The Journal of Neurophysiology Vol. 78 No. 4 October 1997, pp. 1800-1810
Copyright ©1997 by the American Physiological Society

Two Types of Actions of Norepinephrine on Identified Auditory Efferent Neurons in Rat Brain Stem Slices

Xueyong Wang and Donald Robertson

The Auditory Laboratory, Department of Physiology, The University of Western Australia, Nedlands, Western Australia 6907, Australia

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Wang, Xueyong and Donald Robertson. Two types of actions of norepinephrine on identified auditory efferent neurons in rat brain stem slices. J. Neurophysiol. 78: 1800-1810, 1997. Whole cell voltage-clamp recordings were performed on auditory olivocochlear neurons in the ventral nucleus of the trapezoid body (VNTB) of brain stem slices from neonatal rats. Each neuron was identified by retrograde labeling with Fast Blue injected into the cochlea. Bath application of norepinephrine (NE; 1-10 µM) reversibly induced an inward current in 26 of 38 labeled neurons that were voltage clamped at 75 mV. This was responsible for the membrane depolarization to NE observed in current-clamp mode. The NE-induced inward current appeared to be more prominent at 55 mV than at 75 mV and reversed at around 100 mV. It was attenuated but not prevented by 20 mM tetraethylammonium, and it persisted when the perfusate contained 2 mM Cs⁺ or 100 µM Cd²⁺. However, the NE-induced inward current was attenuated to varying degrees in a zero-Ca²⁺ solution. Current-voltage plots revealed that NE caused a decrease in membrane K⁺ conductance. A suppression of voltage-gated Ca²⁺ currents by NE was also observed. The excitatory action of NE was blocked by the -adrenoreceptor antagonist phentolamine. The ₁-adrenoreceptor agonist phenylephrine had an effect similar to that of NE. In 6 of 38 labeled neurons, an inhibitory action of NE (1-10 µM) was observed that appeared to be due to an activation of an inwardly rectified K⁺ current, which caused hyperpolarization of resting membrane potentials in current-clamp mode. This inhibitory response was independent of external Ca²⁺ and was abolished by 2-5 mM Cs⁺ or 0.5 mM Ba²⁺ applied in the perfusate. The receptors involved in the inhibitory actions of NE are not clear. The effect was partially and reversibly blocked by propranolol (10 µM), a -adrenoreceptor antagonist. However, isoprenaline (10 µM), a -adrenoreceptor agonist, failed to induce any effect. On the other hand, the inhibitory effect was irreversibly blocked by pretreatment with phentolamine (5-10 µM). Phenylephrine (5-10 µM) had no effect.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

The ventral nucleus of the trapezoid body (VNTB) and the rostral periolivary zone of the rat brain stem constitute an important source of auditory efferent projections to the inner ear (Aschoff and Ostwald 1988; Robertson et al. 1989; Vetter et al. 1993; Warr and Guinan 1979; White and Warr 1983). In particular, these regions give rise to the medial olivocochlear system, which projects bilaterally to terminate on the outer hair cells. This system has been shown, under a variety of conditions, to exert effects on cochlear sensitivity and susceptibility to loud sounds (Collet et al. 1990; Giraud et al. 1995; Guinan and Gifford 1988a-c; Kawase and Takasaka 1995; Patuzzi and Thompson 1991; Rajan 1988; Warren and Liberman 1989) and therefore may participate in the central control of hearing processes under both normal and extreme conditions. Our previous studies in which both immunohistological and electrophysiological methods were used have shown that the VNTB is innervated by noradrenergic fibers and that the excitability of many neurons in the VNTB is modulated by norepinephrine (NE) and by other neurotransmitters (Wang and Robertson 1997b; Wynne and Robertson 1996). It is of interest to know the electrophysiological and pharmacological nature of the neurochemical influences on these auditory efferent neurons in the brain stem. Such knowledge may lead to a future understanding of how convergence and integration of auditory efferent and afferent signals occur at the level of the brain stem and how chemical disturbances in the brain might alter auditory perception and the susceptibility to deafening influences. In addition, with such knowledge might come the possibility of therapeutic manipulation of inner ear function through centrally rather than peripherally acting pharmacological agents. In the present study, by combining retrograde labeling with slice patch-clamp recording techniques, we record directly from identified auditory efferent neurons in the VNTB and provide direct evidence that they are targets of NE modulation that involves regulation of K⁺ channels. Some of the results have been briefly reported previously (Wang and Robertson 1997a).

	METHODS

Abstract Introduction Methods Results Discussion References

Retrograde labeling

Wistar rats of 5-10 days postnatal age were anesthetized with methoxyflurane or ether (in later experiments) and the cochlea on one side was exposed. 0.1-0.5 µl of Fast Blue (1% in distilled water) was injected into the cochlea through the round window visualized under a dissection microscope. Rats were allow to recover from the operation and were kept 1-5 days before slice experiments.

Slice preparation

Brain stem slices were prepared from 7- to 15-day-old rats with the use of a variation of a method described previously (Robertson 1996). Briefly, the animals were decapitated and the brain stem with attached cerebellum was quickly removed and placed in ice cold artificial cerebrospinal fluid (ACSF). A block of tissue was cut into 300- to 350-µm sections with a Vibratome (series 1000). Two to three slices containing the VNTB were incubated in ACSF for 1.5 h before recording was commenced. The composition of ACSF was as follows (in mM): 130 NaCl, 3 KCl, 1.2 KH₂PO₄, 20 NaHCO₃, 2.4 CaCl₂, 1.3 MgCl₂, and 10 D-glucose, pH 7.4 after equilibration with 95% O₂-5% CO₂.

Electrical recordings

Whole cell patch-clamp recordings were made from Fast-Blue-labeled neurons within VNTB with the use of an Axopatch 200B amplifier (Axon Instruments). Output data were low-pass filtered at 5 kHz and collected through a DigiData 1200 interface with the use of pClamp software (Axon Instruments). Patch pipettes were pulled from borosilicate glass (Clark Electromedical Instruments, Reading, UK; 1.5 mm OD, 0.86 mm ID) with the use of a Brown-Flaming P87 puller. Electrodes had tip resistances ranging from 2 to 4 M when filled with a routinely used solution that contained (in mM): 115 potassium gluconate, 35 KCl, 10 N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid, 2 MgSO₄, 10 ethylene glycol-bis-(-aminoethyl ether) N,N,N,N-tetraacetic acid (EGTA), 2 Na-ATP, and 0.3 sodium guanosine 5-triphosphate (GTP), pH 7.3. Before the experiment, 0.5-1 mg/ml Lucifer yellow (potassium salt; Sigma) was added to the pipette solution. Seal formation between the tip of the electrode and the membrane of Fast-Blue-labeled neurons was directly visualized under a fluorescence microscope (×400; Leitz) and a successful recording was confirmed and validated if 1) Lucifer yellow filled the cell after the membrane patch was ruptured and 2) the membrane potential was more negative than 55 mV. An example is shown in Fig 1. All experiments were performed at ~25°C, temperature maintained by preheating the reservoir solutions. During the experiment, the slice was perfused constantly at a rate of 3-6 ml/min with a standard external solution that was composed of ACSF with the addition of 10⁶ M tetrodotoxin (TTX). All drugs were added into the perfusates of separate reservoirs at the final concentrations and applied to the slices by switching reserviors. The time required for the new solution to travel from its reservoir to the chamber was 15 s to 1 min. When tetraethylammonium (TEA) and CsCl were used, the equivalent amount of NaCl was reduced to maintain ionic strength and osmolarity. Clamped membrane voltages have been corrected off-line for junction potential that was measured according to Neher (1992).

View larger version (98K): [in this window] [in a new window]   FIG. 1. Fluorescence photomicrographs obtained during patch-clamp recording from 2 different neurons in the ventral nucleus of the trapezoid body (VNTB). a and c: each cell is shown labeled with Fast Blue before attachment of patch pipette. b and d: same 2 cells after rupture of seal and diffusion of Lucifer yellow from pipette into cell. Note filling of cell body and dendrites in b. In d, although filling of cell is incomplete, note persistence of Lucifer yellow fluorescence in recorded cell and severe fading of Fast Blue in nonrecorded cells (photo in d obtained 15 min after rupture of patch). Scale bar: 50 µm (applies to a-d).

Chemicals and statistics

NE, phenylephrine, phentolamine, and propranolol were from Sigma. These drugs were dissolved in distilled water at concentrations of 10³ M, stored in aliquots at less than 30°C, and diluted in ACSF immediately before use. Lucifier yellow, TEA chloride (TEA-Cl), EGTA, ATP, GTP, TTX, CsCl, CdCl₂, and BaCl₂ were from Sigma. Fast Blue was from Dr. Illing, GMBH, GrossUmstadt, Germany.

Unless otherwise stated, results are expressed as means ± SE (n), where n refers to number of neurons. Data measurements and analysis were performed with Clampfit (Axon Instruments). All statistics and curve fitting were performed with the use of Sigmaplot (version 3.0).

	RESULTS

Abstract Introduction Methods Results Discussion References

Whole cell currents

Whole cell recordings were made from 51 Fast-Blue-labeled neurons in VNTB. When the slice was perfused with the standard external solution, the average resting membrane potential was 64.2 ± 1.2 (SE) mV (n = 28) as measured in current-clamp mode immediately after establishment of the whole cell configuration. In voltage-clamp mode, whole cell currents, elicited with successive depolarizing and hyperpolarizing pulses from a holding potential of 55 mV, consisted of at least five components. A large fast TTX-sensitive inward current was activated at around 40 mV and had a maximum peak at about 20 mV. A slowly developed Cs⁺-sensitive inward current was activated at around 65 mV and a transient rapidly inactivated outward current was activated at around 45 mV and suppressed by 4-aminopyridine. Finally, a slowly inactivating outward current was activated at potentials more positive than 30 mV. This consisted of subcomponents that can be further dissected according to their sensitivities to blockage by 30 mM TEA. Dissections of these currents were demonstrated by examining the additive effects of known ion channel blockers. Typical whole cell recordings from a labeled neuron are shown in Fig. 2. The whole cell currents of these labeled medial olivocochlear neurons are generally similar to those commonly observed and are well characterized in neurons of other brain regions (see for example Hille 1992).

View larger version (28K): [in this window] [in a new window]   FIG. 2. Whole cell recordings in an identified auditory efferent neuron showing additive effects of known channel blockers. Recordings made with pipette filled with potassium gluconate solution. Slice perfused with standard external solution (A), after addition of 106 M tetrodotoxon (TTX; B), after addition of 2 mM CsCl (C), after addition of 100 µM 4-aminopyridine (D), and after 30 mM tetraethylammonium (TEA) replaced equal molar NaCl (E). Voltage protocol is shown in F. Expanded sections for A and B are given to show the effect of TTX on sodium currents.

The effect of NE was tested in 38 neurons. On switch to NE-containing perfusate for 30 s, two types of responses were observed. In 26 of 38 cells, when clamped continuously at 75 mV, NE induced an inward current that was responsible for membrane depolarization in current-clamp mode, whereas in 6 of 38 cells, an outward current to NE was observed that was responsible for membrane hyperpolarization. The remaining cells tested (n = 6) showed no detectable responses to NE.

Excitatory actions of NE

With the slice perfused by a standard external solution and the cell voltage clamped at a holding potential close to its membrane potential, bath application of micromolar concentrations of NE for 30 s induced an inward current that lasted several minutes and was fully reversible. A typical example is shown in Fig. 3A. The average amplitude of the inward currents induced by 5 µM NE at holding potentials of 55 and 75 mV were 91.8 ± 27.8 pA (n = 4) and 75.4 ± 27.8 pA (n = 9), respectively (Fig. 3B). The Ca²⁺ dependence of these reponses was tested in two cells. Although not blocked by Cd²⁺ (see below), the NE-induced inward current was attenuated to varying degrees in a zero-Ca²⁺ solution (the standard solution with Ca²⁺ omitted and Mg²⁺ raised to 4 mM). Recordings from one cell are shown in Fig. 3C.

View larger version (19K): [in this window] [in a new window]   FIG. 3. A: inward currents induced by 5 µM norepinephrine (NE) from the same neuron [resting membrane potential (Vm) = 64 mV] at holding potentials (i.e., pipette potentials; Vp) of 55 and 75 mV. B: comparison of NE-induced inward currents at 55 mV (4 cells) and 75 mV (9 cells;P = 0.034). C: NE-induced inward current in another cell was attenuated in 0-Ca2+ external solution switched on at the time indicated by upward arrow. In each case, NE was applied for 30 s as indicated.

The NE-induced inward current was further investigated with the use of a voltage protocol that allowed both inward and outward currents induced by a voltage step to be monitored (Fig. 4C). As shown in Fig. 3, the neuron tested was held at 75 mV. Both hyperpolarizing (to 127 mV) and depolarizing (to 15 mV) pulses of 120 ms were applied at 10-s intervals. An inward current was evoked by bath application of 1, 5, and 10 µM NE for 30 s. This baseline inward current was observed in all 26 cells. At the same time, the currents induced by both hyperpolarizing and depolarizing voltage steps were reduced in 17 neurons (Fig. 4) and were virtually unchanged in 10 of 26 neurons. A dose-response curve was constructed by measuring the net increase of the inward currents to various concentrations of NE at a holding potential of 75 mV and fitting to a hyperbolic equation (see legend for Fig. 4D). A half-activation concentration of 8.8 ± 1.4 µM was obtained with a maximum activation of 200.3 ± 11.2 pA.

View larger version (22K): [in this window] [in a new window]   View larger version (10K): [in this window] [in a new window]   FIG. 4. Typical changes in whole cell currents caused by bath application of 3 different concentrations of NE. Representative current traces are shown in A; their corresponding locations in the continuous recording (B) are indicated by the letters. Data in B are constructed from voltage-clamp recordings with the use of a voltage protocol (C) applied repeatedly. C: details of voltage protocol (bottom) and representative evoked current. Arrows indicate where the measurements were taken to construct elements of Fig. 4B with corresponding symbols. D: dose-response curve for NE-induced inward currents. Peak inward currents were measured after bath application of NE for 30 s and fitted with the rectangular hyperbolic function INE = CNE × Imax/(Kd + CNE), where INE is obtained by subtraction of the peak inward current from the baseline current, CNE is the concentration applied, Kd is the dissociation constant, and Imax is the predicted maximal inward current induced by NE. Number of experiments at each concentration indicated in parentheses. Kd and Imax were 8.8 ± 1.4 (SE) µM and 200.3 ± 11.2 pA, respectively.

With the use of the same voltage protocol (Fig. 4C), the effects of the following chemicals on NE-induced inward current were tested; the representative results are shown in Fig. 5. First, phenylephrine, a known ₁-receptor agonist, induced a response similar to NE in two of two cells tested. The response induced by NE was blocked by pretreatment with phentolamine (3 of 3 cells; Fig. 5A). Second, the inwardly rectifying current was blocked by 2 mM Cs⁺ in the perfusate, whereas NE-induced inward current still persisted in the presence of Cs⁺ (n = 1; Fig. 5B). Third, a similar inward current was also observed when 20 mM TEA was applied externally (3 of 3 cells); however, a small additive effect was seen in two experiments when 5 µM NE was added into this TEA-containing bathing solution (Fig. 5C). Finally, the NE-induced inward current appeared not to be blocked by 100 µM Cd²⁺ added into the bathing solution (3 of 3 cells; Fig. 5D).

View larger version (13K): [in this window] [in a new window]   View larger version (13K): [in this window] [in a new window]   FIG. 5. Representative whole cell current recordings from 4 labeled neurons (A-D). A: 2.5 µM phenylephrine induced a similar response to NE. NE-induced response was blocked when cell was pretreated with phentolamine. B: 10 µM NE induced an inward current in the presence of 2 mM Cs+ in the perfusate that blocked the inwardly rectifying K+ current. C: 20 mM TEA in bathing solution suppressed the voltage step-induced outward current and caused an inward shift of the baseline current. NE (5 µM) produced a further small shift of the baseline current. D: Cd2+ 100 µM had no detectable effect on NE-induced responses.

Current-voltage relationship of the excitatory responses

As a first attempt at revealing the current-voltage (I-V) relationship of the excitatory responses, a 500-ms voltage ramp was conducted before and after NE application under two different experimental conditions. Typical results are shown in Fig. 6. In three experiments, with the slice perfused with the standard external solution, the whole cell currents evoked from 111 to 42 mV were reduced by NE. The NE-induced current obtained by subtraction was active around the resting membrane potential and became more activated at more positive potentials. This current reversed at101 ± 3.7 mV (n = 4), which is close to the K⁺ equilibrium potential of 99 mV predicted under the experimental conditions (Fig. 6A). A further two experiments were carried out to examine the involvement of calcium channels in NE actions. The slices were perfused with a solution in which all the NaCl was replaced by an equal amount of TEA-Cl. Under these conditions, an inward current was evoked at voltages between 70 and 0 mV; this current could be abolished by 0.5 mM Cd²⁺ and it therefore appeared to consist of voltage-dependent Ca²⁺ currents uncovered by the replacement of NaCl with TEA-Cl. Although NE application did not produce any detectable change in the baseline current when the cell was held at 75 mV, a voltage ramp from 117 to 41 mV revealed a block of the Ca²⁺ channels by NE in two instances. In one of these cells, the block of Ca²⁺ conductance was prevented by pretreating the cell with 10 µM phentolamine. A representative recording is shown in Fig. 6B. The subtracted NE-induced current showed a voltage-dependent activation from 70 to 40 mV, with peak at around 50 mV. It is evident from these results that the NE-induced inward current is mainly due to a decrease of K⁺ conductance. In addition, NE also seemed to affect Ca²⁺ currents in these cells.

View larger version (14K): [in this window] [in a new window]   FIG. 6. Changes caused by bath application of 5 µM NE in whole cell currents elicited with voltage ramp (500 ms) under 2 different experimental conditions. A: in standard bathing solution, NE suppressed whole cell currents. NE-induced current, obtained by subtracting the control trace from the NE trace, is voltage dependent and reversed close to 100 mV. B: in a different neuron, with all NaCl in the bathing solution replaced by TEA chloride (TEA-Cl), the whole cell current elicited from 117 to 42 mV shows a prominent inward current with its peak at around 50 mV, which was attenuated by NE. Subtraction of the 2 traces revealed an outward current induced by NE, which activated at 70 mV, reached its peak at 50 mV and gradually declined to 0 at ~40 mV.

A further three experiments were carried out to examine the I-V relationship of the NE-induced responses with the use of a voltage protocol that consisted of successive hyperpolarizing and depolarizing pulses from a holding potential of 95 mV (Fig. 7B) while the slice was perfused with the standard external solution. A typical example is shown in Fig. 7. Under these conditions, a reduction of the outward currents was observed when 5 µM NE was applied into the bathing solution (Fig. 7A). In three experiments, the sustained outward currents were reduced by 30 ± 5.1%, which is significant as compared with the reduction in the transient outward currents by 18 ± 4.0% (P < 0.05). The subtracted currents that were blocked by NE are also shown in Fig. 7B. The I-V relationship of the NE-induced current (see Fig. 7C) was very similar to that obtained by ramp tests (Fig. 6A) in its voltage dependence and reversal potential, and both these techniques therefore strongly suggest that NE suppressed an outward K⁺ current.

View larger version (26K): [in this window] [in a new window]   FIG. 7. Whole cell currents elicited from 101 mV with the protocol shown below B. B: NE-sensitive whole cell currents obtained by subtracting the current recording made in presence of NE from control. C: current-voltage (I-V) relationship obtained by plotting the transient and the sustained outward currents against pipette potentials. Current measurements were taken at points indicated by symbols (A). NE-induced current was obtained by subtracting the control-sustained with NE-sustained curves.

NE-induced outward current (inhibitory action of NE)

An inhibitory action of NE was detected in 6 of 38 labeled neurons. A typical example is shown in Fig. 8. On application of NE in the bathing solution, there are at least two detectable changes in whole cell currents recorded with the use of the protocol mentioned previously (Fig. 4C): a positive shift of the baseline current when the cell was held at 75 mV and an increase in amplitude of the inward current elicited by hyperpolarizing pulses to 127 mV. At 5 µM NE, an average shift of the baseline current was 24 ± 5.2 pA (n = 4) and the maximal increase in the inward current was 57 ± 9.5% (n = 4). The time course of these responses was similar to that of the excitatory action of NE. In the same cell shown in Fig. 8, when tested in current-clamp mode, the cell resting membrane potential was hyperpolarized from 66 to 72 mV by 5 µM NE.

View larger version (22K): [in this window] [in a new window]   FIG. 8. Inhibitory action of NE. A: representative whole cell current traces taken from a continuous recording and superimposed. a-d: samples taken at times shown in B. B: whole cell currents recorded with the same protocol as shown in Fig. 4C and plotted against time. NE was applied 3 times at 3 concentrations as indicated. Major changes in whole cell currents are an increase in amplitude of the inward current that was elicited by hyperpolarizing pulses and a positive shift of the baseline current.

These responses seemed to be mediated by receptors on the recorded cell rather than by synaptic modulation, because they were still prominent in the Ca²⁺-free solution (data not shown). Both the baseline current shift and the increase of inward current to NE were abolished when pretreated with 5 µM phentolamine, an -receptor antagonist (2 of 2 cells). However, in both cases, the block by phentolamine was not reversible on wash for up to 1 h. In another cell that showed inhibitory responses, 1-10 µM phenylephrine failed to induce any detectable response. On the other hand, when cells were pretreated with 10 µM propranolol, a -receptor antagonist, the responses to NE were also greatly attenuated (2 of 2 cells; Fig. 9A). However, no response could be induced by 10 µM isoprenaline (2 of 2 cells), a -receptor agonist.

View larger version (20K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   FIG. 9. Effect of propranolol, Cs+, and phentolamine. A: inhibitory responses to NE were attenuated by propranolol and abolished by 5 mM Cs+ in the bathing solution. B: in another experiment, the inhibitory action was abolished by 0.5 mM Ba2+ added into the perfusate. C: inhibitory action of NE was blocked by phentolamine. Plots were constructed in the same way as those in Fig. 3B. Refer to Fig. 4C for details of the recording protocols.

The inward currents evoked by hyperpolarizing voltage steps were blocked by 2-5 mM Cs⁺, a blocker for the inward rectifying K⁺ channels, in the perfusate (n = 3), and also by 0.5 mM Ba²⁺ added in the bathing solution (n = 2). In these cases, the responses to NE were also abolished. It was noted throughout the experiments that although the inwardly rectifying current could be effectively blocked by up to 2 mM Cs²⁺, a complete block of NE-induced responses required as high as 5 mM Cs⁺. It was also observed that the blocking by Ba²⁺ was not fully reversible in both cells tested. Representative results are shown in Fig. 9.

The I-V relationships of the inhibitory responses to NE were determined with the use of approaches similar to those mentioned previously. When the cell was perfused with the standard external solution, I-V relations obtained by both successive voltage pulses (n = 2) and voltage ramp (n = 3) revealed a reversal potential of 95.6 ± 3.4 mV (n = 5). Combining the above results, the inhibitory action of NE appeared to involve mainly the activation of an inward rectifying current, which was prevented in the presence of 2-5 mM external Cs⁺. Representative results are shown in Fig. 10.

View larger version (19K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   FIG. 10. I-V relationship for inhibitory effect. A and B: whole cell currents, elicited with the protocol shown, recorded in absence (left) and presence (middle) of 5 µM NE when the slice was perfused with a standard external solution (A) and a Cs+-containing solution in which 5 mM NaCl was replaced by 5 mM CsCl (B). Right: current traces obtained by subtraction (middle  left). C: I-V curves constructed by measuring the currents at the end of each voltage pulse. I-V relationship of NE activated current was obtained by subtracting the NE curve from the control curve. D: whole cell currents elicited by voltage ramp before and after NE application. NE-activated current was obtained by subtracting the NE trace from the control trace.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

The auditory efferent pathways have been well defined by morphological studies and some functional aspects have been examined in living animals (Liberman and Brown 1986; Robertson and Gummer 1985, 1988). However, detailed investigations of signal transmission at the neuronal level of the efferent circuitry have been hampered by the fact that these neurons are few in number and dispersively located in the brain stem. In this study, we have for the first time demonstrated that patch-clamp recordings can be obtained directly from retrogradely labeled auditory efferent neurons in the VNTB. The excitability of these neurons was directly modulated by NE. There are at least two types of efferent neurons on the basis of their distinct responses to NE. These responses appeared to involve mainly regulation of membrane K⁺ conductance. In the majority of cells tested, NE selectively suppressed a type of voltage-dependent K⁺ channel, caused membrane depolarization, and increased the cell excitability, whereas in 15% of cells an inhibitory effect was revealed that appeared to be associated with an activation of an inward rectifying K⁺ channel.

There may be a number of uncontrolled factors contributing to the variability of responses to NE between cells. For example, it is possible that adrenergic receptors are not restricted to the somata of the neuron recorded. If this is the case, NE responses could be affected by poor space clamp on the distant synaptic sites, as has recently been reported for reconstructed hippocampal CA1 pyramidal neurons (Mainen et al. 1996). At present, however, there is no other information available on either the type or distribution of adrenoreceptors on the neurons we have studied.

In addition, we do not know to what extent desensitization of receptors occurred because of the method of drug delivery. Desensitization of adrenoreceptor-mediated responses has been observed with repetitive applications of NE in neuronal preparations (Diverse-Pierluissi et al. 1996; Scanziani et al. 1993). In our preliminary experiments, we observed that the excitatory responses to NE started to decay on reaching their peak at ~2-3 min during prolonged NE application. To minimize possible desensitization, we applied NE for 20 s to 1 min and allowed 5 min for washing before each repetitive test, which yielded good reproducibility of the NE response. Nevertheless, the relatively slow perfusion of NE made it difficult to detect rapid desensitization if NE was present.

NE decreases K⁺ currents and Ca²⁺ currents in efferent neurons

The NE-induced inward current observed in the present study was mimicked by phenylephrine and blocked by phentolamine. It was resistant to block by Cs⁺ and reversed at around 100 mV, which is close to the K⁺ equilibrium potential of 99 mV predicted in the experimental condition. These results are in agreement with those observed in cultured spinal cord neurons (Legendre et al. 1988), dorsal lateral geniculate neurons (McCormick 1992), dorsal raphe neurons (Pan et al. 1994), and facial motoneurons (Larkman and Kelly 1992), and this indicates that the excitatory actions of NE are caused mainly by decrease of an outward K⁺ current via ₁-adrenoreceptors.

On the other hand, the present results also revealed a voltage-dependent decrease in Ca²⁺ current component by NE. Inhibition of voltage-gated Ca²⁺ currents (mainly N type) has recently been reported in sympathetic neurons (Boehm et al. 1996; Carrier and Ikeda 1992), nucleus tractus solitarii neurons (Ishibashi and Akaike 1995), and olfactory bulb neurons (Bischofberger and Schild 1995; Trombley 1992), and is probably mediated by ₂-adrenoreceptors involving both G protein coupling (see also Hille et al. 1996) and phorbol-ester-insensitive protein kinase C (Boehm et al. 1996). It is likely that similar mechanisms are also present in the auditory efferent neurons in VNTB, and they may functionally participate in the regulation of excitability within the efferent circuits.

At the present stage it is still unknown what receptors are involved and whether and to what extent the inhibition of calcium channels contributes to the excitatory actions of NE. One mechanism that might be expected would be a reduction of Ca²⁺-activated K⁺ conductance resulting from the inhibition of Ca²⁺ currents by NE. In the present experiments, the NE-induced inward current showed dependence on the external Ca²⁺ and became smaller and less stable in a zero-Ca²⁺ solution. However, this appeared to be unrelated to the reduction ofCa²⁺ influx through voltage-gated Ca²⁺ channels, because the inward current still persisted in the presence of 100 µM Cd²⁺. A similar paradox has also beenreported by Legendre et al. (1988) on cultured spinalcord neurons and Pan et al. (1994) on dorsal raphe neurons, and a possible interaction between external Ca²⁺ and receptor binding has been suggested (Pan et al. 1994).

Inhibitory action of NE

In ~15% of identified efferent neurons in the present study, an inhibitory action of NE was clearly demonstrated that seemed to be associated with an enhancement of an inward rectifying K⁺ current. This result is similar to that reported in neurons of the locus coeruleus, dorsal motor nucleus of the vagus, and spinal cord (see Nicoll et al. 1990 for review), and recently in cerebral cortex neurons (Blanton and Kriegstein 1992) and mesopontine cholinergic neurons (Williams and Reiner 1993).

Attempts in this study to characterize the receptor involved in the inhibitory effect yielded results that are difficult to interpret. The rather small number of neurons exhibiting such an action of NE also made this aspect of the study inconclusive. Thus the effect was attenuated reversibly by propranolol (a -receptor antagonist), in conformity with our previous results obtained with the use of intracellular microelectrode recordings (Wang and Robertson 1997b), whereas it was not mimicked by isoprenaline (a -receptor agonist). The -receptor antagonist phentolamine blocked the effect, but this was not reversible at the concentrations employed. In addition, the ₁-receptor agonist phenylephrine had no effect in these cells, in contrast to its ability to reproducibly mimic the excitatory action of NE in other cells. Furthermore, we found that the concentration of Cs⁺ required to block the inhibitory action of NE was higher than that required to block the inwardly rectifying K⁺ current under control conditions. These results suggest that the inhibitory action of NE may be complicated and may involve more than one type of receptor or novel receptors so far not identified (see also Williams and Reiner 1993).

Functional importance

The existence of noradrenergic innervation of auditory structures in the mammalian brain, including the brain stem and the VNTB in particular, has been known for some time (Klepper and Herbert 1991; Wynne and Robertson 1996), and the functional role of this innervation is still under investigation. In anesthetized animal preparations, NE has been found to alter aspects of afferent neuronal properties in cochlear nucleus (Ebert 1996), possibly in the medial nucleus of trapezoid body (Banks et al. 1993), and in auditory cortex (Edeline 1995; Shinba et al. 1992). In this study we have investigated the effects of NE on a specific group of neurons in the VNTB that are labeled by intracochlear injection: the so-called "medial olivocochlear" neurons. These neurons innervate the peripheral receptor organ, the cochlea, where they terminate on the outer hair cells (see Warr 1992 for review). The medial olivocochlear system has also been divided into two groups on the basis of cell responses to sound delivered ipsilaterally or contralaterally to the target cochlea. These two groups are believed to be differentially located on either side of the brain stem with respect to their target (Liberman and Brown 1986; Robertson and Gummer 1985). It could be of interest to know whether these different groups respond differently to NE. Unfortunately, in the present study, we did not record the ipsilateral or contralateral location of the cells studied, and we therefore cannot draw conclusions on this issue.

Electrical stimulation of the medial efferents has been shown to exert inhibitory effects on the cochlear neural output, largely via their action on the micromechanical function of the outer hair cells. The medial olivocochlear efferent neurons have been shown to be driven by acoustic stimulation (Chery-Croze et al. 1993; Liberman and Brown 1986; Mott et al. 1989; Robertson 1984; Robertson and Gummer 1985, 1988) and therefore provide a substrate not only for centrally driven control of cochlear function but also for feedback regulation. However, attempts to establish their role in normal hearing with the use of animal preparations have been largely unsuccessful (Littman et al. 1992; Rajan et al. 1990). Recent interesting experiments in awake humans with vestibular nerve transection (deefferented cochleas) suggest a possible role for the efferents in peripheral lateral inhibitory processes subserving selective attention (Scharf et al. 1997).

If such a role for the medial efferents can be confirmed, it will become important to know the sorts of pharmacological influences on their excitability. In this study we provide clear evidence that NE has such an influence and we have gone some way toward clarifying the mechanisms by which NE exerts its excitatory and inhibitory actions on these neurons. Interestingly, medial olivocochlear neurons have recently also been shown to change their discharge characteristics as a result of sound conditioning (Kujawa et al. 1996). The cell biological basis of this plasticity is unknown, but the present results suggest that NE may be a candidate neuromodulator capable of altering medial olivocochlear neuron characteristics. An outstanding question that remains is the role of the inhibition of Ca²⁺ currents. Apart from direct actions such changes may have on the excitability of action-potential-initiating zones through effects on Ca²⁺-activated K⁺ channels, the role of such Ca²⁺ current inhibition in regulating transmitter release at presynaptic sites (Boehm et al. 1996; Hille et al. 1996) within the VNTB circuitry needs further investigation.

	ACKNOWLEDGEMENTS

  The authors thank G. Bennett for preparation of solutions and care of experimental animals.

  This work was supported by grants from the Australia Research Council and the University of Western Australia.

	FOOTNOTES

  Address reprint requests to D. Robertson.

  Received 8 April 1997; accepted in final form 12 June 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ASCHOFF, A., OSTWALD, J. Distribution of cochlear efferents and olivo-collicular neurons in the brainstem of rat and guinea pig. Exp. Brain Res. 71: 241-251, 1988.[Medline]
• BANKS, M. I., PEARCE, R. A., SMITH, R. H. Hyperpolarization-activated cation current (I_h) in neurons of the medial nucleus of the trapezoid body: voltage-clamp analysis and enhancement by norepinephrine and cAMP suggest a modulatory mechanism in the auditory brain stem. J. Neurophysiol. 70: 1420-1432, 1993.[Abstract/Free Full Text]
• BISCHOFBERGER, J., SCHILD, D. Different spatial patterns of [Ca²⁺] increase caused by N- and L-type Ca²⁺ channel activation in frog olfactory bulb neurones. J. Physiol. (Lond.) 487: 305-317, 1995.[Medline]
• BLANTON, K. J., KRIEGSTEIN, A. R. Norepinephrine activates potassium conductance in neurons of the turtle cerebral cortex. Brain Res. 570: 42-48, 1992.[Medline]
• BOEHM, S., HUCK, S., FREISSMUTH, M. Involvement of a phorbol ester-insensitive protein kinase C in the ₂-adrenergic inhibition of voltage-gated calcium current in chick -conotoxin-sensitive voltage operated calcium channels in vertebrate cells. J. Neurosci. 16: 4596-4603, 1996.[Abstract/Free Full Text]
• CARRIER, G. O., IKEDA, S. R. TTX-sensitive Na⁺ channels and Ca²⁺ channels of the L- and N-type underlie the inward current in acutely dispersed coeliac-mesenteric ganglia neurons of adult rats. Pflügers Arch. 421: 7-16, 1992.[Medline]
• CHERY-CROZE, S., MOULIN, A., COLLETT, L. Effects of contralateral sound stimulation on the distortion product 2f₁-f₂ in humans: evidence of a frequency specificity. Hear. Res. 68: 53-58, 1993.[Medline]
• COLLET, L., KEMP, D. T., VEUILLET, E., DUCLAUX, R., MOULIN, A., MORGON, A. Effect of contralateral auditory stimuli on active cochlear micro-mechanical properties in human subjects. Hear. Res. 43: 251-261, 1990.[Medline]
• DIVERSE-PIERLUISSI, M., INGLESE, J., STOFFEL, R. H., LEFKOWITZ, R. J., DUNLAP, K. G protein-coupled receptor kinase mediates desensitization of norepinephrine-induced Ca²⁺ channel inhibition. Neuron 16: 579-585, 1996.[Medline]
• EBERT, U. Noradrenaline enhances the activity of cochlear nucleus neurons in the rats. Eur. J. Neurosci. 8: 1306-1314, 1996.[Medline]
• EDELINE, J. M. The alpha 2-adrenergic antagonist idazoxan enhances the frequency selectivity and increases the threshold of auditory cortex neurons. Exp. Brain Res. 107: 221-240, 1995.[Medline]
• GIRAUD, A. L., COLLET, L., CHERY-CROZE, S., MAGNAN, J., CHAYS, A. Evidence of a medial olivocochlear involvement in contralateral suppression of otoacoustic emissions in humans. Brain Res. 705: 15-23, 1995.[Medline]
• GUINAN, J. J., GIFFORD, M. L. Effects of electrical stimulation of efferent olivocochlear neurons on cat auditory-nerve fibers. I. Rate-level functions. Hear. Res. 33: 97-113, 1988a.[Medline]
• GUINAN, J. J., GIFFORD, M. L. Effects of electrical stimulation of efferent olivocochlear neurons on cat auditory-nerve fibers. II. Spontaneous rate. Hear. Res. 33: 115-127, 1988b.[Medline]
• GUINAN, J. J., GIFFORD, M. L. Effects of electrical stimulation of efferent olivocochlear neurons on cat auditory-nerve fibers. III. Tuning curves and thresholds at CF. Hear. Res. 37: 29-45, 1988c.[Medline]
• HILLE, B. In: Ionic Channels of Excitable Membranes, , 2nd ed.. Sunderland, MA: Sinauer, 1992, p. 115-139
• HILLE, B., SCHEUER, T., CATTERALL, W. A. Modulation of Ca²⁺ channels by G-protein beta gamma subunits. Nature 380: 258-262, 1996.[Medline]
• ISHIBASHI, H., AKAIKE, N. Norepinephrine modulates high voltage-activated calcium channels in freshly dissociated rat nucleus tractus solitarii neurons. Neuroscience 68: 1139-1146, 1995.[Medline]
• KAWASE, T., TAKASAKA, T. The effects of contralateral noise on masked compound action potential in humans. Hear. Res. 91: 1-6, 1995.[Medline]
• KLEPPER, A., HERBERT, H. Distribution and origin of noradrenergic and serotonergic fibres in the cochlear nucleus and inferior colliculus of the rat. Brain Res. 557: 190-201, 1991.[Medline]
• KUJAWA, S. G., BROWN, M. C., LIBERMAN, M. C. Sound conditioning enhances reponses and peripheral effects of olivocochlear neurons. Soc. Neurosci. Abstr. 22: 648, 1996.
• LARKMAN, P. M., KELLY, J. S. Ionic mechanisms mediating 5-hydroxytryptamine- and noradrenaline-evoked depolarization of adult rat facial motoneurones. J. Physiol. (Lond.) 456: 473-490, 1992.[Abstract/Free Full Text]
• LEGENDRE, P., DUPOUY, B., VINCENT, J. D. Excitatory effect of noradrenaline on pacemaker cells in spinal cord primary cultures. Neuroscience 24: 647-685, 1988.[Medline]
• LIBERMAN, M. C., BROWN, M. C. Physiology and anatomy of single olivocochlear neurons in the cat. Hear. Res. 24: 17-36, 1986.[Medline]
• LITTMAN, J. A., CULLEN, J. K., BOBBIN, R. P. The effect of olivocochlear bundle transection on tuning curves and otoacoustic distortion products. J. Acoust. Soc. Am. 92: 1945-1952, 1992.[Medline]
• MAINEN, Z. F., CARNEVALE, N. T., ZADOR, A. M., CLAIBORNE, B. J., BROWN, T. H. Electrotonic architecture of hippocampal CA1 pyramidal neurons based on three-dimensional reconstructions. J. Neurophysiol. 76: 1904-1923, 1996.[Abstract/Free Full Text]
• MCCORMICK, D. A. Cellular mechanisms underlying cholinergic and noradrenergic modulation of neuronal firing mode in the cat and guinea pig dorsal lateral geniculate nucleus. J. Neurosci. 12: 278-289, 1992.[Abstract]
• MOTT, J. B., NORTON, S. J., NEELY, S. T., WARR, W. B. Changes in spontaneous otoacoustic emissions produced by acoustic stimulation of the contralateral ear. Hear. Res. 38: 229-242, 1989.[Medline]
• NEHER, E. Correction for liquid junction potentials in patch clamp experiments. In: Ion channels. Methods Enzymol. 207: 123-131, 1992.
• NICOLL, R. A., MALENKA, R. C., KAUER, J. A. Functional comparison of neurotransmitter receptor subtypes in mammalian central nervous system. Physiol. Rev. 70: 513-565, 1990.[Free Full Text]
• PAN, Z. Z., GRUDT, T. J., WILLIAMS, J. T. ₁-Adrenoceptors in rat dorsal raphe neurons: regulation of two potassium conductances. J. Physiol. (Lond.) 478: 437-447, 1994.[Medline]
• PATUZZI, R. B., THOMPSON, M. L. Cochlear efferent neurons and protection against acoustic trauma: protection of outer hair cell receptor current and interanimal variability. Hear. Res. 54: 45-58, 1991.[Medline]
• RAJAN, R. Effect of electrical stimulation of the crossed olivocochlear bundle on temporary threshold shifts in auditory sensitivity. I. Dependence on electrical stimulation parameters. J. Neurophysiol. 60: 549-568, 1988.[Abstract/Free Full Text]
• RAJAN, R., ROBERTSON, D., JOHNSTONE, B. M. Absence of tonic activity of the crossed olivocochlear bundle in determining compound action potential thresholds, amplitudes and masking phenomena in anaesthetized guinea pigs with normal hearing sensitivities. Hear. Res. 44: 195-208, 1990.[Medline]
• ROBERTSON, D. Horseradish peroxidase injection of physiologically characterised afferent and efferent neurons in the guinea pig spiral ganglion. Hear. Res. 15: 113-121, 1984.[Medline]
• ROBERTSON, D. Physiology and morphology of cells in the ventral nucleus of the trapezoid body and rostral periolivary regions of the rat superior olivary complex studied in slices. Auditory Neurosci. 2: 12-31, 1996.
• ROBERTSON D AND GUMMER, M. Physiological and morphological characterization of efferent neurones in the guinea-pig cochlea. Hear. Res. 20: 63-77, 1985.[Medline]
• ROBERTSON, D., GUMMER, M. Physiology of cochlear efferents in the mammal. In: Auditory Pathway, edited by J. Syka, and R. B. Masterton . New York: Plenum, 1988, p. 269-278
• ROBERTSON, D., HARVEY, A. R., COLE, K. S. Postnatal development of the efferent innervation of the rat cochlea. Dev. Brain Res. 47: 197-207, 1989.[Medline]
• SCANZIANI, M., GAHWILER, B. H., THOMPSON, S. M. Presynaptic inhibition of excitatory synaptic transmission mediated by alpha adrenoreceptors in area CA3 of the rat hippocampus in vitro. J. Neurosci. 13: 5393-5401, 1993.[Abstract]
• SCHARF, B., MAGNAN, J., CHAYS, A. On the role of the olivocochlear bundle in hearing: 16 cases. Hear. Res. 103: 101-122, 1997.[Medline]
• SHINBA, T., ANDO, Y., OZAWA, N., YAMAMOTO, K. Auditory-evoked response of the cortex after yohimbine administration: phase advance effect of central noradrenergic activation. Brain Res. Bull. 28: 463-471, 1992.[Medline]
• TROMBLEY, P. Q. Norepinephrine inhibits calcium currents and EPSPs via a G-protein-coupled mechanism in olfactory bulb neurons. J. Neurosci. 12: 3992-3998, 1992.[Abstract]
• VETTER, D. E., SALDANA, E., MUGNAINI, E. Inputs from the inferior colliculus to medial olivocochlear neurons in the rat: a double label study with PHA-L and cholera toxin. Hear. Res. 70: 173-186, 1993.[Medline]
• WANG, X. Y., ROBERTSON, D. Patch clamp study of the effect of noradrenaline on auditory efferent neurons in rat brainstem slices (Abstract). Proc. Aust. Neurosci. Soc. 8: 129, 1997a.
• WANG, X. Y., ROBERTSON, D. Effects of bioamines and peptides on neurons in the ventral nucleus of trapezoid body and rostral periolivary regions of the rat superior olivary complex: an in vitro investigation. Hear. Res. 106: 20-28, 1997b.[Medline]
• WARR, W. B. Organization of olivocochlear efferent systems in mammals. In: Mammalian Auditory Pathway: Neuroanatomy, Springer Handbook of Auditory Research, edited by D. B. Webster, A. N. Popper, and R. R. Fay . New York: Springer-Verlag, 1992, p. 410-448
• WARR, W. B., GUINAN, J. J. Efferent innervation of the organ of Corti: two separate systems. Brain Res. 173: 152-155, 1979.[Medline]
• WARREN, E. H., LIBERMAN, M. C. Effects of contralateral sound on auditory-nerve responses. I. Contributions of cochlear efferents. Hear. Res. 37: 89-104, 1989.[Medline]
• WHITE, J. S., WARR, W. N. Dual origins of the olivocochlear bundle in the albino rat. J. Comp. Neurol. 219: 203-214, 1983.[Medline]
• WILLIAMS, J. A., REINER, P. B. Noradrenaline hyperpolarizes identified rat mesopontine cholinergic neurons in vitro. J. Neurosci. 13: 3878-3883, 1993.[Abstract]
• WYNNE, B., ROBERTSON, D. Localization of dopamine--hydroxylase-like immunoreaactivity in the superior olivary complex of the rat. Audiol. Neurootol. 1: 54-64, 1996.

0022-3077/97 $5.00 Copyright ©1997 The American Physiological Society