Patterns of canal and otolith afferent input convergence in frog second-order vestibular neurons

Second-order vestibular neurons (2 degrees VN) were identified in the isolated frog brain by the presence of monosynaptic excitatory postsynaptic potentials (EPSPs) after separate electrical stimulation of individual vestibular nerve branches. Combinations of one macular and the three semicircular canal nerve branches or combinations of two macular nerve branches were stimulated separately in different sets of experiments. Monosynaptic EPSPs evoked from the utricle or from the lagena converged with monosynaptic EPSPs from one of the three semicircular canal organs in ~30% of 2 degrees VN. Utricular afferent signals converged predominantly with horizontal canal afferent signals (74%), and lagenar afferent signals converged with anterior vertical (63%) or posterior vertical (37%) but not with horizontal canal afferent signals. This convergence pattern correlates with the coactivation of particular combinations of canal and otolith organs during natural head movements. A convergence of afferent saccular and canal signals was restricted to very few 2 degrees VN (3%). In contrast to the considerable number of 2 degrees VN that received an afferent input from the utricle or the lagena as well as from one of the three canal nerves (~30%), smaller numbers of 2 degrees VN (14% of each type of 2 degrees otolith or 2 degrees canal neuron) received an afferent input from only one particular otolith organ or from only one particular semicircular canal organ. Even fewer 2 degrees VN received an afferent input from more than one semicircular canal or from more than one otolith nerve (~7% each). Among 2 degrees VN with afferent inputs from more than one otolith nerve, an afferent saccular nerve input was particularly rare (4-5%). The restricted convergence of afferent saccular inputs with other afferent otolith or canal inputs as well as the termination pattern of saccular afferent fibers are compatible with a substrate vibration sensitivity of this otolith organ in frog. The ascending and/or descending projections of identified 2 degrees VN were determined by the presence of antidromic spikes. 2 degrees VN mediating afferent utricular and/or semicircular canal nerve signals had ascending and/or descending axons. 2 degrees VN mediating afferent lagenar or saccular nerve signals had descending but no ascending axons. The latter result is consistent with the absence of short-latency macular signals on extraocular motoneurons during vertical linear acceleration. Comparison of data from frog and cat demonstrated the presence of a similar organization pattern of maculo- and canal-ocular reflexes in both species.     

Vestibular primary afferent responses to sound and vibration in the guinea pig

This study tested whether air-conducted sound and bone-conducted vibration activated primary vestibular afferent neurons and whether, at low levels, such stimuli are specific to particular vestibular sense organs. In response to 500 Hz bone-conducted vibration or 500 Hz air-conducted sound, primary vestibular afferent neurons in the guinea pig fall into one of two categories--some neurons show no measurable change in firing up to 2 g peak-to-peak or 140 dB SPL. These are semicircular canal neurons (regular or irregular) and regular otolith neurons. In sharp contrast, otolith irregular neurons show high sensitivity: a steep increase in firing as stimulus intensity is increased. These sensitive neurons typically, but not invariably, were activated by both bone-conducted vibration and air-conducted sound, they originate from both the utricular and saccular maculae, and their sensitivity underpins new clinical tests of otolith function.     

Commissural effects in the otolith system

We examined whether otolith-activated second- and third-order vestibular nucleus neurons received commissural inhibition from the contralateral otolithic macula oriented in the same geometric plane. For this purpose we performed intracellular recording in vestibular nucleus neurons after stimulation of the ipsi- and contralateral utricular and saccular nerves. More than half (41/72) of the utricular-activated second-order vestibular nucleus neurons received commissural inhibition from the contralateral utricular nerve. The remaining neurons (31/72) showed no visible response to contralateral utricular nerve stimulation. About half (17/36) of utricular-activated third-order neurons also received commissural inhibition from the contralateral utricular nerve. Approximately 10% (7/67) of saccular-activated second-order vestibular neurons received polysynaptic commissural inhibition, whereas 16% (11/67) received commissural facilitation. The majority (49/67) of saccular second-order vestibular neurons, and almost all (22/23) third-order neurons, showed no visible response to stimulation of the contralateral saccular nerve. The present findings suggest that many utricular-activated vestibular nucleus neurons receive commissural inhibition, which may provide a mechanism for increasing the sensitivity of vestibular neurons to horizontal linear acceleration and lateral tilt of the head. Commissural inhibition in the saccular system was less prominent than in the utricular system.     

Second-order vestibular neurons form separate populations with different membrane and discharge properties

Membrane and discharge properties were determined in second-order vestibular neurons (2 degrees VN) in the isolated brain of grass frogs. 2 degrees VN were identified by monosynaptic excitatory postsynaptic potentials after separate electrical stimulation of the utricular nerve, the lagenar nerve, or individual semicircular canal nerves. 2 degrees VN were classified as vestibulo-ocular or -spinal neurons by the presence of antidromic spikes evoked by electrical stimulation of the spinal cord or the oculomotor nuclei. Differences in passive membrane properties, spike shape, and discharge pattern in response to current steps and ramp-like currents allowed a differentiation of frog 2 degrees VN into two separate, nonoverlapping types of vestibular neurons. A larger subgroup of 2 degrees VN (78%) was characterized by brief, high-frequency bursts of up to five spikes and the absence of a subsequent continuous discharge in response to positive current steps. In contrast, the smaller subgroup of 2 degrees VN (22%) exhibited a continuous discharge with moderate adaptation in response to positive current steps. The differences in the evoked spike discharge pattern were paralleled by differences in passive membrane properties and spike shapes. Despite these differences in membrane properties, both types, i.e., phasic and tonic 2 degrees VN, occupied similar anatomical locations and displayed similar afferent and efferent connectivities. Differences in response dynamics of the two types of 2 degrees VN match those of their pre- and postsynaptic neurons. The existence of distinct populations of 2 degrees VN that differ in response dynamics but not in the spatial organization of their afferent inputs and efferent connectivity to motor targets suggests that frog 2 degrees VN form one part of parallel vestibulomotor pathways.     

Differential inhibitory control of semicircular canal nerve afferent-evoked inputs in second-order vestibular neurons by glycinergic and GABAergic circuits

Labyrinthine nerve-evoked monosynaptic excitatory postsynaptic potentials (EPSPs) in second-order vestibular neurons (2°VN) sum with disynaptic inhibitory postsynaptic potentials (IPSPs) that originate from the thickest afferent fibers of the same nerve branch and are mediated by neurons in the ipsilateral vestibular nucleus. Pharmacological properties of the inhibition and the interaction with the afferent excitation were studied by recording monosynaptic responses of phasic and tonic 2°VN in an isolated frog brain after electrical stimulation of individual semicircular canal nerves. Specific transmitter antagonists revealed glycine and GABA_A receptor-mediated IPSPs with a disynaptic onset only in phasic but not in tonic 2°VN. Compared with GABAergic IPSPs, glycinergic responses in phasic 2°VN have larger amplitudes and a longer duration and reduce early and late components of the afferent nerve-evoked subthreshold activation and spike discharge. The difference in profile of the disynaptic glycinergic and GABAergic inhibition is compatible with the larger number of glycinergic as opposed to GABAergic terminal-like structures on 2°VN. The increase in monosynaptic excitation after a block of the disynaptic inhibition in phasic 2°VN is in part mediated by a N-methyl-D-aspartate receptor-activated component. Although inhibitory inputs were superimposed on monosynaptic EPSPs in tonic 2°VN as well, the much longer latency of these IPSPs excludes a control by short-latency inhibitory feed-forward side-loops as observed in phasic 2°VN. The differential synaptic organization of the inhibitory control of labyrinthine afferent signals in phasic and tonic 2°VN is consistent with the different intrinsic signal processing modes of the two neuronal types and suggests a co-adaptation of intrinsic membrane properties and emerging network properties.     

Saccular and utricular inputs to single vestibular neurons in cats

Saccular and utricular organs are essential for postural stability and gaze control. Although saccular and utricular inputs are known to terminate on vestibular neurons, few previous studies have precisely elucidated the origin of these inputs. We investigated the saccular and utricular inputs to single vestibular neurons in whole vestibular nuclei of decerebrated cats. Postsynaptic potentials were recorded from vestibular neurons after electrical stimulation of the saccular and utricular nerves. Ascending and descending axonal projections were examined by stimulating the oculomotor/trochlear nuclei and the cervical segment of the spinal cord, respectively. After each experiment, locations of recorded neurons were identified. The recorded neurons (140) were classified into vestibulo-spinal (79), vestibulo-oculo-spinal (9), and vestibulo-ocular (3) neurons based on antidromic responses; 49 other vestibular neurons were unidentified. The majority of recorded neurons were mainly located in the lateral vestibular nucleus. Most of the otolith-activated vestibular nuclei neurons seemed to participate in vestibulospinal reflexes. Of the total 140 neurons recorded, approximately one third (51) received saccular and utricular inputs (convergent neurons). The properties of these 51 convergent neurons were further investigated. Most (33/51) received excitatory postsynaptic potentials (EPSPs) after saccular and utricular nerve stimulation. These results implied that most of the convergent neurons in this study additively coded mixed information for vertical and horizontal linear acceleration. Based on the latencies of convergent neurons, we found that an early integration process for vertical and horizontal linear acceleration existed at the second-order level.     

Evidence for the utricular origin of the vestibular short-latency-evoked potential (VsEP) to bone-conducted vibration in guinea pig

Previous studies have shown that the vestibular short-latency-evoked potential (VsEP) in response to the brief head acceleration stimulus is a compound action potential of neurons innervating the otolith organs. However, due to the lack of direct evidence, it is currently unclear whether the VsEP is primarily generated by the activity of utricular or saccular afferent neurons, or some mixture of the two. Here, we investigated the origin of the VsEP evoked by brief bone-conducted vibration pulses in guinea pigs, using selective destruction of the cochlea, semicircular canals (SCCs), saccule, or utricle, along with neural blockade with tetrodotoxin (TTX) application, and mechanical displacements of the surgically exposed utricular macula. To access each end organ, either a dorsal or a ventral surgical approach was used. TTX application abolished the VsEP, supporting the neurogenic origin of the response. Selective cochlear, SCCs, or saccular destruction had no significant effect on VsEP amplitude, whereas utricular destruction abolished the VsEP completely. Displacement of the utricular membrane changed the VsEP amplitude in a non-monotonic fashion. These results suggest that the VsEP evoked by BCV in guinea pigs represents almost entirely a utricular response. Furthermore, it suggests that displacements of the utricular macula may alter its response to bone-conduction stimuli.     

Otolith and canal integration on single vestibular neurons in cats

In this review, based primarily on work from our laboratory, but related to previous studies, we summarize what is known about the convergence of vestibular afferent inputs onto single vestibular neurons activated by selective stimulation of individual vestibular nerve branches. Horizontal semicircular canal (HC), anterior semicircular canal (AC), posterior semicircular canal (PC), utricular (UT), and saccular (SAC) nerves were selectively stimulated in decerebrate cats. All recorded neurons were classified as either projection neurons, which consisted of vestibulospinal (VS), vestibulo-oculospinal (VOS), vestibulo-ocular (VO) neurons, or non-projection neurons, which we simply term vestibular (V) neurons. The first three types could be successfully activated antidromically from oculomotor/trochlear nuclei and/or spinal cord, and the last type could not be activated antidromically from either site. A total of 1228 neurons were activated by stimulation of various nerve pair combinations. Convergent neurons were located in the caudoventral part of the lateral, the rostral part of the descending, and the medial vestibular nuclei. Otolith-activated vestibular neurons in the superior vestibular nucleus were extremely rare. A high percentage of neurons received excitatory inputs from two nerve pairs, a small percentage received reciprocal convergent inputs and even fewer received inhibitory inputs from both nerves. More than 30% of vestibular neurons received convergent inputs from vertical semicircular canal/otolith nerve pairs. In contrast, only half as many received convergent inputs from HC/otolith-nerve pairs, implying that convergent input from vertical semicircular canal and otolith-nerve pairs may play a more important role than that played by inputs from horizontal semicircular canal and otolith-nerve pairs. Convergent VS neurons projected through the ipsilateral lateral vestibulospinal tract (i-LVST) and the medial vestibulospinal tract (MVST). Almost all the VOS neurons projected through the MVST. Convergent neurons projecting to the oculomotor/trochlear nuclei were much fewer in number than those projecting to the spinal cord. Some of the convergent neurons that receive both canal and otolith input may contribute to the short-latency pathway of the vestibulocollic reflex. The functional significance of these convergences is discussed.     

Gradual and reversible central vestibular reorganization in frog after selective labyrinthine nerve branch lesions

Postlesional reorganization of vestibular afferent and commissural inputs onto second-order vestibular neurons was studied in the isolated brain after unilateral section of the N.VIII, of the ramus anterior (RA) of N.VIII, of the utricular (UT) or of the anterior vertical and horizontal canal nerves in combination. RA nerve section eliminated the inputs from utricular, anterior vertical and horizontal canal organs. In the first set of experiments we recorded field potentials on the operated side of the vestibular nuclei 2 months after RA nerve section. These responses were evoked by electrical stimulation of the RA nerve or of the posterior vertical canal nerve on the operated or on the intact side. The amplitudes of afferent field potentials evoked by stimulation of the spared posterior vertical canal nerve were increased. The amplitudes of afferent field potentials evoked by stimulation of the axotomized RA nerve remained unaltered. After N.VIII section the commissural, but not the afferent, field potentials increased significantly on the operated side following stimulation of N.VIII on the intact and on the operated side, respectively. After UT nerve section no change in commissural but an increase in the amplitude of afferent field potentials from each of the three intact canal nerves was observed on the operated side. In the context of earlier results these findings imply that second-order vestibular neurons, disfacilitated due to afferent nerve section, became receptive to additional, excitatory synaptic inputs, preferentially from intact vestibular nerve afferent fibers. The reduced excitation via afferent nerve inputs was thereby replaced by other afferent nerve inputs from spatially inadequate vestibular end-organs. The synaptic terminals of inactivated afferent nerve fibers were maintained and not repressed. The process of central reorganization after vestibular nerve lesion was activity related, the expansion of signals restricted to inputs from intact fibers, its extent graded and its onset delayed with respect to the onset of corresponding spinal changes and to the onset of postural recovery after the same type of nerve lesion. After the section of RA nerve or of an individual nerve branch the labyrinthine end-organs remained intact and were not removed as after unilateral labyrinthectomy (UL). Peripheral reinnervation of the end-organs was thus excluded after UL, but expected after one of the former types of lesion. Functional reinnervation of the utricular macula was mirrored behaviorally by the reappearance of severe postural deficits following a second RA nerve section. These lesion-induced postural deficits began to reappear if the repeated RA nerve section was delayed with respect to the first by about 3 months. We therefore studied postlesional reorganization in the brainstem 3 months after the first RA nerve section. Reinnervation of the utricular macula was accompanied by a rapid decline of the increased amplitudes of afferent and commissural vestibular field potentials towards control values, suggesting the reversibility of the lesion-induced central reorganization. Electronic Publication     

Otolith-activated vestibulothalamic neurons in cats

The components of the vestibular ascending pathway that transmit otolith information to the thalamus were studied electrophysiologically in anesthetized cats. Thalamic-projecting vestibular neurons (confirmed antidromically) were recorded extracellularly in the various vestibular nuclei. Otolith inputs to these neurons were examined with selective stimulation of the utricular (UT) or the saccular (SAC) nerves. Vestibular nerve branches other than the tested nerve were transected. Of 40 UT-activated vestibulothalamic neurons, 40% (16/40) were activated by UT nerve stimulation with latencies ranging between 0.9-1.4 ms, suggesting they were second-order neurons from the UT nerve. UT-activated vestibulothalamic neurons were recorded in the medial vestibular nucleus (MVN; 24/40), the lateral vestibular nucleus (LVN; 9/40), the descending vestibular nucleus (DVN; 6/40), and the superior vestibular nucleus (SVN; 1/40). Most of the neurons (38/40) were antidromically activated by focal stimulation of the ventral part of the ipsilateral thalamus. Antidromic stimulation of the pontine area revealed that trajectories of the ascending axons (14 of 38 neurons) to the ipsilateral thalamus passed through the pontine reticular formation, ventral to the ascending tract of Deiters (ATD) and the medial longitudinal fasciculus (MLF). Only three SAC-activated vestibulothalamic neurons were encountered in the LVN. All these neurons were second-order neurons from the SAC nerve and were antidromically activated by stimulation of the contralateral thalamus, in marked contrast to the UT-activated vestibulothalamic neurons. Only three UT-activated and two SAC-activated neurons sent descending collaterals to the spinal cord.     

Properties of utricular and saccular nerve-activated vestibulocerebellar neurons in cats

Properties of otolith inputs to vestibulocerebellar neurons were investigated in 14 adult cats. In the vestibular nuclei, we recorded single-unit activities that responded orthodromically after stimulation of the utricular and/or saccular nerves and antidromically after stimulation of the cerebellum (uvula-nodulus and anterior vermis). Descending axonal projections to the spinal cord were also examined by antidromic stimulation of the caudal end of the C1 segment. Forty-seven otolith-activated neurons that projected to the uvula-nodulus were recorded. Thirteen (28%) of the 47 neurons received convergent inputs from the utriculus and sacculus. The remaining 34 (72%) vestibular neurons were non-convergent neurons: 18 (38%) received utricular input alone, and 16 (34%) received saccular input alone. Most (35/47) vestibulocerebellar neurons were located in the descending vestibular nucleus and only one of these projected to the spinal cord. Seven of the 47 vestibulocerebellar neurons were located in the lateral vestibular nucleus and most of these neurons projected to the spinal cord. The remaining neurons were located in group X (two neurons) and the superior vestibular nucleus (three neurons). In a different series of experiments, 37 otolith-activated vestibular neurons were tested to determine whether they projected to the uvula-nodulus and/or the anterior vermis. Nineteen of the 37 neurons projected to the anterior vermis, 13/37 projected to the uvula-nodulus, and 5/37 projected to both. The utricular and/or saccular nerve-activated vestibulocerebellar neurons projected to not only the uvulanodulus, but also to the anterior vermis. In summary, the results of this study showed that vestibular neurons receiving inputs from the utriculus and/or sacculus projected to the cerebellar cortex. This indirect otolith-cerebellar pathway terminated both in the anterior lobe and in the uvula/nodulus.     

Convergence of the anterior semicircular canal and otolith afferents on cat single vestibular neurons

The convergence between the anterior semicircular canal (AC) and utricular (UT) inputs, as well as the convergence between the AC and saccular (SAC) inputs in single vestibular neurons of decerebrated cats were investigated. Postsynaptic potentials were recorded intracellularly after selective stimulation of each pair of vestibular nerves AC/UT or AC/SAC. Neurons were recorded from the central parts of the vestibular nuclei, where the otolith afferents mainly terminate. Of a total of 105 neurons that were activated after stimulation of the AC and UT nerves, 42 received convergent inputs. Thirty-eight of these neurons received excitatory inputs from both afferents. Convergent neurons were further classified into vestibulospinal (n=28) and vestibulooculospinal (n=6) neurons by antidromic activation from the border between the C1 and C2 spinal cord and the oculomotor or trochlear nucleus. Eight neurons that were not antidromically activated from either site were classified as vestibular neurons. Forty three percent of the convergent vestibulospinal neurons and most of the convergent vestibulooculospinal neurons projected to the spinal cord through the medial vestibulospinal tract. The remaining vestibulospinal and vestibulooculospinal neurons descended through the ipsilateral lateral vestibulospinal tract. Of a total of 118 neurons that were activated after stimulation of the AC and/or SAC nerves, 51 received convergent inputs (27 vestibulospinal, 4 vestibulooculospinal, 5 vestibuloocular and 15 vestibular neurons). Forty-two of the convergent neurons received excitatory inputs from both afferents. Thirty seven percent of the convergent vestibulospinal neurons and all of the convergent vestibulooculospinal neurons projected to the spinal cord through the medial vestibulospinal tract. The remaining vestibulospinal and vestibulooculospinal neurons descended through the ipsilateral lateral vestibulospinal tract. Electronic Publication     

Response of vestibular neurons to head rotations in vertical planes. I. Response to vestibular stimulation

1. We have studied, in decerebrate cats, the responses of neurons in the lateral and descending vestibular nuclei to whole-body rotations in vertical planes that activated vertical semicircular canal and utricular receptors. Some neurons were identified as vestibulospinal by antidromic stimulation with floating electrodes placed in C4. 2. The direction of tilt that caused maximal excitation (response vector orientation) of each neuron was determined. Neuron dynamics were then studied with sinusoidal stimuli closely aligned with the response vector orientation, in the range 0.02-1 Hz. A few cells, for which we could not identify a response vector, probably had spatial-temporal convergence. 3. On the basis of dynamics, neurons were classified as receiving their input primarily from vertical semicircular canals, primarily from the otolith organs, or from canal+otolith convergence. 4. Response vector orientations of canal-driven neurons were often near +45 degrees or -45 degrees with respect to the transverse (roll) plane, suggesting these neurons received excitatory input from the ipsilateral anterior or posterior canal, respectively. Some neurons had canal-related dynamics but vector orientations near roll, presumably because they received convergent input from the ipsilateral anterior and posterior canals. Few neurons had their vectors near pitch. 5. In the lateral vestibular nucleus, neurons with otolith organ input (pure otolith or otolith+canal) tended to have vector orientations closer to roll than to pitch. In the descending nucleus the responses were evenly divided between the roll and pitch quadrants. 6. We conclude that most of our neurons have dynamics and response vector orientations that make them good candidates to participate in vestibulospinal reflexes acting on the limbs, but not those acting on the neck.     

Bone conducted vibration selectively activates irregular primary otolithic vestibular neurons in the guinea pig

The main objective of this study was to determine whether bone-conducted vibration (BCV) is equally effective in activating both semicircular canal and otolith afferents in the guinea pig or whether there is preferential activation of one of these classes of vestibular afferents. To answer this question a large number (346) of single primary vestibular neurons were recorded extracellularly in anesthetized guinea pigs and were identified by their location in the vestibular nerve and classed as regular or irregular on the basis of the variability of their spontaneous discharge. If a neuron responded to angular acceleration it was classed as a semicircular canal neuron, if it responded to maintained roll or pitch tilts it was classified as an otolith neuron. Each neuron was then tested by BCV stimuli—either clicks, continuous pure tones (200–1,500 Hz) or short tone bursts (500 Hz lasting 7 ms)—delivered by a B-71 clinical bone-conduction oscillator cemented to the guinea pig's skull. All stimulus intensities were referred to that animal's own auditory brainstem response (ABR) threshold to BCV clicks, and the maximum intensity used was within the animal's physiological range and was usually around 70 dB above BCV threshold. In addition two sensitive single axis linear accelerometers cemented to the skull gave absolute values of the stimulus acceleration in the rostro-caudal direction. The criterion for a neuron being classed as activated was an audible, stimulus-locked increase in firing rate (a 10% change was easily detectable) in response to the BCV stimulus. At the stimulus levels used in this study, semicircular canal neurons, both regular and irregular, were insensitive to BCV stimuli and very few responded: only nine of 189 semicircular canal neurons tested (4.7%) showed a detectable increase in firing in response to BCV stimuli up to the maximum 2 V peak-to-peak level we delivered to the B-71 oscillator (which produced a peak-to-peak skull acceleration of around 6–8 g and was usually around 60–70 dB above the animal's own ABR threshold for BCV clicks). Regular otolithic afferents likewise had a poor response; only 14 of 99 tested (14.1%) showed any increase in firing rate up to the maximum BCV stimulus level. However, most irregular otolithic afferents (82.8%) showed a clear increase in firing rate in response to BCV stimuli: of the 58 irregular otolith neurons tested, 48 were activated, with some being activated at very low intensities (only about 10 dB above the animal's ABR threshold to BCV clicks). Most of the activated otolith afferents were in the superior division of the vestibular nerve and were probably utricular afferents. That was confirmed by evidence using juxtacellular injection of neurobiotin near BCV activated neurons to trace their site of origin to the utricular macula. We conclude there is a very clear preference for irregular otolith afferents to be activated selectively by BCV stimuli at low stimulus levels and that BCV stimuli activate some utricular irregular afferent neurons. The BCV generates compressional and shear waves, which travel through the skull and constitute head accelerations, which are sufficient to stimulate the most sensitive otolithic receptor cells.     

Canal-otolith interaction in the fastigial nucleus of the alert monkey

To determine the contribution of the otoliths as well as the horizontal and vertical semicircular canals to the response of "vestibular only" neurons in the rostral fastigial nucleus of the alert monkey, we applied natural sinusoidal vestibular stimuli (0.6 Hz; +/-15 deg) around different axes. During the experiment the monkey sat erect in a primate chair with the head immobile. Semicircular canal responses were investigated during tilted yaw stimulation around an earth vertical axis. The tilt angle was varied by 30 deg and included the optimal plane for horizontal canal stimulation (15 deg nose down from the stereotactic plane). The otoliths and mainly the vertical canals made contributions during stimulation around an earth-fixed horizontal axis (vertical stimulation). Head orientation was also slowly altered (2-3 deg/s) over a range of 180 deg under both stimulus conditions (tilted yaw and vertical stimulation). Neuronal data for each paradigm were fitted by a least squares best-sine function. Computation of the hypothetical contributions made by all three pairs of semicircular canals and the otoliths to these responses showed that 74% of the 46 neurons investigated received an otolith input; in most instances it was combined with a canal input. Neurons most often received input from the horizontal and vertical canals as well as the otoliths. Only a minority of neurons received a purely otolith (13%), vertical canal (13%), or horizontal canal (4%) input. Conventional criteria (head position-related activity, spatiotemporal convergence, STC) failed to detect an otolith contribution in several such instances. Thus, canal-otolith convergence is the general rule at this central stage of vestibular information processing in the fastigial nucleus. The large variety of response types allows these neurons to participate in multiple tasks of vestibulospinal movement control.     

Properties of utricular nerve-activated vestibulospinal neurons in cats

The axonal pathway, conduction velocities, and locations of the cell bodies of utricular nerve-activated vestibulospinal neurons were studied in decerebrated or anesthetized cats using the collision test of orthodromic and antidromic spikes. For orthodromic stimulation, bipolar tungsten electrodes were placed on the utricular nerve and the other vestibular nerve branches were transected. Monopolar tungsten electrodes were positioned on both sides of the upper cervical segments (C2–4), caudal end of the cervical enlargement (C7-T1), and from the lower thoracic to the upper lumbar segments (T12-L3) and were used for antidromic stimulation of the spinal cord. Another monopolar electrode was also placed in the oculomotor nucleus to study whether utricular nerve-activated vestibulospinal neurons have ascending branches to the oculomotor nucleus. Of the 173 vestibular neurons orthodromically activated by the stimulation of the utricular nerve, 46 were second-order vestibulospinal neurons and 5 were third-order neurons. The majority of the utricular nerve-activated vestibulospinal neurons were located in the rostral part of the descending vestibular nucleus and the caudal part of the ventral lateral nucleus. Seventy-three percent of the utricular nerve-activated vestibulospinal neurons descended through the ipsilateral lateral vestibulospinal tract. Approximately 80% of these neurons reached the cervicothoracic junction, but a few reached the upper lumbar spinal cord. Twenty-seven percent of the utricular nerve-activated vestibulospinal neurons descended through the medial vestibulospinal tract or the contralateral vestibulospinal tracts. Those axons terminated mainly in the upper cervical segments. Almost none of the utricular nerve-activated vestibular neurons had ascending branches to the oculomotor nucleus.     

Tonic influence of the efferent vestibular system on the spontaneous afferent activity from semicircular canals in the frog (Rana esculenta L.)

Summary In the frog we have recorded the spontaneous activity of single afferent fibres of the ampullary nerves of the left horizontal semicircular canal (HC) and vertical anterior canal (VAC) in isolated head preparations. The recordings have been made in 4 experimental situations: intact preparations; preparations whose brain was destroyed; preparations whose contralateral vestibular nerve had been cut between Scarpa's ganglion and the brain; preparations where either an ampullary nerve (that of the HC or of the VAC) or the utricular nerve had been cut contralaterally. From our results, it appears that the efferent vestibular system has a tonic influence on the afferent activity from HC and VAC; the influence of the part of the efferent vestibular system (EVS) activity depending on contralateral vestibular inputs is inhibitory, while the influence of the part of the EVS activity depending on ipsilateral vestibular inputs might be facilitatory.     

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

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.     

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.     

Basic organization principles of the VOR: lessons from frogs

Locomotion is associated with a number of optical consequences that degrade visual information processing in the absence of appropriate compensatory movements. The resulting retinal image flow is counteracted by coordinated eye-head reflexes that are initiated by optokinetic and vestibular inputs. The contribution of the vestibulo-ocular reflex (VOR) for stabilizing retinal images is relatively small in amplitude in frogs but important in function by compensating for the non-linearities of the neck motor system. The spatial tuning of the VOR networks underlying the angular (AVOR) and linear (LVOR) with respect to canal and extraocular motor coordinates is organized in a common, canal-related reference frame. Thereby, the axes of head and eye rotation are aligned, principle and auxiliary VOR connections transform vestibular into motor signals and parallel AVOR and LVOR circuits mediate vergence and version signals separately. Comparison of these results with data from other vertebrates demonstrates a number of fundamental organization principles common to most vertebrates. However, the fewer degrees of behavioral freedom of frogs are reflected by the absence of, e.g. a functioning velocity storage network or of a fixation suppression of the VOR. In vitro experiments with the isolated brainstem and branches of N.VIII attached were used to study the putative transmitters of vestibular nerve afferent inputs, the postsynaptic receptor subtypes of second-order vestibular neurons and their dynamic response properties. Evidence is presented that suggests that afferent vestibular nerve fibers with different dynamic response properties activate different subtypes of glutamate receptors. The convergence pattern of monosynaptic afferent nerve inputs from different labyrinthine organs onto second-order vestibular neurons is remarkably specific. As a rule, second-order vestibular neurons receive converging afferent nerve inputs from one semicircular canal and from a specific sector of hair cells on one otolith organ. This convergence pattern remains malleable even in adulthood and reorganization is initiated by activity-related changes in vestibular nerve afferent fibers. The output of second-order vestibular neurons is modified by at least three inhibitory control loops. Uncrossed inhibitory vestibular side loops appear to control specifically the dynamic response tuning, whereas coplanar commissural inhibitory inputs improve mainly the spatial tuning and the cerebellar feedback loop controls the response gain. Among the targets of second-order vestibular projection neurons are extraocular motoneurons and internuclear neurons. Extraocular motoneurons differ among each other by the presence of very different response dynamics. These differences may represent a co-adaptation to the response dynamics of twitch and non-twitch extraocular muscle fibers. Different dynamical properties are required for a rapid acceleration of the globe at the one end and for the maintenance of a stable eccentric eye position over long periods of time at the other end of a continuum of variations in dynamic response properties. The maintenance of a given eccentric eye position over long periods of time is especially well developed in frogs and assists visual surveillance during lurking in the absence of saccades.