Spontaneous firing patterns in the medial geniculate nucleus in a guinea pig model of tinnitus.

The mechanism of tinnitus, the perception of sound in the absence of acoustic stimulation, remains as yet unknown. It has been proposed that tinnitus is caused by altered spontaneous activity in the auditory pathway following cochlear damage in combination with inadequate gating at the level of the auditory thalamus, the medial geniculate nucleus (MGN). To investigate this further we made electrophysiological recordings in MGN of guinea pigs (n = 9) with and without tinnitus after acoustic trauma (continuous loud tone at 10 kHz, 124 dB SPL for 2 h). Parameters of interest were spontaneous tonic and burst firing. After acoustic trauma, 5 out of 9 guinea pigs developed signs of tinnitus as determined by the gap prepulse inhibition of acoustic startle. Spontaneous firing rates were significantly increased in the tinnitus animals as compared to the non-tinnitus animals and this change was specific to pure-tone responsive MGN neurons. However, burst firing parameters, including number of bursts per minute, burst duration, number of spikes in each burst, and percentage of spikes occurring in a burst, were not different between tinnitus and non-tinnitus animals. In addition, our data showed a strong dependence of spontaneous firing rates with heart rate, which implies that monitoring physiological status in animals is pertinent to obtaining reliable data when recording at higher levels of the auditory pathway. Our results suggest that increases in the tonic spontaneous fining rate of pure-tone responsive MGN neurons but not changes in burst firing parameters, are a robust neural signature of tinnitus in anaesthetised animals.

Changes in auditory thalamus neural firing patterns after acoustic trauma in rats.

Tinnitus is an abnormal phantom perception associated with cochlear trauma, and is thought to cause changes in the rates and patterns of firing neurons in the central auditory pathway. Recent studies have suggested a key role for the auditory thalamus, the medial geniculate nucleus (MGN), in the generation of tinnitus as it may serve a gating function for information en route to cortex. Dysfunctional gating would lead to abnormal activity reaching cortex and hence inappropriate perception, tinnitus, would occur. In this study we compared spontaneous MGN firing rates and burst firing parameters in Wistar rats with and without behavioural evidence of tinnitus following an acoustic trauma. Data were also compared with animals subjected to sham surgery and at an early time-point (2 weeks) after acoustic trauma. Acoustic trauma resulted in a temporary but not a permanent threshold loss and no differences were found in spontaneous firing rate between any of the groups. However, acoustic trauma, whether resulting in tinnitus or not, was accompanied by a significant decrease in the percentage of neurons showing burst firing. In bursting neurons, the number of spikes occurring in a burst and the number of burst per minutes was also significantly reduced compared to the sham group. Our results show that in our rat model without permanent threshold loss, elevated spontaneous firing rates are not associated with acoustic trauma and/or tinnitus and that burst firing parameters are associated with acoustic trauma but are not a neural signature for tinnitus.

Medial geniculate neurons show diverse effects in response to electrical stimulation of prefrontal cortex.

Phantom perceptions have been proposed to arise due to dysfunctional sensory gating at the level of the thalamus. Recently, it has been suggested that tinnitus, a phantom perception of sound, may arise from altered cortico-limbic circuitry and its connection with the auditory thalamus, the medial geniculate nucleus (MGN). Indeed, some elements of this cortico-limbic circuitry, such as the prefrontal cortex (PFC), as well as elements of the auditory pathway, have been shown to be altered in humans with tinnitus. However, the functional connectivity between PFC and MGN has not yet been explored. We therefore investigated the effects of activation of the PFC on neuronal activity in MGN in normal anaesthetized Wistar rats. Bipolar electrical stimulation was delivered to the PFC while recording single neuron activity in the MGN. The majority (81%) of MGN neurons sampled showed a change in their spontaneous firing rate in response to electrical stimulation of the PFC. The effects observed varied greatly between neurons and included combinations of inhibitory and excitatory effects with a wide range of latencies. The effects were not dependent on acoustic response type or MGN subdivision. These data demonstrate that PFC activation can modulate MGN neuronal activity and this connection could potentially play a role in sensory gating of auditory signals.

Effects of pulsatile electrical stimulation of the round window on central hyperactivity after cochlear trauma in guinea pig.

Partial hearing loss induced by acoustic trauma has been shown in animal models to result in an increased spontaneous firing rate in central auditory structures. This so-called hyperactivity has been suggested to be involved in the generation of tinnitus, a phantom auditory sensation. Although there is no universal cure for tinnitus, electrical stimulation of the cochlea, as achieved by a cochlear implant, can result in significant reduction of the tinnitus percept. However, the mechanism by which this tinnitus suppression occurs is as yet unknown and furthermore cochlear implantation may not be an optimal treatment option for tinnitus sufferers who are not profoundly deaf. A better understanding of the mechanism of tinnitus suppression by electrical stimulation of the cochlea, may lead to the development of more specialised devices for those for whom a cochlear implant is not appropriate. This study aimed to investigate the effects of electrical stimulation in the form of brief biphasic shocks delivered to the round window of the cochlea on the spontaneous firing rates of hyperactive inferior colliculus neurons following acoustic trauma in guinea pigs. Effects during the stimulation itself included both inhibition and excitation but spontaneous firing was suppressed for up to hundreds of ms after the cessation of the shock train in all sampled hyperactive neurons. Pharmacological block of olivocochlear efferent action on outer hair cells did not eliminate the prolonged suppression observed in inferior colliculus neurons, and it is therefore likely that activation of the afferent pathways is responsible for the central effects observed.

Modulation of medial geniculate nucleus neuronal activity by electrical stimulation of the nucleus accumbens.

Dysfunctional sensory gating has been proposed to result in the generation of phantom perceptions. In agreement, it has been recently suggested that tinnitus, a phantom perception of sound commonly associated with hearing loss, is the result of a breakdown of circuitry involving the limbic system and the medial geniculate nucleus (MGN) of the thalamus. In humans with tinnitus, structural changes and abnormal activity have been found to occur in the auditory pathway as well as parts of the limbic system such as the nucleus accumbens (NAc). However, at present, no studies have been conducted on the influence of the NAc on the MGN. We investigated the functional connectivity between the NAc and MGN single neurons. Bipolar electrical stimulation was delivered to the NAc while recording single neuron activity in MGN in anesthetized Wistar rats. Histological analysis was used to confirm placement of electrodes. NAc electrical stimulation generally decreased spontaneous firing rates in MGN neurons and, in a limited number of neurons, caused an increase in firing rate. This suggests that NAc can modulate the activity of auditory neurons in the MGN and may play a role in the development of tinnitus.

Suppression of putative tinnitus-related activity by extra-cochlear electrical stimulation.

Studies on animals have shown that noise-induced hearing loss is followed by an increase of spontaneous firing at several stages of the central auditory system. This central hyperactivity has been suggested to underpin the perception of tinnitus. It was shown that decreasing cochlear activity can abolish the noise-induced central hyperactivity. This latter result further suggests that an approach consisting of reducing cochlear activity may provide a therapeutic avenue for tinnitus. In this context, extra-cochlear electric stimulation (ECES) may be a good candidate to modulate cochlear activity and suppress tinnitus. Indeed, it has been shown that a positive current applied at the round window reduces cochlear nerve activity and can suppress tinnitus reliably in tinnitus subjects. The present study investigates whether ECES with a positive current can abolish the noise-induced central hyperactivity, i.e., the putative tinnitus-related activity. Spontaneous and stimulus-evoked neural activity before, during and after ECES was assessed from single-unit recordings in the inferior colliculus of anesthetized guinea pigs. We found that ECES with positive current significantly decreases the spontaneous firing rate of neurons with high characteristic frequencies, whereas negative current produces the opposite effect. The effects of the ECES are absent or even reversed for neurons with low characteristic frequencies. Importantly, ECES with positive current had only a marginal effect on thresholds and tone-induced activity of collicular neurons, suggesting that the main action of positive current is to modulate the spontaneous firing. Overall, cochlear electrical stimulation may be a viable approach for suppressing some forms of (peripheral-dependent) tinnitus.

Hyperactivity following unilateral hearing loss in characterized cells in the inferior colliculus.

Hyperactivity (increased spontaneous firing rates) following cochlear trauma and hearing loss has been well documented in the inferior colliculus (IC). This hyperactivity is associated with frequency regions in the IC that are closely related to regions of peripheral hearing loss. In other auditory nuclei, notably cochlear nucleus, hyperactivity has been shown to be more prevalent in particular cell types but this has not been investigated in the IC. Single-neuron spontaneous firing rates were recorded in the IC of animals after acoustic trauma (10-kHz tone at 124dB for 2h) and in sham surgery controls. Single-neuron recordings were made 2weeks later. Evoked responses to ipsi- and contralateral sound were used for classification. Classifications were based on peri-stimulus time histograms, input-output functions, frequency response areas and monaural/binaural responses. Results showed increased spontaneous firing rates in the IC following trauma, in regions corresponding to the frequencies at which there was peripheral hearing loss (12-20kHz). Most response categories, with the exception of cells showing an onset response classification, showed a significantly increased average spontaneous firing rate. These data suggest that hyperactivity in the IC is not confined to a particular response type in contrast to findings in the cochlear nucleus. This may be the result of factors intrinsic to the IC, or because of convergent input to the IC from a range of other auditory structures.

Development of hyperactivity after acoustic trauma in the guinea pig inferior colliculus.

The time of onset of hyperactivity (increased spontaneous firing rates) was investigated by single neuron recording in the inferior colliculus (IC) of guinea pigs subjected to unilateral acoustic trauma (exposure to a loud 10 kHz tone). Hyperactivity was present by 12 h post acoustic trauma whereas data obtained within approximately 4 h of the cessation of acoustic trauma found no evidence of hyperactivity. These data suggest that hyperactivity in the IC begins at some time between 4 and 12 h post trauma and is a relatively rapid plastic event beginning within hours rather than days post cochlear trauma. This is consistent with results reported in the cat auditory cortex (Norena and Eggermont, 2003). Hyperactivity did not show any further systematic increase between 12 h and up to 2 weeks post acoustic trauma. At recovery times of 12 and 24 h hyperactivity was widespread across most regions of the IC but at longer recovery times, it became progressively more restricted to ventral regions corresponding to the regions of the cochlea where there was persistent damage.

Progressive centralization of midbrain hyperactivity after acoustic trauma.

Partial hearing loss is known to cause increased spontaneous activity at several stages of the central auditory pathways, and this phenomenon has been suggested as a possible neural substrate for tinnitus, a phantom hearing sensation. One recent study in guinea pig has suggested that approximately 6 weeks after acoustic trauma, the increased spontaneous activity in inferior colliculus is not intrinsically generated in the central nucleus but is dependent on afferent input from the cochlea. This was unexpected in view of the fact that tinnitus in human patients can persist after severing of the auditory nerve. In this study, we show that when recovery time after acoustic trauma is extended to 8 and 12 weeks, cochlear ablation does not significantly decrease the increased spontaneous activity measured in the inferior colliculus. This result demonstrates for the first time that central hyperactivity that develops after acoustic trauma transitions from an early stage when it is dependent on continued peripheral afferent input to a later stage in which the hyperactivity is intrinsically generated within the central nervous system.

Relationship between auditory thresholds, central spontaneous activity, and hair cell loss after acoustic trauma.

Acoustic trauma caused by exposure to a very loud sound increases spontaneous activity in central auditory structures such as the inferior colliculus. This hyperactivity has been suggested as a neural substrate for tinnitus, a phantom hearing sensation. In previous studies we have described a tentative link between the frequency region of hearing impairment and the corresponding tonotopic regions in the inferior colliculus showing hyperactivity. In this study we further investigated the relationship between cochlear compound action potential threshold loss, cochlear outer and inner hair cell loss, and central hyperactivity in inferior colliculus of guinea pigs. Two weeks after a 10-kHz pure tone acoustic trauma, a tight relationship was demonstrated between the frequency region of compound action potential threshold loss and frequency regions in the inferior colliculus showing hyperactivity. Extending the duration of the acoustic trauma from 1 to 2 hours did not result in significant increases in final cochlear threshold loss, but did result in a further increase of spontaneous firing rates in the inferior colliculus. Interestingly, hair cell loss was not present in the frequency regions where elevated cochlear thresholds and central hyperactivity were measured, suggesting that subtle changes in hair cell or primary afferent neural function are sufficient for central hyperactivity to be triggered and maintained.

Tonotopic changes in GABA receptor expression in guinea pig inferior colliculus after partial unilateral hearing loss.

Immunohistochemistry was used to investigate the topographic distribution of the alpha1 subunit of the GABA receptor (GABRA1) in guinea pig inferior colliculus after treatments that caused a unilateral loss of peripheral neural sensitivity in the high-frequency regions of the cochlea. Both forms of treatment (direct mechanical lesion of the cochlea and acoustic overstimulation) resulted in a significant decrease in GABRA1 labeling in regions of the contralateral inferior colliculus in which high-frequency sound stimuli are represented. This localized region of reduced inhibitory receptor expression corresponds to the region in which hyperactivity of inferior colliculus neurons has been shown to develop after such treatments. The results strengthen the notion of a causal link between reduced GABRA1 expression and neural hyperactivity in central auditory nuclei and provide a possible mechanism for the development of phantom auditory sensations, or tinnitus.

Hyperactivity in the auditory midbrain after acoustic trauma: dependence on cochlear activity.

Plasticity in the adult mammalian brain can occur after damage to peripheral nerves and has also been described in the auditory system. Acoustic trauma, resulting in a loss of cochlear sensitivity, can lead to elevated levels of spontaneous activity, that is hyperactivity, in central nuclei such as the inferior colliculus. The current view is that this hyperactivity is centrally generated as a result of altered input. We investigated acute and chronic effects of acoustic trauma on cochlear sensitivity and development of hyperactivity in the inferior colliculus of guinea pigs. In addition, we investigated whether hyperactivity in the inferior colliculus, once established, is dependent on neural activity in the cochlea. Acoustic trauma (1 h continuous, 10 kHz tone at 124 dB SPL) resulted in a small but permanent, frequency restricted threshold loss in the cochlea up to 6 weeks post-exposure (maximum recovery time used). This was accompanied by hyperactivity in restricted frequency areas of the inferior colliculus, broadly corresponding to the cochlear threshold loss. We found that hyperactivity in the inferior colliculus depended on neural activity in the cochlea at all recovery times, since it disappeared after cochlear ablation and treatments blocking spontaneous firing of primary afferents. We suggest that the dependency of the central hyperactivity on the integrity of the peripheral receptor indicates hyperexcitability within the CNS resulting in greater neuronal firing in response to normal levels of peripheral spontaneous activity.

Synaptic responses in cochlear nucleus neurons evoked by activation of the olivocochlear system.

The action of olivocochlear collaterals to the cochlear nucleus is not fully established. Synaptic ultrastructure suggests an excitatory role. Extracellular recordings show spikes evoked by electrical stimulation of olivocochlear axons, but these spikes in the cochlear nucleus may be antidromic (activation of output axons) or orthodromic (synaptic input). We therefore recorded intracellular responses to shocks to olivocochlear axons in anaesthetized guinea pigs. In chopper and primary-like neurons shocks caused either no response or an inhibitory synaptic response (IPSP), but never an excitatory one (EPSP). In contrast, onset neurons never showed IPSPs but showed a variety of other responses; antidromic spikes, EPSPs, orthodromic spikes or no effect. The results agree with earlier extracellular observations in that olivocochlear collaterals provide excitatory input to onset neurons. Because some onset neurons are inhibitory they may be the source of the IPSPs observed in other cochlear nucleus neurons. The data also show that electrical stimulation at the floor of the IVth ventricle results in antidromic spikes as well. However, intracellular recording enabled the orthodromic action to be verified and the presumed olivocochlear action to be better understood. Our data support the hypothesis that olivocochlear collaterals initiate excitatory input onto onset-chopper neurons.

Changes in neuronal activity and gene expression in guinea-pig auditory brainstem after unilateral partial hearing loss.

Spontaneous neural hyperactivity in the central auditory pathway is often associated with deafness, the most common form of which is partial hearing loss. We quantified both peripheral hearing loss and spontaneous activity in single neurons of the contralateral inferior colliculus in a guinea-pig model 1 week after a unilateral partial deafness induced by cochlear mechanical lesion. We also measured mRNA levels of candidate genes in the same animals using quantitative real-time PCR. Spontaneous hyperactivity was most marked in the frequency region of the peripheral hearing loss. Expression of glutamate decarboxylase 1 (GAD1), GABA-A receptor subunit alpha-1 (GABRA1), and potassium channel subfamily K member 15 (KCNK15) was decreased ipsilaterally in the cochlear nucleus and bilaterally in the inferior colliculus. A member of RAB family of small GTPase (RAB3A) was decreased in both ipsilateral cochlear nucleus and contralateral inferior colliculus. RAB3 GTPase activating protein subunit 1 (RAB3GAP1) and glycine receptor subunit alpha-1 (GLRA1) were reduced ipsilaterally in the cochlear nucleus only. These results suggest that a decrease in inhibitory neurotransmission and an increase in membrane excitability may contribute to elevated neuronal spontaneous activity in the auditory brainstem following unilateral partial hearing loss.

Effects of centrifugal pathways on responses of cochlear nucleus neurons to signals in noise.

The medial olivocochlear (MOC) system, which originates in the brainstem and projects to the outer hair cells in the cochlea, is thought to be involved in improving signal detection in noisy backgrounds. This proposition arises from the observation that the input-output functions of auditory primary afferent fibres to pure tones recorded in a continuous background noise are unmasked by MOC activation, improving the dynamic range, and is supported by both animal and human behavioural experiments. However, it is not known how the unmasking effects observed in the cochlea are translated into higher auditory brain centres, such as the cochlear nucleus, where intrinsic circuitry can potentially modulate any effect. In this study we have investigated the effects of continuous background noise without and with MOC system activation, on responses of different neuron types in the ventral cochlear nucleus of the guinea pig. Results show that the unmasking effects of MOC system activation on tone responses in continuous background noise are present in the cochlear nucleus. These unmasking effects manifest themselves as decompression of input-output functions as well as an improved slope, which results in an improvement in intensity discrimination of the tones. The data show, however, that the strength of the unmasking effects of MOC system activation varies between the different neuronal types. Unmasking was not detected in onset chopper neurons despite its demonstrable presence in other neuronal types in the same animals. These observations may reflect the level of involvement of different neuronal types in intensity discrimination.

Gentamicin abolishes all cochlear effects of electrical stimulation of the inferior colliculus.

Electrical stimulation of the inferior colliculus (IC) has been shown to result in suppression of cochlear output, due to activation of the medial olivocochlear system. This auditory efferent system originates in the brainstem and terminates on the outer hair cells in the cochlea. Recently, excitatory effects of IC stimulation have also been reported, both on cochlear gross potentials and on primary auditory afferents. It has been hypothesized that this excitation is due to co-activation of the lateral olivocochlear system, which synapses on the primary auditory afferent fibres contacting the inner hair cells. If stimulation of the IC leads to the activation of both the medial and lateral olivocochlear system, resulting in a mixture of inhibitory and excitatory effects in the cochlea, then removal of the inhibitory effects, by blocking the medial system, should lead to more pronounced excitatory effects out in the periphery. To investigate this hypothesis, we recorded the effect of IC stimulation on cochlear gross potentials as well as on single auditory primary afferents in guinea pigs following block of the medial olivocochlear system with gentamicin. We found that administration of gentamicin, whether intraperitoneally or by intracochlear perfusion, blocked all effects of IC stimulation, whether inhibitory or excitatory. These data strongly suggest that all effects observed after IC stimulation, both inhibitory as well as excitatory, are due to the activation of the medial olivocochlear system.

Catecholaminergic innervation of guinea pig superior olivary complex.

In mammals, olivocochlear neurons in the superior olivary complex project to the cochlea, providing input to outer hair cells and auditory afferents contacting inner hair cells. In the rat it has been demonstrated that olivocochlear neurons receive noradrenergic input, arising from the locus coeruleus and it has been demonstrated in this species using in vitro brain slices that noradrenaline exerts a direct, mostly excitatory effect on an olivocochlear subpopulation. The guinea pig is a more commonly used animal in auditory physiology than the rat and anatomical data on noradrenaline in the auditory brainstem in this species are lacking. Because it has been shown that a compact locus coeruleus is not present in the guinea pig, subtle species differences might be expected. Therefore, using immunohistochemical and tracing techniques we have investigated in the guinea pig (1) the noradrenergic and dopaminergic innervation of the superior olivary complex, (2) the anatomical relationship between noradrenergic fibres and olivocochlear neurons and (3) the origin of the noradrenergic input to this brainstem region. The results show that the guinea pig superior olivary complex receives moderately dense noradrenergic innervation and no dopaminergic innervation. In addition, noradrenergic fibres and varicosities were observed in close contact with both somata and dendrites of olivocochlear neurons, strongly suggestive of synaptic contacts. Finally the results show that a significant component of the noradrenergic innervation of the guinea pig superior olivary complex arises in the locus subcoeruleus, which is a structure likely to be the homologue of the locus coeruleus in rats and other species.

Noradrenergic modulation of brainstem nuclei alters cochlear neural output.

The peripheral auditory sense organ, the cochlea, receives innervation from lateral and medial olivocochlear neurons in the brainstem. These neurons are able to modulate cochlear neural output. Anatomical studies have shown that one of the neurotransmitters which is present in varicosities surrounding the olivocochlear neurons in the brainstem is noradrenaline and previous work on brainstem slices has demonstrated a generally excitatory effect of noradrenaline on medial olivocochlear neurons. In order to assess in vivo the function of the noradrenergic inputs to olivocochlear neurons, we injected noradrenaline in the brainstem of anaesthetised guinea pigs and recorded ipsilateral cochlear electrical activity. Injections of noradrenaline close to the lateral olivocochlear neurons evoked increases in the sound-driven neural activity from the cochlea, measured as compound action potential (CAP) amplitude, as well as in the spontaneous activity, measured as amplitude of the 900 Hz peak of the spectrum of the neural noise in the cochlear fluids. In contrast, noradrenaline in the vicinity of the medial olivocochlear neurons evoked inhibitory effects on both the CAP amplitude and 900 Hz peak. These results indicate most likely an excitatory action of noradrenaline on both the lateral and medial olivocochlear neurons in the brainstem, and show that such noradrenergic inputs can modulate cochlear function.

Diverse responses of single auditory afferent fibres to electrical stimulation of the inferior colliculus in guinea-pig.

Medial olivocochlear (MOC) neurons in the auditory brainstem project to the cochlea and inhibit cochlear neural output by their action on the cochlear outer hair cells. The function of the lateral olivocochlear (LOC) neurons, projecting to the auditory primary afferents is still under debate. Recent studies have suggested that the olivocochlear system can have frequency-specific, spatially restricted effects within the cochlea. It has been shown that the inferior colliculus (IC) projects to the MOC neurons in a tonotopic manner and that electrical stimulation of the IC can activate the MOC system, suppressing cochlear gross potentials. In addition, it has been shown that stimulation of the IC may be able to activate the LOC neurons. We investigated the effect of IC stimulation on single units in the cochlea of guinea-pigs and searched for evidence of spatially restricted effects of the MOC system and effects of the LOC system. We found a variety of effects on single units. About 40% of units were unchanged whereas others (53%) showed inhibitory effects, reflected in a rightward shift of their rate-level function, sometimes accompanied by a suppression of the spontaneous rate. About 18% of the inhibited neurons showed an increased spontaneous rate. In 5% of the units we observed an excitatory effect of IC stimulation, resulting in a leftward shift of the rate-level functions. We also found that the effect could vary greatly between units of the same and adjacent frequencies within a single animal. These results imply an involvement of another regulatory system besides the MOC system, possibly the LOC system, which acts directly on the primary afferents. These data also demonstrate that the olivocochlear system is capable of eliciting highly localized effects on different frequency regions in the cochlea.

Dopaminergic olivocochlear neurons originate in the high frequency region of the lateral superior olive of guinea pigs.

Dopaminergic neurons are known to exist within the lateral superior olive (LSO). The LSO is the nucleus of origin of the lateral olivocochlear neurons, which project to the cochlea and synapse onto the primary afferents contacting the inner hair cells. We investigated whether the dopaminergic neurons in the LSO are part of the lateral olivocochlear neuron population. We combined intracochlear injections of a fluorescent retrograde tracer with immunofluorescent staining of tyrosine hydroxylase (TH). TH was used as a marker for dopaminergic neurons. After the injection with retrograde tracer most of the TH-labelled neurons in the LSO also contained the tracer, which directly demonstrates for the first time that the TH-labelled, dopaminergic neurons in the LSO are lateral olivocochlear neurons. TH-labelled neurons were not equally distributed over the LSO as is observed for the lateral olivocochlear neurons in general. TH-labelled neurons were almost exclusively seen in the medial, high frequency, limb of the LSO. Since the projection of the lateral olivocochlear neurons to the cochlea is known to be tonotopic, we investigated the TH-labelling in the cochlea as well. We found that the staining pattern of TH in the cochlea is in broad agreement with the distribution of TH-labelling in the LSO. Cochlear sections showed dense labelling in the basal and second, high frequency, turns and decreasing intensity of staining in the third turn, while the extreme apical, low frequency, turn was almost devoid of any positive TH-labelling. These observations imply that the dopaminergic neurons of the lateral olivocochlear system may play a role in the selective suppression of the high frequency fibers of the auditory system.