| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Laboratoire de Neurobiologie des Réseaux Sensorimoteurs, Centre National de la Recherche Scientifique UMR 7060, Université Paris 5, Centre Universitaire des Saints-Pères, 75270 Paris Cédex 06, France 2 The Rockefeller University, New York, New York 10021
Submitted 3 March 2003; accepted in final form 8 May 2003
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
) before transmitting it to their efferent target cells. To understand the physiology of central, vestibular-related pathways, several groups have therefore performed in vitro intracellular recordings of neurons in the vestibular nuclei of rats (Gallagher et al. 1985; Johnston et al. 1994), mice (Dutia and Johnston 1998), guinea pigs (Serafin et al. 1991a,b), or chicks (Du Lac and Lisberger 1995a) using brain stem slices. In addition Babalian et al. (1997) characterized the membrane properties of identified, second-order vestibular neurons recorded in the isolated whole brain of guinea pig. These recordings were all done in the medial vestibular nucleus (MVN), where the density of neurons is the highest and which is easier to locate on slices than the other vestibular subnuclei (for reviews see Darlington et al. 1995; Vidal et al. 1999).
In all species studied so far, these recordings have led to the identification of two main types of MVN neurons (MVNn), the type A and type B neurons, which differ in their action potential profile and membrane properties. It is generally admitted that MVNn represent a continuum of cells whose properties are distributed between those of type A and type B MVNn (Babalian et al. 1997; Beraneck et al. 2003; Du Lac and Lisberger 1995a; Johnston et al. 1994). In the guinea pig (Serafin et al. 1991a), as in other species, type A neurons are characterized by their single, deep after-hyperpolarization (AHP), and have on average a wider action potential than that of type B neurons. They display an A-like rectification when released from hyperpolarization, or in response to depolarizing current pulses given from a hyperpolarized level. This rectification is also visible as an inflection point delaying the depolarization of the neuron during interspike intervals. Interestingly, this A-like rectification is not blocked by 4-aminopyridine (4-AP) and is therefore not mediated by the rectifying A potassium conductance described in other neurons of the CNS (Hagiwara et al. 1961; Llinás 1988). Type B neurons have a biphasic, significantly smaller AHP, no sizable A-like rectification, and narrower action potentials. Some of the type B neurons, the B + LTS MVNn, display low-threshold calcium spikes when released from a strong hyperpolarization. We assume that a large majority (>80%) of the MVNn recorded in guinea pig slices are second-order vestibular neurons. Indeed, 80 to 85% of the neurons recorded in the MVN area of the isolated whole brain of guinea pig, using similar electrodes to those used in slices, could be identified as second-order vestibular neurons (Babalian et al. 1997). This high proportion of second-order cells is in agreement with anatomic (Carleton and Carpenter 1983; Sato and Sasaki 1993) and physiological studies (Chen-Huang et al. 1997; Goldberg et al. 1987).
Extracellular recordings of neurons in the other vestibular subnuclei—the superior, lateral, and inferior (or descending) vestibular nuclei— have been obtained in vivo, and recently in rat brain stem slices in vitro (Sun et al. 2002). However, to our knowledge, in vitro intracellular recordings of central vestibular neurons outside the MVN have never been performed. We have therefore decided to undertake intracellular recordings of neurons in the lateral vestibular nucleus (LVN) to determine their membrane properties and to see to what extent the results obtained in the MVN generalize to another group of vestibular neurons.
Anatomic and in vivo electrophysiological studies have shown that the afferent and efferent connections, as well as the functional role of LVN neurons (LVNn), were distinct from those of MVNn (for reviews see Carleton and Carpenter 1983; Uchino and Isu 1992; Wilson and Melvill-Jones 1979; Wilson and Peterson 1988). The MVN receives extensive canal and otolith input (Carleton and Carpenter 1984; Gacek 1969; Gstoettner et al. 1992; Walberg et al. 1958) and is strongly involved in relaying afferent signals coming from the horizontal semicircular canals (Sato and Sasaki 1993; Sato et al. 1989; Schaefer and Meyer 1992; Uchino and Isu 1992). The LVN also appears to receive input from both the otolith receptors and the semicircular canals (Boyle and Pompeiano 1980; Gstoettner et al. 1992; Wilson and Schor 1999). Input to the LVN includes strong signals from the vertical semicircular canals (Kasper et al. 1988), but the LVN would be less involved in relaying horizontal semicircular canal signals (Uchino and Isu 1992). Whereas the whole extent of the MVN is innervated by sensory nerve fibers coming from the ipsilateral labyrinth, the dorsal part of the LVN does not receive direct, monosynaptic vestibular input (for reviews see Carleton and Carpenter 1983; Gacek 1969; Korte 1979; Walberg et al. 1958). Another important difference is that extensive inhibitory commissural connections exist between the two MVN (Carleton and Carpenter 1983; Shimazu and Precht 1966), whereas there are no or only very few commissural connections at the level of the lateral vestibular nuclei (Epema et al. 1988; Uchino et al. 2001).
Axons from MVNn project both to various oculomotor nuclei through the medial longitudinal fasciculus and to the upper segments of the spinal cord through the medial vestibulospinal tract (Wilson and Melvill-Jones 1979). In contrast, the LVNn have mainly vestibulospinal projections through the lateral vestibulospinal tract, which reach all levels of the spinal cord (for reviews see Carleton and Carpenter 1983; Sarkisian 2000; Wilson and Melvill-Jones 1979). Whether the ascending tract of Deiters, which is involved in the control of eye movements, originates in the LVN is still a matter of controversy (see DISCUSSION). As well as projecting to the spinal cord, both the MVN and LVN receive numerous proprioceptive afferents from the neck, trunk, and limbs (Bankoul et al. 1995; Jian et al. 2002; Xiong and Matsushita 2001). Altogether, the MVN would be involved in both eye and head control, particularly in the horizontal plane. In contrast, the LVN appears to be mainly involved in the control of posture, particularly in response to linear displacements or gravito-inertial force variations including roll-and-pitch movements in the vertical plane.
Are the different connectivity and functional roles of the LVN and MVN matched by differences in the static and dynamic intrinsic, biophysical properties of their neurons? To answer that question, we determined some membrane properties and dynamic responses to various types of current injections of intracellularly recorded LVNn in guinea pig brain stem slices. The results were compared with those obtained for a set of 89 MVNn recorded under the same conditions (Beraneck et al. 2003). The data from the LVNn and MVNn were collected and analyzed using identical protocols (Ris et al. 2001).
|
|
METHODS |
|---|
|
Experiments were carried out on young adult, pigmented guinea pigs of both genders weighing 150–350 g (Elevage de la Garenne, Saint-Pierre d'Exideuil, France). The animals were handled in accordance with the European Communities Council Directive of November 24, 1986, and following the procedures issued by the French Ministère de l'Agriculture.
After sodium pentobarbital anesthesia (50 mg/kg), the animals were killed by decapitation, and thick (400–500 µm), coronal brain stem slices containing the LVN were cut and maintained using standard techniques (Serafin et al. 1991a; Vibert et al. 1999b). However, in most experiments, a cold (4°C) sucrose-containing artificial cerebrospinal fluid (sucrose ACSF) was used instead of our standard ACSF during the dissection and preparation of the slices. The slices containing the LVN were then transferred to individually oxygenated vials and incubated in the sucrose ACSF at room temperature for 15 min. Then, we replaced two thirds of the liquid in each incubation vial with the same volume of normal ACSF. This operation was repeated twice more at 10-min intervals. After this procedure, the slices were kept at room temperature, in what was almost exclusively normal ACSF, for at least 1 h until they were transferred to the recording chamber. During recordings, the slices were superfused as usual with normal ACSF maintained at 31–32°C (Serafin et al. 1991a; Vibert et al. 1999b). Compared with the normal ACSF (composition in mM: NaCl 124, NaHCO3 26, KCl 5, KH2PO4 1.25, MgSO4 1.3, CaCl2 2.4, glucose 10), the sucrose ACSF was obtained by replacing NaCl with an equimolar concentration of sucrose (i.e., 124 mM), in accordance with the method used by Devor et al. (2001). Use of the sucrose solution during the preparation of the slices was extremely beneficial for the viability of LVNn. The mechanism of this action of sucrose is not clearly understood, and the concentration of sucrose added in the ACSF to replace NaCl varies from preparation to preparation (Aghajanian and Rasmussen 1989; Moyer and Brown 1998; Rothman 1985; for review see Aitken et al. 1995). However, it is generally assumed that removal of NaCl from the ACSF reduces the sodium influx coincident with the tendency for neurons to depolarize during the initial and unavoidable anoxia that occurs with decapitation, dissection, and slicing. Furthermore, it minimizes excitotoxic cell swelling that results from passive chloride influx followed by cation and water entry.
As mentioned in the INTRODUCTION, data obtained from LVNn were compared with those obtained on a sample of previously recorded MVNn, obtained without using sucrose ACSF (Beraneck et al. 2003; Ris et al. 2001). Even though the LVNn were ultimately recorded in the same normal ACSF as MVNn, the question arises whether the use of sucrose ACSF during the dissection and first 35 min of incubation of the slices induced significant modifications of their membrane properties. Because we obtained a few LVNn (5 out of 42) without using sucrose ACSF, we verified that neither the static nor the dynamic membrane properties were different from those of the LVNn obtained using the sucrose ACSF. Although the use of sucrose ACSF led to a clear increase in the number of successful penetrations of LVNn, it did not modify the spontaneous discharge rate, resting membrane potential, or other membrane properties of the neurons.
Electrophysiological recordings
In a few preliminary experiments, LVNn were recorded from extracellularly, using 2 M NaCl-containing glass microelectrodes (resistance: 2–8 M
). During these extracellular recordings, many silent cells were visible during the descent of the electrode within the tissue. Indeed, the pressure of the electrode triggered firing in silent cells. When we saw well-isolated spikes triggered by electrode movements that did not persist when the electrode remained still, we recorded examples of these spikes and considered that the corresponding neuron had no spontaneous discharge and was silent at rest. In subsequent experiments, intracellular electrophysiological recordings were obtained with sharp, 3 M potassium acetate–containing glass microelectrodes from neurons within the LVN. We identified this nucleus using landmarks such as the shape of the 4th ventricle, the acoustic stria, and the inferior cerebellar peduncle. Intracellular injections of neurobiotin were performed for 12 out of the 42 neurons we studied (29%), to confirm that they were localized within the LVN boundaries (see following text). All measurements were done with an Axoclamp 2A system (Axon Instruments, Union City, CA) in either the bridge or switching discontinuous current clamp (DCC) mode (Moore et al. 1993). The electrode resistance varied from 80 to 150 M. Both series resistance (bridge balance) and capacitance compensation were checked throughout the recording of each individual neuron (Ris et al. 2001). Part of the current injections and all data acquisition were performed with a PC-compatible computer using the "Acquis 1" program (version 4.0; Bio-logic S.A., Gif-sur-Yvette, France). The sampling rate used for acquisition varied between 2,000 and 5,000 Hz, depending on the length of the data acquisition sequence. Consequently, the amplitudes of the digitized spikes were variable. However, oscilloscope traces verified that the size of the action potential was constant at any given membrane potential. The data were analyzed using program scripts working under Mathematica 4.0 (Wolfram Research, Champaign, IL), or MATLAB 6.5 (The MathWorks, Natick, MA). The scripts used to quantify the properties and responses of LVNn were exactly the same as those used to analyze the data obtained from MVNn (Beraneck et al. 2003; Ris et al. 2001).
Basic membrane and firing properties of LVNn
Because the majority of LVNn are spontaneously active in slices, the membrane potential was filtered with a 1-Hz low-pass filter to obtain an estimate of its average "resting" level (i.e., of the mean level of its membrane potential in the absence of any current injection through the electrode). This was taken as the "mean resting membrane potential" of each neuron. For each cell this membrane potential value was corrected by measuring and subtracting the extracellular voltage offset found after withdrawal of the electrode from the neuron. No correction was made for the liquid junction potential, but this can be assumed to be constant between LVNn and MVNn because they were recorded using similar electrodes. Cells that had resting membrane potentials more negative than –50 mV and spike amplitudes higher than 50 mV were retained for further study. Six neurons that displayed smaller spike height (between 45 and 50 mV), but did not show any sign of deterioration (they exhibited normal membrane potentials and spike widths), were also included in the sample.
Recordings of the neurons at rest (i.e., with no holding current being injected through the recording electrode) were used to calculate mean spontaneous firing rate, its coefficient of variation (CV) expressed as a percentage, and to measure the mean amplitude of the spike. For each neuron, an average of the spike shape and the following interspike interval profile was obtained by averaging successive spontaneous spikes taken either at the resting membrane potential or while the cell was slightly depolarized (for the neurons that were silent at rest). The spikes (mean number about 90) were synchronized to their thresholds, taken at the point on the rising phase of the action potential where the slope of the potential trace reached an arbitrary threshold of 10 V/s (Krawitz et al. 2001). The averaged spike shape was used to determine the amplitude of the after-hyperpolarization (AHP) and the width of the spike (taken at threshold). The AHP amplitude was calculated as the membrane potential difference between the spike threshold and the membrane potential minimum after the falling phase of the spike (Fig. 2A2).
|
The cell's firing threshold, that is, the membrane potential at which the cell begins to fire action potentials (in mV), was assessed as the potential reached at the threshold of the first spike triggered by a slow, depolarizing current ramp (see Injection of depolarizing ramplike currents). The threshold of this first spike was measured in the same way as other spike thresholds; that is, it was taken at the point on the rising phase of the action potential where the slope of the potential trace reached an arbitrary threshold of 10 V/s. For each cell, we determined whether long-lasting, subthreshold plateau potentials could be triggered by low-amplitude (0.1–0.2 nA), short-duration (10 ms) current pulses delivered while the neuron was maintained just below its firing threshold (Babalian et al. 1997; Serafin et al. 1991a,b). If so, we measured their mean duration (Fig. 2A4).
For a few representative LVNn (see RESULTS below), some of the ionic conductances shaping the spike and AHP profile were assessed by pharmacological studies. The following drugs, obtained from Sigma-Aldrich (Saint-Quentin Fallavier, France), were applied in the perfusion ACSF at known concentrations. Tetraethylammonium (TEA) was used at 5–10 mM, 4-AP at 1–2 mM, and apamin at 0.1–0.2 µM.
Quantitative analysis of the spike shape of central vestibular neurons
As presented in the INTRODUCTION, MVNn were first categorized into type A and type B using qualitative criteria. MVNn were characterized as a type A, type B, or type B with low-threshold calcium spikes (B + LTS) neurons according to their action potential profile (Serafin et al. 1991a). Neurons that displayed intermediate properties, or did not clearly fit into any of these two categories, were grouped as type C MVNn.
To remove possible biases in the classification of MVNn, we recently developed quantitative criteria to characterize them (Beraneck et al. 2003). For each neuron, the averaged spike profile obtained during spontaneous firing and its first derivative were used [as already done by Johnston et al. (1994)] to assess the presence of the A-like rectification and double AHP, the two main criteria used previously for the qualitative classification.
The presence of an A-like rectification characterizing type A MVNn is visible as an inflection of the voltage trace (V) within the interspike interval (see Fig. 2A). This inflection is better seen on the first derivative of the voltage trace (dV/dt) as a transient decrease of the rate of the depolarization leading to the next spike. In MVNn, the A-like rectification begins 
2–3 ms after the end of the spike whatever the spontaneous discharge rate of the cell, and the derivative of the voltage trace remains positive. The strength of the A-like rectification was quantified as the algebraic decrease 
(dV/dt) in the rate of the interspike depolarization (dV/dt in V/s) associated with this phenomenon (Fig. 2A3). In the absence of any A-like rectification, this parameter was set at zero. Figure 2B demonstrates that the presence of an A-like rectification was not dependent on the rate of decay of the double-component AHP.
The presence of an early fast AHP followed by a delayed slow one (i.e., of the double-component AHP characterizing type B MVNn) was assessed on the averaged spike profile, and then confirmed using the first derivative (see Fig. 2A for the following description). Figure A3 shows the first derivative of the A2 voltage trace (dV/dt in V/s). The strongly negative (< –1 V/s), initial parts of the derivative, which correspond to the falling phase of the spike, are not to scale. When a double AHP is present, its second component is seen as a transient decrease of the rate of depolarization, which occurs within 2 ms of the end of the spike. During this period, the first derivative becomes transiently negative. The strength of the double AHP was quantified as the algebraic decrease (dV/dt) in the rate of the interspike depolarization (dV/dt in V/s) associated with the second component of the phenomenon. As shown on Fig. 2A3, this algebraic decrease is measured between the positive peak of the derivative, reached while the potential trace increases after the first component of the double AHP, and the minimum of the derivative reached during the second, transient hyperpolarization of the membrane potential. The strength of the double AHP was set at zero when no double AHP was present.
Quantitative values describing the strength of the A-like rectification and the strength of the double AHP were measured for the set of 89 MVNn recorded by Beraneck et al. (2003). Quantitative criteria for classification of MVNn in type A, type B, or type C neurons were designed (Beraneck et al. 2003). The MVNn displaying an A-like rectification lower than 0.15 V/s (50%) were classified as type B MVNn. The MVNn displaying an A-like rectification stronger than 0.15 V/s and no double AHP (47%) were classified as type A MVNn. The 3 MVNn (3%) that displayed both a double AHP and an A-like rectification greater than 0.15 V/s, which could not be unambiguously categorized as either type A or type B MVNn, were therefore considered as type C MVNn. To compare the spike profile of LVNn with those of type A and type B MVNn, the presence and strength of an A-like rectification and/or double AHP were assessed in the same way, using the very same Matlab scripts.
After the assessment of its basic membrane and firing properties, each LVNn was submitted to the stereotyped stimulation protocol described below. The instantaneous firing rate of the cell (IF in spikes/s) was estimated at any time with a Mathematica script that measured the time intervals between two successive action potentials. The time at the end of each interval between action potentials was used to indicate the time for each IF value.
Assessment of the passive input resistance of LVNn using current steps
The passive input resistance of each neuron was assessed using series of hyperpolarizing current steps of decreasing amplitudes. The cell was maintained by steady-state hyperpolarization at a few millivolts (mV = 0 to 10) below its threshold for discharge, to suppress spikes. Although the responses were multiexponential, their principal component could be approximated by a single exponential that provided an estimate of the whole cell resistance for each LVNn (input resistance = voltage deflection/current input).
Injection of depolarizing ramplike currents
Ramp currents of 0.3 nA amplitude were applied at five different slopes up to a final steady-state value, as described in detail in Ris et al. (2001), while the cell was maintained at about 10 mV below its firing threshold (Fig. 4A). In other words, the membrane potential at which the ramps were delivered was set relative to the firing threshold of each cell. The five slopes corresponded to times to reach the plateau of current of 5,000 ms (0.06 nA/s), 3,400 ms (0.09 nA/s), 1,800 ms (0.17 nA/s), 600 ms (0.5 nA/s), and 200 ms (1.5 nA/s), respectively. Because the whole stimulus was 5,000 ms long, there was no plateau after the slowest ramp, which was used, as stated above, to assess the cell's firing threshold. For each ramp we computed the rate of increase of the instantaneous firing rate of the cell (kIF in spikes s–1 nA–1), which gives an indication of the sensitivity of the cell to current injections. We also measured in each case the difference between the firing rate reached at the end of the depolarizing current injection and the stable discharge rate reached at the end of the plateau (overshoot in spikes/s). This parameter gives an indication of the nonlinear, dynamic properties of neurons. To assess how the level of polarization of LVNn influenced their responses, the whole sequence of ramp stimulations was repeated while the neuron was at its resting membrane potential (or held slightly above its firing threshold if the neuron was silent at rest) and spontaneously fired action potentials.
|
Injection of sinusoidal currents
A third series of stimuli consisted of current sine waves applied for 5,000 ms at various frequencies ranging from 0.2 to 50 Hz (Du Lac and Lisberger 1995b; for details see Ris et al. 2001). The amplitude of the stimulus (I) was adjusted at the 0.2-Hz frequency to keep the membrane potential variation between 5 and 10 mV peak-to-peak, and was typically within the 0.05- to 0.15-nA range. The first series of sinusoidal currents was delivered while the cell was at its resting membrane potential (or held slightly above its firing threshold if the neuron was silent at rest) and spontaneously fired action potentials (Fig. 5B). For each frequency of stimulation inferior or equal to about one third of the neuron's background discharge, the modulation of the IF rate of MVNn was fitted with a sine wave that was then used to calculate the amplitude and the phase of the IF modulation (IF, Fig. 5B). Ris et al. (2001) have shown that in this condition, the IF modulation of MVNn was linear. When the frequency of stimulation passed a third of the neuron's background discharge, it was inappropriate to continue fitting the modulation of the IF rate of MVNn with a sine wave because of sampling limitations. The amplitude of the IF modulation of MVNn was calculated in an empirical way as the difference between the minimum and maximum IF reached by the neuron during the 5 s of the stimulation. By definition this difference represents the raw peak-to-peak amplitude of the IF modulation induced by the injection of sinusoidal current, and eventually reaches a minimum as the frequency of stimulation approaches the background discharge rate. No phase measurements were obtained in this situation. Using this method, we could evaluate IF from 0.2 Hz to a maximum stimulus frequency that varied from cell to cell according to its background discharge and the sensitivity of its discharge to current injection, but could reach 50 Hz in some cases.
|
The underlying mean membrane potential excursion (V) was computed for each stimulus frequency using a Mathematica script, which performed a Fourier analysis of the total membrane potential response. The magnitude of the Fourier component corresponding to the stimulation frequency was taken as the potential response. However, this procedure was valid only when the components attributed to the shape and frequency of the action potentials were not overlapping those of the stimulation frequency. This requirement was true for frequencies below 1 Hz (Fig. 5B). IF and I were used to evaluate at 0.4 Hz the cell sensitivity to current injection by dividing IF by the amplitude of the injected current (IF/I in spikes s–1 nA–1; Table 2). The sensitivity of the firing rate of the cell to variations of the mean membrane potential IF/V was quantified in spikes s–1 mV–1. We calculated the "active" impedance Z of the cell as the amplitude of the membrane potential change obtained for the 0.4-Hz stimulus divided by the amplitude of the injected current (V/I in M).
|
When possible, a similar series of sinusoidal stimuli was given while the cell was maintained at a depolarized membrane potential by a steady-state current injection of 0.15–0.25 nA, to assess how the level of discharge of MVNn modified their responses.
Some of the cells were also submitted to the same series of sinusoidal current injections while they were maintained at 10 to 20 mV below their threshold for discharge, so that no spike was evoked by the stimulation (Fig. 5A). The amplitude of the membrane potential change (Vh) was computed for each frequency using a Mathematica script, and was used to evaluate at each frequency the impedance Zh of the cell maintained under a steady-state hyperpolarization (Zh = Vh/I in M).
As reported by Ris et al. (2001), the amplitude of the modulation of the membrane potential or instantaneous firing rate of MVNn by sinusoidal currents displayed resonant properties. For each MVNn, the response increased with increasing stimulation frequency to reach a maximum at what was defined as the peak frequency of resonance (R-peak). Then, the modulation progressively dropped to lower levels. The "amplitude" of the resonance (R-ratio) was defined as the ratio between the maximum amplitude of the firing rate modulation at the peak frequency of resonance and the amplitude obtained at the lowest frequency we used, 0.2 Hz. The resonant properties of LVNn were also assessed by another index: the ratio between the amplitudes of the firing rate modulation obtained at 4 Hz and 0.4 Hz (IF 4/0.4). The two ratios that characterized the resonance of LVNn were measured in the same way for the membrane potential when the neurons were hyperpolarized to suppress action potentials.
Histological procedures
In about one third of the experiments, neurobiotin tracer (Vector Laboratories, Burlingame, CA) was dissolved at 2% in the potassium acetate used to fill the intracellular electrodes. At the end of each experiment, neurobiotin was injected into an intracellularly recorded LVNn by passing depolarizing, rectangular current pulses of 1.0 nA amplitude and 200 ms duration at 2.5 Hz, for 20 min. Brain stem slices were then fixed by submersion in 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) overnight. After several rinses with phosphate-buffered saline (PBS), sections were treated with 0.5% H2O2 in PBS for 2 h, rinsed 3 times with PBS, and then incubated overnight with 1/50 Vectorstain ABC Elite kit (Vector Laboratories) in PBS containing 0.2% Triton X-100. The neurobiotin-injected neurons were visualized by reaction with 0.05% diaminobenzidine (Sigma) in phosphate buffer for 10 min, and by further addition of 0.05% H2O2 for 5–10 min. The sections were mounted onto slides, dried, and coverslipped.
To precisely assess the boundaries of the LVN, two brains were taken from guinea pigs of the same age/weight as the animals we normally used. Transverse cryostat sections of 14 µm thickness were obtained from the brain stem at the level of the LVN, and colored with either cresyl violet or toluidine blue. The anatomic location of each of the neurons labeled with neurobiotin was assessed using this homemade "atlas" of the LVN. The borders between the different subnuclei of the vestibular nuclear complex were defined according to the cytoarchitectonic studies performed in the guinea pig by Gstoettner and Burian (1987) and Ris et al. (1999) (also see DISCUSSION). The size of each neurobiotin-labeled cell was assessed by measuring the maximum diameter of the soma along its long, major axis, and the minimum width of the soma along its short, minor axis. The mean soma diameter was then obtained by averaging these two values. Depending on its diameter, each cell was considered as either small (10–20 µm), medium (21–35 µm), large (36–60 µm), or giant (50–80 µm) (Gstoettner and Burian 1987).
Statistical analysis
Calculations of means ± SDs and further processing of all results were carried out using the Systat 8.0 software (SPSS, Chicago, IL) on a PC-compatible computer. The data obtained from LVNn were compared with those obtained for the set of 89 MVNn recorded in the same conditions by Beraneck et al. (2003). Statistical comparisons between numerical values were achieved through nonparametric tests, with the threshold for significance set at P 
0.05. Kruskal–Wallis ANOVA was first performed to search for significant differences between the mean values obtained for LVNn with respect to type A and type B MVNn (which defined 3 categories of neurons). Two by two comparisons between the 3 cell groups were then performed using Mann–Whitney U tests. Paired nonparametric tests (Friedman ANOVA followed by Wilcoxon signed-rank tests) were used to compare for each cell type the responses evoked by ramps of different slopes. They were also used to determine how the responses to ramps and sinusoidal currents were modified according to the level of steady-state polarization of the cell (2 levels for the ramps and 3 levels for the sinusoidal currents).
|
|
RESULTS |
|---|
|
; Ris et al. 2001; see METHODS). Values are means ± SD. Histological identification of the recorded cells
Because in slices the boundaries of the LVN are less easy to assess than those of the MVN, we injected 12 of the 42 neurons we recorded with 2% neurobiotin to check whether they were located in the LVN. All injected neurons were located inside the boundaries of the LVN, as shown in Fig. 1A. The ventro-medial portion of the most caudal part of the LVN (vL in Fig. 1A1), from which the ascending tract of Deiters originates, is a controversial area that has been considered as part of either the LVN or MVN, depending on the authors (see DISCUSSION). However, there were no injected neurons in this area; they were all located within the main body of the LVN, clearly separated from MVN by the acoustic stria. The other, noninjected neurons were all obtained from the area where the injected neurons had been recorded, using landmarks such as the shape of the 4th ventricle, the acoustic stria, and the inferior cerebellar peduncle. The soma diameter of the LVNn injected with neurobiotin ranged from 45.0 to 74.1 µm around a mean of 61.2 ± 10.3 µm, which means that they were all large or giant cells (Fig. 1B). They were significantly larger than MVNn, which had a mean soma diameter of 31.4 ± 7.9 µm (P = 0.004, Fig. 1C).
|
Characterization of LVNn according to their firing and membrane properties (Table 1)
Twenty-nine out of the 42 intracellularly recorded LVNn (69%) spontaneously fired action potentials at rest (i.e., when no current was injected into the cell). Altogether, the spontaneous firing rate of LVNn varied from 0 to 64.7 spikes/s around a mean of 16.5 ± 18.9 spikes/s and a median of 10.6 spikes/s (including silent cells; second row of Table 1). When silent cells were not considered, the mean discharge of LVNn reached 23.9 ± 18.4 spikes/s for a median of 17.8 spikes/s (third row of Table 1). Extracellular recordings of LVNn gave quite similar results, given that 20 out of the 30 cells (67%) were spontaneously firing at rest. Altogether, the discharge rate of the 30 extracellularly recorded LVNn ranged from 0 to 66.0 spikes/s, with an average value of 14.8 ± 18.4 spikes/s and a median of 6.0 spikes/s (including silent cells). Compared with intracellularly recorded MVNn, the proportion of intracellularly recorded LVNn that were silent at rest was much higher (31 vs. 7%). In accordance with this difference, the spontaneous firing rate of LVNn was significantly lower than that of both types of MVNn when silent cells were included (second row of Table 1). When silent cells were not considered, the spontaneous firing rate of LVNn still tended to be lower than that of MVNn, but this trend did not reach significance (P = 0.14; third row of Table 1).
|
The firing rate of spontaneously active LVNn tended to be more irregular than in MVNn. The coefficient of variation (CV) of LVNn was not significantly different from the CV of type B MVNn, but was significantly higher than the CV of type A MVNn (Table 1, Fig. 2, A1 and D). There was a significant, negative relationship between the CV and spontaneous firing rate of LVNn (correlation coefficient r = –0.43, P = 0.02). In spite of their lower spontaneous firing rate, the average resting membrane potential of LVNn was –59.4 ± 4.6 mV and was not significantly different from the resting membrane potential of either type of MVNn (Table 1). In accordance with this discrepancy, LVNn had a more depolarized firing threshold than that of MVNn. On average, the firing threshold of LVNn (i.e., the membrane potential at which they begin to fire action potentials spontaneously) was only slightly more hyperpolarized than their resting membrane potential.
When looking at the shape of the action potential and following AHP, the LVNn we recorded formed a more homogeneous population than MVNn did. Forty out of 42 (95%) displayed a very sharp spike, with a width at threshold <1 ms. Although the amplitudes of LVNn and MVNn spikes were similar, the spikes of LVNn were significantly narrower than the spikes of either type A or type B MVNn with a mean width of 0.89 ± 0.10 ms (Table 1). Ninety percent of the LVNn displayed during spontaneous firing a double-component AHP similar to the one observed in type B MVNn (Table 1, Fig. 2, A and D). In accordance with these results, the overall amplitude of the AHP of LVNn was comparable to the value obtained for type B MVNn, and lower than in type A MVNn (Table 1). About one quarter of LVNn (10 out of 42) displayed an A-like rectification stronger than 0.15 V/s, which was the criterion used to define the type A MVNn (Fig. 2, A, B, and D). Altogether, 25 of the LVNn (60%) showed a measurable A-like rectification ranging from 0.05 to 0.33 V/s, and there was a continuous distribution of the neurons in terms of the strength of their A-like rectification. The average strength of the A-like rectification of LVNn reached 0.09 ± 0.10 V/s (Table 1) and was much lower than in type A MVNn (P < 0.001), but also significantly higher than in type B MVNn (P < 0.001). Figure 2B shows that the presence and amplitude of the A-like rectification was independent of the rate of decay of the double-component AHP.
Although 10% of type B MVNn displayed low-threshold calcium spikes (LTS) and could be classified as B + LTS MVNn, we could not find any LTS in LVNn. However, most LVNn (30 out of 35, 86%) displayed subthreshold plateau potentials after short-duration, small depolarizing current pulses delivered from just below the firing threshold (Fig. 2C). This type of plateau potential was also found in 91% of type B MVNn, and was shown to result from the activation of tetrodotoxin-sensitive, persistent sodium currents (Serafin et al. 1991b). The duration of the plateau potentials triggered in LVNn varied from 0 to 102.2 ms around a mean of 36.4 ± 39.5 ms and was comparable to what was observed in type B (Table 1).
The passive input resistance of LVNn was estimated from their response to hyperpolarizing current pulses delivered from a slightly hyperpolarized potential in the absence of spontaneous firing. The resistance of the LVNn ranged from 31.2 to 106.1 M, around an average value of 58.8 ± 18.4 M, and was significantly lower than the input resistance of either type of MVNn (Table 1).
Altogether, the 42 LVNn we recorded constituted a homogeneous population of cells displaying very sharp spikes, strong double-component AHP, and often a weak A-like rectification.
Ionic conductance contributing to spike repolarization and AHP
Blockers of different potassium channels were applied to six representative LVNn to assess their respective contribution in shaping the spike and AHP profile, as has already been done for MVNn (Johnston et al. 1994; Serafin et al. 1991b). 4-AP, which antagonizes the voltage-sensitive A potassium conductance (KA) and the delayed rectifier (KV), delayed the repolarization of the action potential. Consequently, the amplitude and width of the action potential were increased, whereas the first, early component of the double-AHP disappeared (Fig. 3A). As in type A MVNn (see INTRODUCTION), 4-AP did not modify the A-like rectification observed in some LVNn during the interspike interval (not shown). Tetraethylammonium, which antagonizes high- and small-conductance Ca2+-sensitive K+ channels (BK and SK) in addition to the delated-rectifier KV, delayed the repolarization of the action potential and induced the appearance of a plateau potential after the peak of each action potential. The amplitude and width of the action potential were increased as with 4-AP (Fig. 3B). The amplitude of the AHP was reduced, and its latency, duration, and shape were completely modified. The selective SK channel blocker apamin induced a selective decrease of the second, late component of the AHP, without modifying the shape of the action potential (Fig. 3C). In summary, the pharmacological effects of the potassium channel blocker were similar to those previously observed in type B MVNn (Johnston et al. 1994; Peusner et al. 1998).
|
Responses of LVNn to the injection of depolarizing ramplike currents
Out of the 5 ramps applied to each cell, the 600-ms (slope of 0.5 nA/s) ramp gave the most significant results and was taken as the main index of the response of LVNn to ramplike currents. Figure 4A shows the typical response of an LVNn to a 600-ms ramp delivered from a hyperpolarized level, below the firing threshold. In most LVNn tested (25 out of 27), the instantaneous firing (IF) frequency reached at the end of the ramp was higher than the one observed at the end of the current plateau; that is, an overshoot was observed (Fig. 4A). The mean overshoot of LVNn (6.9 ± 6.6 spikes/s) was similar to the one displayed by type B MVNn, and significantly higher than the one displayed by type A MVNn (P = 0.014, Fig. 4B). The same was true for the ramps delivered from the resting membrane potential (Fig. 4B). Whatever the level of polarization of the neurons, the sensitivity of LVNn to steep, ramplike current injection, given by kIF, was similar to that of both types of MVNn, as shown in Fig. 4B (upper panels).
The slope of increase of the firing rate in response to the injected current (kIF) and the overshoot were higher when the slope of the ramp increased. For example, the mean values obtained for 600-ms ramps delivered from a hyperpolarized level (kIF = 114.9 ± 47.0 spikes s–1 nA–1, overshoot = 6.9 ± 6.6 spikes/s) were significantly higher than those obtained for 3,400-ms ramps (kIF = 98.7 ± 39.6 spikes s–1 nA–1, P = 0.004; overshoot = 2.0 ± 1.9 spikes/s, P < 0.001). The same increase of the mean kIF and overshoot of LVNn with the slope of the ramp was observed for the ramps delivered from the resting membrane potential. However, compared with the ramps delivered from a hyperpolarized level, the mean overshoots and kIF of the ramps delivered from rest tended to be smaller for all slopes.
Responses of LVNn to injections of sinusoidal currents delivered during steady-state hyperpolarization, in the absence of action potentials
When sinusoidal currents were delivered in the absence of action potentials, LVNn responded by a sinusoidal modulation of their membrane potential (Fig. 5A). In accordance with their lower input resistance, the impedance Zh of LVNn was significantly lower than that of both types of MVNn at all frequencies tested (Fig. 5C). At 0.4 Hz, for instance, Zh was 58.7 ± 20.2 M for LVNn versus 131.2 ± 58.7 M for MVNn (P < 0.001, Table 2 HP level). As was the case for MVNn, the membrane potential modulation Vh of LVNn displayed a small resonance at a median peak frequency of 0.7 Hz. This median peak frequency was comparable to the one obtained for type A and type B MVNn (Fig. 5C, Table 2). Although the Zh 4/0.4 ratio was not significantly different between LVNn and MVNn, the amplitude of the resonance of LVNn (1.08 ± 0.08) was significantly lower than the amplitude of the resonance obtained for the whole sample of MVNn (P = 0.006, Table 2). However, these results demonstrate that the membrane of LVNn as well as MVNn does not behave in a purely passive way at moderately hyperpolarized levels. In accordance with its slight resonance, the membrane potential response of LVNn displayed a small, nonsignificant phase lead with regard to the injected current at the lowest frequencies of stimulation, which became a phase lag at higher frequencies (Fig. 5C). The phase lags obtained for LVNn were significantly lower than those measured for the whole sample of MVNn at intermediate frequencies (from 2 to 14 Hz), which is mainly attributed to the lower impedance of LVNn. Indeed, independently of the amplitude of the resonance, the lower impedance of LVNn resulted in higher "corner frequencies" and in a shift of the phase lags to higher frequencies compared with those of MVNn.
Responses of LVNn to injections of sinusoidal currents delivered at (or just above) the resting membrane potential
When sinusoidal currents were delivered in the presence of action potentials, LVNn responded by a sinusoidal modulation of their membrane potential and firing rate (Fig. 5B). As pointed out in METHODS, the mean membrane potential excursion could be reliably estimated for frequencies 1 Hz. In accordance with the lower input resistance of LVNn, their active impedance Z (V/I) measured at 0.4 Hz was significantly lower than the one obtained for either type of MVNn (P < 0.001 for type A and P = 0.002 for type B MVNn, Table 2). In terms of instantaneous firing rate, the sensitivity of LVNn to sinusoidal current injection was 105.4 ± 39.2 spikes s–1 nA–1 at 0.4 Hz, which was significantly lower than that for type B MVNn (P = 0.006), but comparable to the value obtained for type A MVNn (Table 2). Indeed, compared with type A MVNn, the lower active impedance of LVNn was compensated by a higher sensitivity of their discharge to membrane potential variations (IF/V, P = 0.007, Table 2).
The sensitivity to sinusoidal current IF/I of LVNn was significantly lower than that of the whole sample of MVNn for all frequencies below 10 Hz (Fig. 6A2). Over this range of frequencies, the sensitivity to sinusoidal current of LVNn was always significantly lower than that of type B MVNn, but only tended to be lower than that of type A MVNn, except at 4 and 8 Hz. However, the main difference between LVNn and MVNn was that the sensitivity of LVNn did not vary as much over the whole range of frequencies tested. In other words, compared with both type A and type B MVNn, the Bode plot showing the variation of IF/I with frequency was flatter (Fig. 6, A1 and A2). The ratio between IF at 4 Hz and 0.4 Hz (IF 4/04) was significantly lower for LVNn than that for either type A or type B MVNn (P values of 0.025 and 0.006, respectively; Table 2). The amplitude of the resonance (R-ratio) was also significantly lower for LVNn than for the whole sample of MVNn (P = 0.039, Table 2). Furthermore, data from individual neurons (Fig. 6A1) showed that most LVNn did not show the gradual decrease of their firing rate modulation by current injection usually observed for MVNn after the peak frequency of resonance (Fig. 6A2). This sudden stop of the modulation of spontaneous firing arose from the fact that LVNn tended to completely synchronize their firing with the depolarizing phase of sinusoidal current injections as soon as the frequency of stimulation became higher than about half of their spontaneous firing rate. The peak frequency of resonance (R-peak) of LVNn was extremely variable around a median of 0.8 Hz (see the data for individual neurons on Fig. 6A1), but was nonetheless lower than that of type A and type B MVNn (P values of 0.009 and 0.013, respectively; Table 2).
|
The firing rate modulation of LVNn displayed a small, but significant phase lead with regard to the injected current at 0.2 Hz (P = 0.002, one-sample t-test) and 0.4 Hz (P = 0.01). This phase lead decreased to zero and became a phase lag at higher frequencies (Fig. 6A3). Compared with the phase values obtained for the membrane potential modulation during steady-state hyperpolarization, both the phase lead observed at low frequency and the phase lag obtained at high frequency tended to be greater. In accordance with their lower peak frequency of resonance, LVNn displayed a larger phase lag than that of MVNn at 0.8, 1, and 2 Hz (Fig. 6A3). At 2 Hz, for instance, the phase lag of LVNn reached –11.3 ± 13.3° versus –1.8 ± 7.0° for type A MVNn (P = 0.003) and –4.7 ± 9.9° for type B MVNn (P = 0.076).
Responses of LVNn to injections of sinusoidal currents during steady-state depolarization
The LVNn maintained under steady-state depolarization displayed a significantly higher spontaneous discharge than that at their resting membrane potential. Thus sinusoidal modulation of this spontaneous discharge was maintained up to higher frequencies of sinusoidal current injection (Fig. 6B). As at rest, the active impedance (V/I) of LVNn measured at 0.4 Hz was significantly lower than that obtained for either type of MVNn (P = 0.013 for type A and P = 0.003 for type B MVNn, Table 2 Depo level). In terms of instantaneous firing rate, the sensitivity of LVNn to sinusoidal current injection tended to be lower than that for type B MVNn (P = 0.072), and was comparable to the value obtained for type A MVNn (Table 2). Compared with both type A and type B MVNn, the lower active impedance of LVNn was compensated by a higher sensitivity of their discharge to membrane potential variations (IF/V, P values of 0.041 and 0.009, respectively; Table 2).
The sensitivity to sinusoidal current IF/I of LVNn was significantly lower than that of the whole sample of MVNn for all frequencies between 1 and 14 Hz (Fig. 6B2). Over this range of frequencies, the sensitivity to sinusoidal current of LVNn was significantly lower than that of type B MVNn from 2 to 10 Hz, but significantly lower than that of type A MVNn only at 4 and 6 Hz. As at rest, the Bode plot showing the variation of IF/I with frequency was flatter for LVNn than that for either type of MVNn (Fig. 6, B1 and B2). The IF 4/04 ratio was significantly lower for LVNn than that for type A or type B MVNn (P = 0.002 in both cases; Table 2). The amplitude of the resonance (R-ratio) was also significantly lower for LVNn than that for the whole sample of MVNn (P = 0.019, Table 2). As at the resting membrane potential, data from individual neurons (Fig. 6B1) showed that most LVNn did not show the gradual decrease of their firing rate modulation observed for MVNn after the peak frequency of resonance (Fig. 6B2). The peak frequency of resonance (R-peak) of LVNn was again very variable, but significantly higher than at the resting membrane potential, with a median value of 8 Hz. The median peak frequency of resonance of LVNn was significantly lower than that of type A MVNn (P = 0.046), but similar to that of type B MVNn (P = 0.30, Table 2).
Compared with the phase values obtained at the resting membrane potential, LVNn displayed a significant phase lead 0.6 Hz (P < 0.001, one sample t-test) instead of 0.4 Hz. In addition, LVNn had a smaller phase lag at intermediate frequencies (Fig. 6B3), in accordance with their higher frequency of resonance. As a consequence, the significant difference observed at rest between the phase lag displayed by LVNn and MVNn at 2 Hz disappeared (P = 0.23; Fig. 6B3).
|
|
DISCUSSION |
|---|
|
Location and identification of the recorded neurons
Before addressing the static and dynamic membrane properties of LVNn in more detail, we need to discuss whether the neurons we recorded from were in the LVN, and whether they can be more precisely identified. In cytoarchitectonic studies of the mammalian vestibular nuclei, the LVN has been classically defined as that part of the vestibular complex that contains a large number of multipolar, giant-size cells, that is, the Deiters' neurons (Brodal and Pompeiano 1957). In general, the boundaries of the LVN are quite easy to define, but there is a controversy concerning the border between the ventro-lateral part of the MVN and the ventro-medial part of the caudal LVN (for review see Ris et al. 1999). Indeed, the area denoted vL in Fig. 1A1 has been reported to belong either to the LVN (Brodal and Pompeiano 1957; Mehler and Rubertone 1985) or to the magnocellular part of the MVN (Epema et al. 1988; Gerrits et al. 1985) in the rabbit, cat, and rat. Two cytoarchitectonic studies performed in the guinea pig have concluded that this vL area was part of the LVN (Gstoettner and Burian 1987; Ris et al. 1999), but other authors disagree (Graf et al. 2002). Although we agree that, in the guinea pig, vL cannot be incorporated into the MVN on cytoarchitectonic criteria, this area appears to be the origin of the ascending tract of Deiters, which is involved in the control of eye movements (Graf et al. 2002; Nguyen et al. 1999). In this respect, then, it belongs more to the MVN, which has both oculomotor and spinal projections, than to the LVN, whose neurons project almost exclusively to the spinal cord (Carleton and Carpenter 1983; Sarkisian 2000).
The 12 representative neurons that were injected with neurobiotin were all localized within the boundaries of the LVN, and none of them was in the controversial vL area. Indeed, they were all recorded more rostrally, at a level where the LVN and MVN are clearly separated from one another by the acoustic stria (Fig. 1, A2 and A3). Because all neurons we recorded were localized using the same landmarks, this suggests that most of them were LVNn, and were not involved in the control of eye movements by the ascending tract of Deiters.
All LVNn injected with neurobiotin were large or giant neurons, with a mean soma diameter of 45 µm. Together with the strong homogeneity of their membrane properties, this suggests that most of the LVNn we recorded had a similar size and were Deiters' neurons much bigger than the usual MVNn (compare Fig. 1B and 1C). In the guinea pig as in other vertebrates, however, the LVN is characterized by a morphological variety of cells (Gstoettner and Burian 1987; Mehler and Rubertone 1985; Pompeiano 1991; Ris et al. 1999). In addition to the typical giant cells (Deiters' cells), the nucleus also contains small and medium-size cells that were apparently not recorded in this study. This sampling bias was probably attributable to the fact that bigger neurons were much easier to penetrate with intracellular recording electrodes than the smaller cells. This bias was not a problem in the MVN, given that MVNn are all densely packed, small to medium-size cells (Gstoettner and Burian 1987).
Basic membrane and firing properties of LVNn compared with those of MVNn
As mentioned above, the homogeneity of the membrane properties of LVNn we recorded contrasts with the heterogeneity of MVNn. Considering together the spike shape, membrane resistance, and spontaneous firing properties of LVNn, they were different from those of either type A or type B MVNn. In a sense, the large LVNn, which tended to display both the double AHP of type B MVNn and the A-like rectification of type A MVNn, looked like a composite of type A and type B MVNn. Although LVNn express the same strong persistent sodium currents as type B MVNn, none of the LVNn displayed the LTS that characterize the B + LTS MVNn. The homogeneity of LVNn may reflect the fact that most of the neurons we penetrated were probably large or giant Deiters' neurons.
LVNn have the same mean membrane potential as that of MVNn, but their average discharge rate tends to be lower. This difference has been observed by others authors in vitro (Sun et al. 2002). Furthermore, both extracellular and intracellular recordings show that about one third of LVNn are silent at rest in vitro. This proportion is higher than the proportion of silent neurons found among intracellularly recorded MVNn (7%), as well as among MVNn recorded extracellularly in an in vitro whole brain preparation (12%; Vibert et al. 1999a). In accordance with their lower discharge rate, the spontaneous discharge of LVNn was more irregular than that of MVNn. Actually, there was a significant negative relationship between the coefficient of variation of the LVNn we recorded and their spontaneous firing rate. Surprisingly, in vivo recordings suggest the average resting frequency of neurons in LVN is similar to, or higher than, that of neurons in the MVN (Ris et al. 1995; Ryu and McCabe 1971). This discrepancy suggests that the resting activity of LVNn in vivo may depend more than that of MVNn on the excitatory inputs they receive from the vestibular nerve, and from other structures like the medullary reticular formation and cerebellar fastigial nucleus. Alternatively, the discrepancy between in vivo and in vitro data may be explained by the fact that we recorded large Deiters' neurons among LVN cells. Indeed, extracellular recordings in the cat have shown that in the LVN, the spontaneous firing rate of identified vestibulospinal neurons was inversely related to their conduction velocity, and hence presumably to their size (Pompeiano 1991). In other words, the largest LVNn would have significantly lower spontaneous discharge rates than those of the other vestibulospinal neurons within the LVN. Similarly, the low input membrane resistance and active membrane impedance displayed by LVNn compared with MVNn must correspond to their bigger size.
According to the results obtained with potassium channel antagonists, the pharmacology of giant LVNn is completely in agreement with what has been found in MVNn (Johnston et al. 1994; Peusner et al. 1998). No fundamental difference was found in the types of potassium currents that shape the action potential and interspike interval.
Responses of LVNn to ramplike and sinusoidal currents
In accordance with their much lower input resistance and active impedance, the sensitivity of the firing rate of large-size LVNn to small-amplitude sinusoidal current injections was significantly lower than that of the whole sample of MVNn. Separate comparisons with each type of MVNn demonstrated that there was a significant difference between LVNn and type B MVNn, but that the trend for LVNn to display a lower sensitivity than that of type A MVNn was often not significant. This means that, at least compared with type A MVNn, the lower active impedance of LVNn was compensated by a higher sensitivity of their discharge to membrane potential variations (IF/V). This looks even truer for large-amplitude, ramplike current injections, given that the trend for the sensitivity of the firing rate of LVNn to be lower than that of MVNn disappeared for such stimuli. In addition, the overshoot displayed by LVNn in response to steep ramps was similar to that of type B MVNn, and significantly higher than that of type A MVNn.
The amplitude of the response of LVNn to current injections was pretty flat over a large range of frequency, with a smaller resonance than that in both types of MVNn. This was not attributed to the lower mean spontaneous discharge rate of LVNn, given that the shape of the Bode plot was not modified when the LVNn were maintained continuously depolarized and therefore had a much higher spontaneous firing rate. As pointed out above, the LVNn we recorded looked like a composite of type A and type B MVNn but were not endowed with radically different membrane properties. This means that just subtle changes in the membrane properties of central neurons are sufficient to modify the profile of their dynamic responses. This is in accordance with results obtained using realistic neuronal models of each type of MVNn (Av-Ron and Vidal 1999; Ris et al. 2001), which show that the dynamics of their responses to current injection can be strongly modified by small changes in one or two of their ionic conductances. In addition, the amplitude of the firing rate modulation of LVNn by sinusoidal currents was limited to a certain cutoff frequency, above which there was no indication of a progressive fall as seen in MVNn past their peak frequency of resonance. This suggests that LVNn tend to synchronize their discharge very efficiently with high-frequency inputs (more than both types of MVNn). Interestingly, the type B MVNn, with their sharper spikes and larger overshoot, already tended to show a better synchronization than did type A neurons (Ris et al. 2001).
Do the different membrane properties displayed by LVNn and MVNn reflect their distinct functional role?
In many respects, and despite their lower sensitivity to current injections, LVNn resemble more type B MVNn than type A MVNn. Indeed, both groups of neurons display similar spike shapes, a high sensitivity of the firing rate to membrane potential variation, and strong nonlinear properties like the overshoot and persistent sodium current. Because the LVNn we recorded are probably mainly vestibulospinal neurons, it is tempting to speculate that among MVNn, type B neurons might correspond to vestibulospinal neurons, whereas type A neurons might be vestibuloocular neurons. However, this is probably not the case because in data from the isolated whole brain, a similar proportion of type A and type B MVNn have been identified as vestibulospinal neurons (Vibert et al. 1997; and unpublished data). In addition, the picture is complicated by the fact that the MVN includes a nonnegligible proportion of vestibulo-oculo-spinal neurons that project both to the oculomotor system and the spinal cord (for reviews see Uchino and Isu 1992; Vibert et al. 1997). We think more of the type A and type B MVNn as mediating parallel pathways going both to the oculomotor and postural systems, but having different dynamic properties (Ris et al. 2001).
Although some of the differences between LVNn and MVNn (such as the different membrane resistance) probably result from their very different sizes, the distinct dynamics displayed by LVNn in response to current injections cannot be directly inferred from their anatomy. Indeed, as mentioned above, realistic neuronal models of MVNn show that the resonant properties of neurons can be modified by slight changes of specific ion conductance (Av-Ron and Vidal 1999; Ris et al. 2001). This suggests that the difference between the resonance displayed by MVNn and LVNn might reflect their distinct functional roles, that is, LVNn are mainly involved in the control of posture, whereas MVNn are rather involved in eye–head coordination and gaze control.
The LVNn we recorded presumably constitute a specific subset of the neurons present in the LVN. According to Boyle and Pompeiano (1980), and Kasper et al. (1988), neurons of the cat LVN contain in vivo a population of cells whose responses to whole body rotations in the vertical plane (i.e., amplitude of the modulation of the firing rate and phase with regard to position) remain relatively unmodified with changes in frequency. These positional responses were attributed to stimulation of macular receptors. Recently Angelaki and Dickman (2000) studied, in monkeys, properties of central otolith neurons that respond to linear accelerations of the head. Among these neurons recorded in the rostral vestibular nuclei, at least 3 groups of central response dynamics were described. 1) "High-pass" neurons exhibited increasing gains and phase values as a function of frequency. 2) "Flat" neurons were characterized by relatively flat gains and constant phase lags (20–55°). 3) A few neurons ("low-pass") were characterized by decreasing gain and phase as a function of frequency. Because the response dynamics of the large LVNn we recorded appears very similar to that of the so-called flat neurons, it is tempting to speculate that the latter might correspond to the giant vestibulospinal Deiters' cells of the LVN.
|
|
DISCLOSURES |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
Address for reprint requests: A. Uno, Laboratoire de Neurobiologie des Réseaux Sensorimoteurs, CNRS UMR 7060, Université Paris 5, 45 rue des Saints-Pères, 75270 Paris Cédex 06, France. (E-mail: auno{at}ent.med.osaka-u.ac.jp).
|
|
REFERENCES |
|---|
|
Aitken PG, Breese GR, Dudek FF, Edwards F, Espanol MT, Larkman PM, Lipton P, Newman GC, Nowak TS Jr, Panizzon KL, Raley-Susman KM, Reid KH, Rice ME, Sarvey JM, Schoepp DD, Segal M, Taylor CP, Teyler TJ, and Voulalas PJ. Preparative methods for brain slices: a discussion. J Neurosci Methods 59: 139–149, 1995.[Web of Science][Medline]
Angelaki DE and Dickman JD. Spatiotemporal processing of linear acceleration: primary afferent and central vestibular neurons responses. J Neurophysiol 84: 2113–2132, 2000.
Av-Ron E and Vidal PP. Intrinsic membrane properties and dynamics of medial vestibular neurons: a simulation. Biol Cybern 80: 383–392, 1999.[Web of Science][Medline]
Babalian A, Vibert N, Assié G, Serafin M, Mühlethaler M, and Vidal PP. Central vestibular networks in the guinea pig: functional characterization in the isolated whole brain in vitro. Neuroscience 81: 405–426, 1997.[Web of Science][Medline]
Bankoul S, Goto T, Yates B, and Wilson VJ. Cervical primary afferent input to vestibulospinal neurons projecting to the cervical dorsal horn: an anterograde and retrograde tracing study in the cat. J Comp Neurol 353: 529–538, 1995.[Web of Science][Medline]
Beraneck M, Hachemaoui M, Idoux E, Ris L, Uno A, Godaux E, Vidal PP, Moore L, and Vibert N. Long-term plasticity of ipsilesional medial vestibular nucleus neurons after unilateral labyrinthectomy. J Neurophysiol 90: 184–203, 2003.
Boyle R and Pompeiano O. Reciprocal responses to sinusoidal tilt of neurons in Deiters' nucleus and their dynamic characteristics. Arch Ital Biol 118: 1–32, 1980.[Web of Science][Medline]
Brodal A and Pompeiano O. The vestibular nuclei in the cat. J Anat 91: 438–454, 1957.[Web of Science][Medline]
Carleton SC and Carpenter MB. Afferent and efferent connections of the medial, inferior and lateral nuclei in the cat and monkey. Brain Res 278: 29–51, 1983.[Web of Science][Medline]
Carleton SC and Carpenter MB. Distribution of primary vestibular fibers in the brainstem and cerebellum of the monkey. Brain Res 294: 281–298, 1984.[Web of Science][Medline]
Chen-Huang C, McCrea RA, and Goldberg JM. Contributions of regularly and irregularly discharging vestibular nerve inputs to the discharge of central vestibular neurons in the alert squirrel monkey. Exp Brain Res 114: 405–422, 1997.[Web of Science][Medline]
Darlington CL, Gallagher JP, and Smith PF. In vitro electrophysiological studies of the vestibular nucleus complex. Prog Neurobiol 45: 335–346, 1995.[Web of Science][Medline]
Devor A, Fritschy JM, and Yarom Y. Spatial distribution and subunit composition of GABAA receptors in the inferior olivary nucleus. J Neurophysiol 85: 1686–1696, 2001.
Du Lac S and Lisberger SG. Membrane and firing properties of avian medial vestibular nucleus neurons in vitro. J Comp Physiol A Sens Neural Behav Physiol 176: 641–651, 1995a.[Medline]
Du Lac S and Lisberger SG. Cellular processing of temporal information in medial vestibular nucleus neurons. J Neurosci 15: 8000–8010, 1995b.[Abstract]
Dutia MB and Johnston AR. Development of action potentials and apamin-sensitive after-potentials in mouse vestibular nucleus neurons. Exp Brain Res 118: 148–154, 1998.[Web of Science][Medline]
Epema AH, Gerrits NM, and Voogd J. Commissural and intrinsic connections of the vestibular nuclei in the rabbit: a retrograde labeling study. Exp Brain Res 71: 129–146, 1988.[Web of Science][Medline]
Gacek RR. The course and central termination of first-order neurons supplying vestibular endorgans in the cat. Acta Otolaryngol Suppl 254: 1–66, 1969.[Medline]
Gallagher JP, Lewis MR, and Shinnick-Galagher P. An electrophysiological investigation of the rat medial vestibular nucleus in vitro. In: Contemporary Sensory Neurobiology, edited by Correia MJ and Perachio AA. New York: A. R. Liss, 1985, p. 293–304.
Gerrits NM, Voogd J, and Magras IN. Vestibular afferents of the inferior olive and the vestibulo-olivo-cerebellar climbing fiber pathway to the flocculus in the cat. Brain Res 332: 325–336, 1985.[Web of Science][Medline]
Goldberg JM, Highstein SM, Moschovakis AK, and Fernandez C. Inputs from regularly and irregularly discharging vestibular nerve afferents to secondary neurons in the vestibular nuclei of the squirrel monkey. I. An electrophysiological analysis. J Neurophysiol 58: 700–718, 1987.
Graf W, Gerrits N, Yatim-Dhiba N, and Ugolini G. Mapping the oculomotor system: the power of transneuronal labeling with rabies virus. Eur J Neurosci 15: 1557–1562, 2002.[Web of Science][Medline]
Gstoettner W and Burian M. Vestibular nuclear complex in the guinea pig: a cytoarchitectonic study and map in three planes. J Comp Neurol 257: 176–188, 1987.[Web of Science][Medline]
Gstoettner W, Burian M, and Cartellieri M. Central projections from singular parts of the vestibular labyrinth in the guinea pig. Acta Otolaryngol 112: 486–495, 1992.[Medline]
Hagiwara S, Kusano K, and Saito N. Membrane changes of Onchodium nerve cells in potassium-rich media. J Physiol 155: 470–489, 1961.
Jian BJ, Shintani T, Emanuel BA, and Yates BJ. Convergence of limb, visceral and vertical semicircular canal or otolith inputs onto vestibular nucleus neurons. Exp Brain Res 144: 247–257, 2002.[Web of Science][Medline]
Johnston AR, MacLeod NK, and Dutia MB. Ionic conductances contributing to spike repolarization and after-potentials in rat medial vestibular nucleus neurons. J Physiol 481: 61–77, 1994.
Kasper J, Schor RH, and Wilson VJ. Response of vestibular neurons to head rotations in vertical planes. I. Response to vestibular stimulation. J Neurophysiol 60: 1753–1764, 1988.
Korte GE. The brainstem projection of the vestibular nerve in the cat. J Comp Neurol 184: 279–292, 1979.[Web of Science][Medline]
Krawitz S, Fedirchuk B, Dai Y, Jordan LM, and McCrea DA. State-dependent hyperpolarization of voltage threshold enhances motoneurone excitability during fictive locomotion in the cat. J Physiol 532: 271–281, 2001.
Lliná RR. The intrinsic electrophysiological properties of mammalian neurons: insights into central nervous system function. Science 242: 1654–1664, 1988.
Mehler WR and Rubertone JA. Anatomy of the vestibular nucleus complex. In: The Rat Nervous System, edited by Paxinos G. New York: Academic Press, 1985, p. 185–219.
Moore LE, Hill RH, and Grillner S. Voltage-clamp frequency domain analysis of NMDA-activated neurons. J Exp Biol 175: 59–87, 1993.[Abstract]
Moyer JR Jr and Brown TH. Methods for whole-cell recording from visually preselected neurons of perirhinal cortex in brain slices from young and aging rats. J Neurosci Methods 86: 35–54, 1998.[Web of Science][Medline]
Nguyen LT, Baker R, and Spencer RF. Abducens internuclear and ascending tract of Deiters inputs to medial rectus motoneurons in the cat oculomotor nucleus: synaptic organization. J Comp Neurol 405: 141–159, 1999.[Web of Science][Medline]
Peusner KD, Gamkrelidze G, and Giaume C. Potassium currents and excitability in second-order auditory and vestibular neurons. J Neurosci Res 53: 511–520, 1998.[Web of Science][Medline]
Pompeiano O. The role of different size vestibulospinal neurons in the static control of posture. Arch Ital Biol 129: 21–41, 1991.[Web of Science][Medline]
Ris L, de Waele C, Serafin M, Vidal PP, and Godaux E. Neuronal activity in the ipsilateral vestibular nucleus following unilateral labyrinthectomy in the alert guinea pig. J Neurophysiol 74: 2087–2099, 1995.
Ris L, Saussez S, Gerrits N, Godaux E, and Pochet R. The subdivisions of the guinea pig vestibular complex revealed by acetylcholinesterase staining. J Vestib Res 9: 73–81, 1999.[Web of Science][Medline]
Ris L, Hachemaoui M, Vibert N, Godaux E, Vidal PP, and Moore LE. Resonance of spike discharge modulation in neurons of the guinea pig medial vestibular nucleus. J Neurophysiol 86: 703–716, 2001.
Rothman SM. The neurotoxicity of excitatory amino acids is produced by passive chloride influx. J Neurosci 5: 1483–1489, 1985.[Abstract]
Ryu JH and McCabe BF. Types of neuronal activity in the lateral vestibular nucleus. Acta Otolaryngol 72: 288–291, 1971.[Medline]
Sarkisian VH. Input-output relation of Deiters' lateral vestibulospinal neurons with different structures of the brain. Arch Ital Biol 138: 295–353, 2000.[Web of Science][Medline]
Sato F and Sasaki H. Morphological correlation between spontaneously discharging primary vestibular afferents and vestibular nucleus neurons in the cat. J Comp Neurol 333: 554–566, 1993.[Web of Science][Medline]
Sato F, Sasaki H, Ishizuka N, Sasaki S, and Mannen H. Morphology of single primary vestibular afferents originating from the horizontal semicircular canal in the cat. J Comp Neurol 290: 423–439, 1989.[Web of Science][Medline]
Schaefer KP and Meyer DL. Eye and head movements as specialized functions of vestibular circuits. In: The Head-Neck Sensory Motor System, edited by Berthoz A, Vidal PP, and Graf W. New York: Oxford University Press, 1992, p. 241–243.
Serafin M, de Waele C, Khateb A, Vidal PP, and Mühlethaler Llinas M. Medial vestibular nucleus in the guinea-pig. I. Intrinsic membrane properties in brainstem slices. Exp Brain Res 84: 417–425, 1991a.[Web of Science][Medline]
Serafin M, de Waele C, Khateb A, Vidal PP, and Mühlethaler M. Medial vestibular nucleus in the guinea-pig. II. Ionic basis of the intrinsic membrane properties in brainstem slices. Exp Brain Res 84: 426–433, 1991b.[Web of Science][Medline]
Shimazu H and Precht W. Inhibition of central vestibular neurons from the contralateral labyrinth and its mediating pathway. J Neurophysiol 29: 467–492, 1966.
Smith MR, Nelson AB, and Du Lac S. Regulation of firing response gain by calcium-dependent mechanisms in vestibular nucleus neurons. J Neurophysiol 87: 2031–2042, 2002.
Sun Y, Waller HJ, Godfrey DA, and Rubin AM. Spontaneous activity in rat vestibular nuclei in brain slices and effects of acetylcholine agonists and antagonists. Brain Res 934: 58–68, 2002.[Web of Science][Medline]
Uchino Y and Isu N. Properties of vestibulo-ocular and/or vestibulo-collic neurons in the cat. In: The Head-Neck Sensory Motor System, edited by Berthoz A, Vidal PP, and Graf W. New York: Oxford University Press, 1992, p. 266–272.
Uchino Y, Sato H, Zakir M, Kushiro K, Imagawa M, Ogawa Y, Ono S, Meng H, Zhang X, Katsuta M, Isu N, and Wilson VJ. Commissural effects in the otolith system. Exp Brain Res 136: 421–430, 2001.[Web of Science][Medline]
Vibert N, Babalian A, Serafin M, Gasc JP, Mühlethaler M, and Vidal PP. Plastic changes underlying vestibular compensation in the guinea-pig persist in isolated, in vitro whole brain preparations. Neuroscience 93: 413–432, 1999a.[Web of Science][Medline]
Vibert N, Bantikyan A, Babalian A, Serafin M, Mühlethaler M, and Vidal PP. Post-lesional plasticity in the central nervous system of the guinea-pig: a "top-down" adaptation process? Neuroscience 94: 1–5, 1999b.[Web of Science][Medline]
Vibert N, de Waele C, Serafin M, Babalian A, Mühlethaler M, and Vidal PP. The vestibular system as a model of sensorimotor transformations. A combined in vivo and in vitro approach to study the cellular mechanisms of gaze and posture stabilization in mammals. Prog Neurobiol 51: 243–286, 1997.[Web of Science][Medline]
Vidal PP, Vibert N, Serafin M, Babalian A, Mühlethaler M, and de Waele C. Intrinsic physiological and pharmacological properties of central vestibular neurons. In: Advances in Oto-rhino-laryngology. Vestibular Dysfunction and its Therapy, edited by Büuttner U. Basel, Switzerland: Karger, 1999, vol. 55, p. 26–81.
Walberg F, Bowsher D, and Brodal A. The termination of primary vestibular fibers in the vestibular nuclei of the cat: an experimental study with silver methods. J Comp Neurol 110: 391–419, 1958.[Web of Science][Medline]
Wilson VJ and Melvill-Jones G. Mammalian Vestibular Physiology. New York: Plenum Press, 1979.
Wilson VJ and Peterson BW. Vestibular and reticular projections to the neck. In: Control of Head Movement, edited by Peterson BW and Richmond FJ. New York: Oxford University Press, 1988, p. 129–140.
Wilson VJ and Schor RH. The neural substrate of the vestibulocollic reflex. What needs to be learned. Exp Brain Res 129: 483–493, 1999.[Web of Science][Medline]
Xiong G and Matsushita M. Ipsilateral and contralateral projections from upper cervical segments to the vestibular nuclei in the rat. Exp Brain Res 141: 204–217, 2001.[Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  N. Pilati, M. Barker, S. Panteleimonitis, R. Donga, and M. Hamann A Rapid Method Combining Golgi and Nissl Staining to Study Neuronal Morphology and Cytoarchitecture J. Histochem. Cytochem., June 1, 2008; 56(6): 539 - 550. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Eugene, S. Deforges, F. Guimont, E. Idoux, P.-P. Vidal, L. E. Moore, and N. Vibert Developmental regulation of the membrane properties of central vestibular neurons by sensory vestibular information in the mouse J. Physiol., September 15, 2007; 583(3): 923 - 943. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Idoux, M. Serafin, P. Fort, P.-P. Vidal, M. Beraneck, N. Vibert, M. Muhlethaler, and L. E. Moore Oscillatory and Intrinsic Membrane Properties of Guinea Pig Nucleus Prepositus Hypoglossi Neurons In Vitro J Neurophysiol, July 1, 2006; 96(1): 175 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Sekirnjak and S. du Lac Physiological and Anatomical Properties of Mouse Medial Vestibular Nucleus Neurons Projecting to the Oculomotor Nucleus J Neurophysiol, May 1, 2006; 95(5): 3012 - 3023. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Beraneck, E. Idoux, A. Uno, P.-P. Vidal, L. E. Moore, and N. Vibert Unilateral Labyrinthectomy Modifies the Membrane Properties of Contralesional Vestibular Neurons J Neurophysiol, September 1, 2004; 92(3): 1668 - 1684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Straka, M. Beraneck, M. Rohregger, L. E. Moore, P.-P. Vidal, and N. Vibert Second-Order Vestibular Neurons Form Separate Populations With Different Membrane and Discharge Properties J Neurophysiol, August 1, 2004; 92(2): 845 - 861. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
