Coding of envelope modulation in the auditory nerve and anteroventral cochlear nucleus.

We have investigated responses of the auditory nerve fibres (ANFS) and anteroventral cochlear nucleus (AVCN) units to narrowband 'single-formant' stimuli (SFSS). We found that low and medium spontaneous rate (SR) ANFS maintain greater amplitude modulation (AM) in their responses at high sound levels than do high SR units when sound level is considered in dB SPL. However, this partitioning of high and low SR units disappears if sound level is considered in dB relative to unit threshold. Stimuli with carrier frequencies away from unit best frequency (BF) were found to generate higher AM in responses at high sound levels than that observed even in most low and medium SR units for stimuli with carrier frequencies near BF. AVCN units were shown to have increased modulation depth in their responses when compared with high SR ANFS with similar BFS and to have increased or comparable modulation depth when compared with low SR ANFS. At sound levels where AM almost completely disappears in high SR ANFS, most AVCN units we studied still show significant AM in their responses. Using a dendritic model, we investigated possible mechanisms of enhanced AM in AVCN units, including the convergence of inputs from different SR groups of ANFS and a postsynaptic threshold mechanism in the soma.     

Frequency sensitivity of auditory neurons in the Caiman cochlear nucleus

Summary The characteristic frequencies of single auditory neurons in Caiman crocodilus (South American Alligator) range from 70 to 2,900 Hz. These neurons in the cochlear nuclei show a striking tonotopic organization which parallels that in birds. The sensitivity curve of all neurons and the number of neurons in each frequency range show features similar to those of birds and mammals.Supported by NSF. grant GB 5697. I thank Dr. Mark Konishi for overseeing this work.     

Auditory information coding by modeled cochlear nucleus neurons

In this paper we use information theory to quantify the information in the output spike trains of modeled cochlear nucleus globular bushy cells (GBCs). GBCs are part of the sound localization pathway. They are known for their precise temporal processing, and they code amplitude modulations with high fidelity. Here we investigated the information transmission for a natural sound, a recorded vowel. We conclude that the maximum information transmission rate for a single neuron was close to 1,050 bits/s, which corresponds to a value of approximately 5.8 bits per spike. For quasi-periodic signals like voiced speech, the transmitted information saturated as word duration increased. In general, approximately 80% of the available information from the spike trains was transmitted within about 20 ms. Transmitted information for speech signals concentrated around formant frequency regions. The efficiency of neural coding was above 60% up to the highest temporal resolution we investigated (20 μs). The increase in transmitted information to that precision indicates that these neurons are able to code information with extremely high fidelity, which is required for sound localization. On the other hand, only 20% of the information was captured when the temporal resolution was reduced to 4 ms. As the temporal resolution of most speech recognition systems is limited to less than 10 ms, this massive information loss might be one of the reasons which are responsible for the lack of noise robustness of these systems.     

Oscillating neurons in the cochlear nucleus: II. Simulation results

A computer model of sustained chopper neurons in the ventral cochlear nucleus is presented and investigated. In the companion paper, the underlying neurophysiological and neuroanatomical data are demonstrated. To explain the preference of chopper neurons for oscillations with periods which are multiples of a 0.4 ms synaptic delay, we suggest a model of circularly connected chopper neurons. In order to simulate chopper neurons within a physiological dynamic range for periodicity encoding, it is necessary to assume that they receive an input from onset neurons. Our computer analysis of the resulting simple neuronal network shows that it can produce stable oscillations. The chopping can be triggered by an amplitude-modulated signal (AM). The dynamic range and the synchronous response of the simulated chopper neurons to AM are enhanced significantly by an additional input from onset neurons. Physiological properties of chopper neurons in the cat, such as mean, standard deviation, and coefficient of variation of the interspike interval are matched precisely by our simulations.     

Thresholds of cat cochlear nucleus neurons to microwave pulses

Action potentials of neurons in cat dorsal and posteroventral cochlear nuclei were recorded extracellularly with glass microelectrodes while the head of the cat was exposed to microwave pulses at 915 MHz using a diathermy applicator. Response thresholds to acoustic tones, acoustic clicks, and microwave pulses were determined for auditory units with characteristic frequencies (CFs) from 278 Hz to 39.2 kHz. Tests with pulsatile stimuli were performed for durations of 20-700 mus, principally 20, 70, and 200 mus. Brainstem midline specific absorption rate (SAR) threshold was as small as 11.1 mW/g per pulse, and specific absorption (SA) threshold was a small as 0.6 muJ/g per pulse. Microwave thresholds were generally lower for CF less than 9 kHz, as were most acoustic thresholds. However, microwave threshold was only weakly related to click threshold and CF-tone threshold of each unit.     

A simulation of chopper neurons in the cochlear nucleus with wideband input from onset neurons

The unique temporal and spectral properties of chopper neurons in the cochlear nucleus cannot be fully explained by current popular models. A new model of sustained chopper neurons was therefore suggested based on the assumption that chopper neurons receive input both from onset neurons and the auditory nerve (Bahmer and Langner in Biol Cybern 95:4, 2006). As a result of the interaction of broadband input from onset neurons and narrowband input from the auditory nerve, the chopper neurons in our model are characterized by a remarkable combination of sharp frequency tuning to pure tones and faithful periodicity coding. Our simulations show that the width of the spectral integration of the onset neuron is crucial for both the precision of periodicity coding and their resolution of single components of sinusoidally amplitude-modulated sine waves. One may hypothesize, therefore, that it would be an advantage if the hearing system were able to adapt the spectral integration of onset neurons to varying stimulus conditions.     

Morphology of neurons cultured from subdivisions of the mouse cochlear nucleus

This study was designed to characterize the dendritic organization of cochlear nucleus (CN) cells grown in primary cell culture and to assess differences among cultures grown from different regions of CN. Cultures were prepared from postnatal mice and processed using microtubule-associated protein 2 (MAP2) or gamma-aminobutyric acid (GABA) immunohistochemistry. CN neurons were successfully cultured from preparations grown from either the anteroventral subdivision of the nucleus (AVCN), the posterior region [posteroventral (PVCN) and dorsal (DCN) subnuclei], or the whole CN, although the cultured neurons did not exhibit complex dendritic patterns characteristic of CN neurons in vivo. Neurons cultured from the entire nucleus exhibited an increased rate of survival compared to those cultured from either the anterior or posterior regions, although similar types of cells were observed in all preparations. The majority of cultured CN neurons were GABA-positive and had soma areas that were similar to the areas of immature GABAergic neurons measured in CN sections. Small cells (soma areas or=120 microm(2)) were also present in significant numbers. Overall, CN cultures consisted of a heterogeneous population of neurons that had less elaborate dendritic organizations than cells of corresponding size that have been described in adult animals in vivo.     

Neural modeling of intrinsic and spike-discharge properties of cochlear nucleus neurons

The purpose of this study was to develop neurobiologically plausible models to account for the response properties of several types of cochlear nucleus neurons. Three cell types--the bushy cells, stellate cells, and fusiform cells--were selected because useful data from intracellular recordings were available for these cell types, and because these three cell types exhibit distinct contrasts in their neuronal signal coding strategies. Stellate cells have primarily linear current-voltage (I-V) characteristics, but both bushy and fusiform cells have highly non-linear I-V characteristics. In light of this, we hypothesize that some of these cells have non-linear voltage-dependent conductances which alter their response properties. We modeled the bushy cell membrane conductance as an exponentially increasing function of membrane voltage, that of the fusiform cell as an exponentially decreasing function of the voltage, and that of the stellate cell as being voltage-independent. We have combined the voltage-dependent non-linear conductances of the cell membrane with a simple R-C circuit type of neuron model. These models reproduced the salient features of the experimentally observed I-V characteristics of the cells. In addition, we found that the models reproduced the spike discharge behavior to intracellularly injected current steps. Moreover, a more detailed study of stellate cell 'chopper'-type response patterns yielded hypotheses regarding the nature of the current that must exist at the soma during a pure-tone stimulus in order for the cells to exhibit various chopper subtype patterns, such as chop-S, chop-T, and Oc. The chop-S pattern requires a steady average current level with a relatively small variability during the tone-burst stimulus. The chop-T pattern, in contrast, requires that the current become more irregular during the tone-burst stimulus. The Oc pattern arises, however, when the input is similar to the chop-T case but the intrinsic properties of the cell model have been changed to increase the accommodation of the threshold. The implications of these findings for circuitry in the cochlear nucleus are discussed. Our analysis of these models revealed that this approach can be used to simulate neuronal cell types where I-V characteristics are known but more detailed ion channel data are not known.     

Computer simulation of shared input among projection neurons in the dorsal cochlear nucleus

Computer simulations of a network model of an isofrequency patch of the dorsal cochlear nucleus (DCN) were run to explore possible mechanisms for the level-dependent features observed in the cross-correlograms of pairs of type IV units in the cat and nominal type IV units in the gerbil DCN. The computer model is based on the conceptual model (of a cat) that suggests two sources of shared input to DCN's projection neurons (type IV units): excitatory input from auditory nerves and inhibitory input from interneurons (type II units). Use of tonal stimuli is thought to cause competition between these sources resulting in the decorrelation of type IV unit activities at low levels. In the model, P-cells (projection neurons), representing type IV units, receive inhibitory input from I-cells (interneurons), representing type II units. Both sets of model neurons receive a simulated excitatory auditory nerve (AN) input from same-CF AN fibers, where the AN input is modeled as a dead-time modified Poisson process whose intensity is given by a computationally tractable discharge rate versus sound pressure level function. Subthreshold behavior of each model neuron is governed by a set of normalized state equations. The computer model has previously been shown to reproduce the major response properties of both type IV and type II units (e.g., rate-level curves and peri-stimulus time histograms) and the level-dependence of the functional type II-type IV inhibitory interaction. This model is adapted for the gerbil by simulating a reduced population of I-cells. Simulations were carried out for several auditory nerve input levels, and cross-correlograms were computed from the activities of pairs of P-cells for a complete (cat model) and reduced (gerbil model) population of I-cells. The resultant correlograms show central mounds (CMs), indicative of either shared excitatory or inhibitory input, for both spontaneous and tone-evoked driven activities. Similar to experimental results, CM amplitudes are a non-monotonic function of level and CM widths decrease as a function of level. These results are consistent with the hypothesis that shared excitatory input correlates the spontaneous activities of type IV units and shared inhibitory input correlates their driven activities. The results also suggest that the decorrelation of the activities of type IV units can result from a reduced effectiveness of the AN input as a function of increasing level. Thus, competition between the excitatory and inhibitory inputs is not required.     

Identification of tuberculo-ventral neurons in the polymorphic layer of the rat dorsal cochlear nucleus

The tuberculo-ventral tract represents a short nervous circuit within the auditory cochlear nuclei. Tuberculo-ventral neurons of the dorsal cochlear nucleus send isofrequency inhibitory inputs to bushy cells of the ventral cochlear nucleus. Injection of wheat germ agglutinin conjugated to horseradish peroxidase into the rat ventral cochlear nucleus, labelled tuberculo-ventral neurons retrogradely in the deep polymorphic layer of the ipsilateral dorsal cochlear nucleus. Five to 20% of the perimeter of these cells was covered by synaptic boutons, most of which contained flat and pleomorphic vesicles. These boutons contained glycine and sometimes GABA. Occasional small axo-somatic boutons contained round vesicles and were immunonegative for both glycine and GABA. This study shows that the synaptic profile of tuberculo-ventral neurons is different from that of other medium-size glycinergic neurons within the polymorphic layer or more superficial regions of the dorsal cochlear nucleus like cartwheel neurons. In fact the latter mostly receive boutons that contain pleomorphic vesicles.     

Risk functions and expected spike density functions for neurons in the cat cochlear nucleus

The temporal properties of spontaneous and (or) evoked discharges of 157 neurons localized in dorsal cochlear nucleus of anaesthetized cats have been studied. Tone bursts were presented at stimulus best frequency in a free field from the side of ipsilateral ear. About half of cells were characterized by paused or build-up types of the discharge. For all such units a long lasting post-spike decrease in excitability could be seen from the analysis of hazard functions of spontaneous and evoked activity. As a result, the time dependence of conditional probability of the first crossing of the threshold (under condition of an absence of previous response spikes) or expecting probability function (EPF) were over the usual peristimulus histograms. Units with chopper discharges usually did not demonstrate alternative peaks in EPF. We interpreted this fact as evidence that chopper discharge pattern is a result of strong post spike decrease in excitability. Such pattern doesn't demonstrate an existence of real periodicity of the unit. In primary-like units the hazard functions demonstrated only minor after-spike decrease of excitability, and the EPFs were similar to the initial part of peristimulus histograms. Type II units (presumably inhibitory cells) were characterized by non-monotonous hazard functions and by a tendency to burst response patterns. In some cells, we observed a tendency to existence of real intrinsic oscillations both in the EPFs and hazard functions.     

Cytology of large neurons in the guinea pig dorsal cochlear nucleus contacting the inferior colliculus

Large neurons in the dorsal cochlear nucleus of the guinea pig which project to the inferior colliculus were identified after injections of the neural tracer WGA-HRP. Retrograde labelled cells (pyramidal and giant neurons) in the dorsal cochlear nucleus were glycine and GABA immunonegative and showed a similar ultrastructure. Between 30 and 60% of their perimeter was covered by axo-somatic boutons, most of which (>50%) contained pleomorphic synaptic vesicles. Other boutons (about 40% of total) contained flat vesicles and few (5-6%) contained round vesicles, a characteristic of the excitatory cells innervating the inferior colliculus. Immunogold-cytochemistry, coupled to silver intensification, showed that more than 50% of axo-somatic pleomorphic boutons and over 90% of boutons containing flat and pleomorphic vesicles store glycine. Rare WGA-HRP labelled axo-somatic boutons containing flat-pleomorphic vesicles were seen on pyramidal and giant neurons. This suggests that a few inhibitory collicular terminals contact the excitatory large neurons in the dorsal cochlear nucleus.     

The kinetics of the response to glutamate and kainate in neurons of the avian cochlear nucleus.

Neurons in the nucleus magnocellularis (nMAG) of the chicken precisely transmit auditory nerve activity via glutamatergic synapses. Using techniques for rapid application of solutions, we have explored the properties of CNQX-sensitive glutamate receptors in whole cells and outside-out patches from the nMAG. Glutamate-evoked current in patches desensitized biphasically to less than 1% of the peak current, with a fast time constant of 960 microseconds at 22 degrees C, decreasing to 570 microseconds at 33 degrees C. Dose-response studies using kainate indicated that at least two agonist molecules bind to gate the channel. We propose a kinetic model that quantitatively describes our experimental observations. The rapid kinetics of this receptor are well suited to allow phase locking of synaptic signals to auditory stimuli.     

Oscillating neurons in the cochlear nucleus: I. Experimental basis of a simulation paradigm

Anatomical and physiological auditory data and pitch measurements are presented including some additional analysis. The data provide the basis for a new computer model of sustained chopper neurons in the ventral cochlear nucleus. New and old evidence indicating a preference for multiples of 0.4 ms in oscillations of chopper neurons in the cochlear nucleus of different species such as man, cats, and Guinea fowls, is summarized. Our hypothesis is that the time constant of 0.4 ms is due to the minimum synaptic delay of chopper neuron connections. Anatomical findings show that chopper neurons are indeed connected and can excite each other; a model of a circular network of neurons that are connected via synapses with a delay of 0.4 ms is thus plausible. Results concerning frequency tuning and dynamical properties of periodicity encoding of chopper neurons are reviewed. It is concluded that chopper neurons receive input both from auditory nerve fibres and onset neurons.     

The polypeptide PEP-19 is a marker for Purkinje neurons in cerebellar cortex and cartwheel neurons in the dorsal cochlear nucleus

This light and electron microscopic immunocytochemical study shows that the polypeptide PEP-19, a presumptive calcium binding protein specific to the nervous system, represents an excellent marker for cerebellar Purkinje cells and dorsal cochlear nucleus (DCoN) cartwheel cells. The polypeptide clearly reveals the entire populations of both types of neurons, including their complete dendritic and axonal arborizations. Other PEP-19 containing neurons in the two regions display weak immunoreactivity restricted to the cell body or to cell body and principal dendrites. Electron microscopic localization of PEP-19-like immunoreactivity reveals similarities between this polypeptide, parvalbumin, and a 28K vitamin D-dependent calcium binding protein. However, calmodulin, which is expressed in both Purkinje and granule cells, may differ from PEP-19. Similarities between the organization of the cerebellar cortex and the DCoN superficial layers have been known for some time, with several types of neurons in one system having their presumed homologue in the other. These data provide further support for the proposed structural and functional homology between Purkinje and cartwheel neurons, and establishes PEP-19 as a useful marker for examining degeneration of these two neuronal populations in murine cerebellar mutants.     

Glutamatergic and GABAergic agonists increase [Ca2+]i in avian cochlear nucleus neurons

Neurons of the avian cochlear nucleus, nucleus magnocellularis (NM), are stimulated by glutamate, released from the auditory nerve, and GABA, released from both interneurons surrounding NM and from cells located in the superior olivary nucleus. In this study, the Ca²⁺ indicator dye Fura-2 was used to measure Ca²⁺ responses in NM stimulated by glutamate- and GABA-receptor agonists using a chicken brainstem slice preparation. Glutamatergically stimulated Ca²⁺ responses were evoked by kainic acid (KA), α-amino-3-hydroxyl-5-methylisoxazole-4-propionic acid (AMPA), and N-methyl-D -aspartate (NMDA). KA- and AMPA-stimulated changes in [Ca²⁺]_i were also produced in NM neurons stimulated in the presence of nifedipine, an L-type Ca²⁺ channel blocker, suggesting that KA- and AMPA-stimulated changes in [Ca²⁺]_i were carried by Ca²⁺-permeable receptor channels. Significantly smaller changes in [Ca²⁺]_i were produced by NMDA. When neurons were stimulated in an alkaline (pH 7.8) superfusate, NMDA responses were potentiated. KA- and AMPA-stimulated responses were not affected by pH. Several agents known to stimulate metabotropic receptors in other systems were tested on NM neurons bathed in a Ca²⁺ free-EGTA–buffered media, including l -cysteine sulfinic acid (L-CSA), trans-azetidine dicarboxylic acid (t-ADA), trans-aminocyclopentanedicarboxylic acid (t-ACPD), and homobromoibotenic acid (HBI). The only agent to reliably and dose-dependently increase [Ca²⁺]_i was HBI, an analog of ibotenate. GABA also stimulated increases in [Ca²⁺]_i in NM neurons. GABA-stimulated responses were reduced by agents that block voltage-operated channels and by agents that inhibit Ca²⁺ release from intracellular stores. Whereas GABA-A receptor agonist produced increases in [Ca²⁺]_i GABA-B and GABA-C receptor agonists had no effect. There appear to be several ways for [Ca²⁺]_i to increase in NM neurons. Presumably, each route represents a means by which Ca²⁺ can alter cellular processes. © 1998 John Wiley & Sons, Inc. J Neurobiol 37: 321–337, 1998     

Concentration-jump analysis of voltage-dependent conductances activated by glutamate and kainate in neurons of the avian cochlear nucleus.

We have examined the mechanisms underlying the voltage sensitivity of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate receptors in voltage-clamped outside-out patches and whole cells taken from the nucleus magnocellularis of the chick. Responses to either glutamate or kainate had outwardly rectifying current-voltage relations. The rate and extent of desensitization during prolonged exposure to agonist, and the rate of deactivation after brief exposure to agonist, decreased at positive potentials, suggesting that a kinetic transition was sensitive to membrane potential. Voltage dependence of the peak conductance and of the deactivation kinetics persisted when desensitization was reduced with aniracetam or blocked with cyclothiazide. Furthermore, the rate of recovery from desensitization to glutamate was not voltage dependent. Upon reduction of extracellular divalent cation concentration, kainate-evoked currents increased but preserved rectifying current-voltage relations. Rectification was strongest at lower kainate concentrations. Surprisingly, nonstationary variance analysis of desensitizing responses to glutamate or of the current deactivation after kainate removal revealed an increase in the mean single-channel conductance with more positive membrane potentials. These data indicate that the rectification of the peak response to a high agonist concentration reflects an increase in channel conductance, whereas rectification of steady-state current is dominated by voltage-sensitive channel kinetics.