semicircularis (mTS) retrogradely filled cells primarily in the dorsal region of the descending octaval nuclei (DON) with contralateral predominance. Fluorescent dextran amines placed in the DON revealed commissural fibers that cross the midline with the internal arcuate tract. The interconnections are consistent with a dorsal-ventral organization of the DON: reciprocal innervation is present for the left and right dorsal zones of the DON and for the left and right ventral zones of the DON. Based on projections to the medial (auditory) TS and the reciprocal connections, the dorsal region of the DON appears to be the major auditory processing site in the medulla and also may be a site for directional, binaural comparisons. The ventral region of the DON may be a site for bilateral vestibular processing. Double-labelling experiments revealed that some of the descending octaval cells projecting to the contralateral DON also project to the mTS. Based on the auditory pathway indicated by this study, future neurophysiological investigations of sensitivity to directional sound stimuli should begin in the dorsal DON of the toadfish.

A 'neural' response with 3-ms latency evoked by loud sound in profoundly deaf patients

A large negative deflection with a latency of 3 ms was observed in the auditory brainstem response (ABR) waveforms of some patients with peripheral profound deafness. This deflection was termed the N3 potential. In this paper, we review patients with the N3 potential and discuss the characteristics of abnormal ABR waveforms. The origin of the N3 potential was also discussed, especially with respect to vestibular evoked potentials. In most of the patients, audiograms showed no response to the maximum output of an audiometer in the high-frequency range and a residual response in the low-frequency range. The N3 potentials were noted at intensities of 80 dB nHL or greater. As the stimulus intensity increased, the amplitude of the potential increased and the latency decreased. A high repetition rate (83.3/s) of the click stimulus influenced the latency and amplitude of the N3 potential. The potential was replicated on retest within less than a month, and had a consistent latency and amplitude over the scalp. The results indicate that the N3 potential is not an electrical artifact but a physiological neural response evoked by a loud sound. The N3 potential is most likely not an auditory evoked response from cochlear or a response from a semicircular canal, because it has a 3-ms latency, a sharp waveform, and is unassociated with vertigo. The results suggest that the N3 potential may be a saccular acoustic response.