| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Oregon Hearing Research Center and Vollum Institute, Portland, Oregon
Submitted 28 August 2007; accepted in final form 17 October 2007
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
). In addition to this computation based on information from primary afferent fibers, the DCN also receives input from diverse brain regions through a system of auditory granule cells and their parallel fiber axons. Granule cells are the smallest neurons in the cochlear nucleus and are found in the granule cell region of the cochlear nuclear complex and distributed at lower density through the outer layers of the DCN itself (Mugnaini et al. 1980b).
Thus the granule cell system provides an essential element in the convergence of somatosensory and auditory signals at an early stage of sensory processing. Such integration may play a role in the suppression of predicted body-generated sounds (Shore 2005; Zhou and Shore 2006) or orientation of the head or ears to a sound source (Kanold and Young 2001; May 2000; Sutherland et al. 1998; Young and Davis 2002). Given the potential importance of this nonauditory pathway, it is surprising that very little is known about the physiology of synaptic inputs to granule cells.
Each granule cell has one to four short dendrites that give rise to small terminal expansions resembling the claw-like endings of cerebellar granule cells (Mugnaini et al. 1980a,b; Wright and Ryugo 1996). Their dendrites may receive input from neurons of higher auditory centers, such as the inferior colliculus (Caicedo and Herbert 1993), auditory cortex (Schofield and Coomes 2005; Weedman and Ryugo 1996), and olivocochlear neurons (Brown et al. 1988). Granule cells also receive nonauditory projections from trigeminal ganglion, dorsal column nuclei, and spinal trigeminal nuclei (Haenggeli et al. 2005; Itoh et al. 1987; Shore et al. 2000; Weinberg and Rustioni 1987; Wright and Ryugo 1996; Zhou and Shore 2004) as well as vestibular ganglion and nucleus (Bukowska 2002; Burian and Gstoettner 1988; Kevetter and Perachio 1989). Ultrastructural studies reveal both bouton-like and classical mossy terminals on granule cell dendrites (Alibardi 2004; Mugnaini et al. 1980a; Wright and Ryugo 1996), and it is likely that these extrinsic inputs are mainly glutamatergic (Ottersen et al. 1990; Wright and Ryugo 1996; Zhou et al. 2007). However, axons of presumably cholinergic efferents also terminate among some granule cells (Benson and Brown 1990). Finally, putative inhibitory inputs have been observed, and these may be GABA- and/or glycinergic (Alibardi 2003a,b, 2004).
We have been interested in how such diverse inputs to granule cells are integrated physiologically and how they are modified by local circuits. Here the electrophysiological features of synaptic transmission in granule cells of the DCN were explored using in vitro patch-clamp recordings. We found that excitation was mediated by AMPA and N-methyl-D-aspartate (NMDA) receptors and that a local, polysynaptic circuit may contribute to this excitation. Powerful synaptic inhibition was mediated by both GABA and glycine and likely serves to gate the entry of nonauditory signals into the DCN.
|
|
METHODS |
|---|
|
Slice preparation and electrophysiology
Coronal brain slices (200 µm thick) containing DCN were made from Wistar rats (P12–P22). Slices were prepared in a warm (37°C) extracellular solution composed of (in mM) 129 NaCl, 15 glucose, 2.5 KCl, 1.4 MgSO4, 2 CaCl2, 1.2 NaH2PO4, 20 NaHCO3, and 3 Na-HEPES, bubbled with 5% CO2-95% O2 (pH 7.4), modified from Wickesberg and Oertel (1988). All recordings were done at 34–36°C with the same extracellular solution used for preparing the slices. In experiments with reduced Ca2+, Ca2+ was replaced by equimolar amount of Mg2+. For current-clamp recordings, a K-gluconate-based internal solution was used composed of (in mM) 113 K-gluconate, 4.5 MgCl2, 9 HEPES, 0.1 EGTA, 4 ATP, 0.3 GTP, and 14 Tris-phosphate and pH adjusted to 7.3 with KOH. Recording pipettes were pulled from borosilicate glass capillaries and had 9- to 11-M
tip resistance when filled with the K-gluconate internal solution. Pipettes were wrapped with Parafilm to minimize capacitance. Voltages were corrected off-line for a 12-mV junction potential. In voltage-clamp recordings, a CsCl-based internal solution was used containing (in mM) 140 CsCl, 4.6 MgCl2, 10 HEPES, 4 ATP, 0.4 GTP, and 0.2 EGTA, and pH adjusted to 7.3 with CsOH. The command voltages were not corrected for a 3-mV junction potential, except as noted.
Recordings were made with a Multiclamp 700B amplifier (Molecular Devices, Sunnyvale, CA). Data were filtered at 2.2 kHz, digitized at 5 kHz, and stored on a laboratory computer using a Digidata 1200 interface (Axon Instruments) and pClamp 9.2 software (Axon Instruments). Synaptic responses were evoked by 5- to 20-V, 100-µs voltage pulses delivered through a bipolar glass electrode that was placed 30–40 µm away from the cell body to activate presynaptic axons. The maximum acceptable series resistance was 
35 M. Recordings with and without compensation on-line by 70–80% (lag: 10 µs) displayed little difference in current amplitudes. Drug concentrations were 20 µM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), 10 µM 6,7-dinitroquinoxaline-2,3-dione (DNQX), 10 µM 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo[f]quinoxaline-2,3-dione (NBQX), 50 µM D-2-amino-5-phosphonopentanoic acid (D-AP5), 0.5 µM strychnine, 1 mM QX-314, 10 µM bicuculline, and 10 µM SR-95531. Tonic current was measured in the presence of 20 µM DNQX, 50 µM D-AP5, 0.5 µM strychnine, and 2 µM GABA, whereas 100 µM bicuculline was puffed on the granule cells held at –70 mV.
The majority of our recordings (n = 69 cells) were made from DCN granule cells occupying the molecular, fusiform and deep layers, which was described as the seventh subregion by Mugnaini et al. (1980). We also compared these with the glycinergic responses of 7 granule cells from the granule cell regions in the medial border of the DCN and the ventral cochlear nucleus (VCN), corresponding to the third subregion of Mugnaini et al. (1980). Granule cells were viewed using a Zeiss Axioskop FS microscope equipped with differential interference contrast optics and a x40 water-immersion objective. Because there are other small cell types in the DCN such as unipolar brush cells and chestnut cells (Weedman et al. 1996), we loaded each recorded neuron with Alexa Fluor-488 (Molecular Probes, Eugene, OR) to identify the cell type under fluorescence illumination. Cells were imaged after recording using a Till Photonics Polychrome IV monochrometer and Sensicam camera. In a few cases (Fig. 1) cells were recorded and then imaged using a Prairie Technologies 2 photon microscope. Drugs were dissolved in extracellular solution. In some cases, drugs were first dissolved in DMSO such that the final concentration was <0.01%.
|
The data were analyzed with Clampfit 9.2 software (Axon Instruments). The decays of evoked postsynaptic currents were fit by single- or double-exponential function (based on the improvement of the summed square error): D(t) = Afast*e–t/
fast + Aslow*e–t/slow, where D(t) is the decay of the evoked current as a function of time (t); Afast and Aslow are amplitude constants, and fast and slow are fast and slow decay time constants, respectively. In some cases, adding the second exponent did not significantly decrease the summed square error, and so Aslow was set to 0. The weighted decay time constant was calculated as wd = (Afast x fast + Aslow x slow)/(Afast + Aslow). Percentage of fast component was calculated as % Fast = [Afast/(Afast + Aslow)] x 100. All results are expressed as means ± SD (except as noted).
Fluctuation analysis
Peak-scaled nonstationary fluctuation (Momiyama et al. 2003; Traynelis et al. 1993) analysis was used to estimate the weighted mean of single-channel currents of receptors underlying glycinergic synaptic currents in granule cells. Quantal fluctuation generated large disparities in IPSC amplitude from trial to trial. Because larger-amplitude events tended to be somewhat slower, events were selected so that the analyzed dataset did not vary in amplitude more than twofold. No time-dependent changes in amplitude were apparent (Momiyama et al. 2003; Traynelis et al. 1993). Traces containing obvious delayed releases of quanta or spurious noise were also eliminated. An analysis routine was implemented in Axograph X 1.0 (AxographX), which scaled an averaged IPSC to each event, subtracted these, and determined the variance along the decay phase of the currents. Mean variance and amplitude were obtained for data binned by dividing the currents into 100 segments by amplitude; these were then plotted as a variance-mean graph and a line was fit to the first 25% of the curve to obtain single-channel current. The accuracy of the analysis routine was checked by simulation. Synaptic currents were simulated using a routine in WCP 3.3.9 (Stratheclyde Electrophysiology Software) which employs a three-state model (unbound-closed-open). Binding and unbinding rates (in 1/s) were 200,000 and 10,000, respectively, whereas opening and closing rates were 50,000 and 500, respectively. One hundred channels were modeled with a coefficient of variation of 20%. Baseline noise was set to 5 pA. With a decay time of 100 µs for transmitter and channel amplitudes of 3–4 pA, this model yielded currents that resembled those we recorded in size and decay time. We then generated datasets with channel amplitudes of either 3 or 4 pA and filtered the data at 10, 1, or 0.3 kHz. For any setting, the measured single-channel current was within 3% of the simulated value.
|
|
RESULTS |
|---|
|
Granule cells were identified based on their size and location in the DCN, as described by Mugnaini et al. (1980b). By imaging dye-loaded neurons, we could safely exclude recordings from other, slightly larger neurons such as unipolar brush cells and chestnut cells in the DCN (Ryugo et al. 2003). In several cases, we also imaged cells using two-photon microscopy; examples are shown in Fig. 1. In these images, and also from our routine imaging after each recording, granule cell somata ranged from 6 to 8 µm in diameter and had one to three short dendrites as reported previously (Mugnaini et al. 1980a). Not every dendrite included a prominent claw-like ending, and in some cells, no claws were apparent, consistent with innervation by nonmossy (bouton) input (Wright and Ryugo 1996; Zhou and Shore 2004).
Intrinsic properties of granule cells
The membrane capacitance of granule cells, assessed with brief, 5-mV voltage steps, was 3.3 ± 1.0 pF (n = 31). At 1 µF/cm2, this value corresponds to a diameter of 10 µm, slightly larger than our direct measurements; the difference in measured and estimated sizes could be explained by the added capacitance of the dendrites. The average input resistance, measured in a potential range near –60 mV, was 2.3 ± 0.8 G (n = 22). Accordingly only small currents were needed to drive spike activity in these cells. Firing properties were studied by long- and short-duration current injections and the results are similar to those of Rusznak et al. (1997). Typical responses to 400-ms current injections of –30 to +35 pA are shown in Fig. 2. Hyperpolarizing pulses did not generate a voltage "sag," suggesting little expression of IH in these cells. Repetitive action potentials could be elicited with injections of as little as +15 pA (Fig. 2, A and B). At +30 pA, the mean firing rate was 115 ± 41 Hz (n = 5). Five consecutive brief current pulses (duration, 1 ms; interval, 3 ms), ranging from 5 to 205 pA, were injected to assess how well the cells could respond to synaptic-like stimuli. Current stimuli between 5 and 25 pA could not elicit action potentials (n = 5; Fig. 2C). With 45-pA short-duration current injections, summating potentials reached spike threshold, and only at 205 pA was each stimulus able to elicit a spike. The average amplitude of brief current pulse needed to trigger a spike was 177 ± 30 pA (n = 5).
|
Glutamatergic synaptic currents were isolated using a cocktail of SR-95531 (10 µM) and strychnine (0.5 µM) to block inhibitory events. At a holding potential of –70 mV, maximal evoked EPSCs had a peak amplitude of –61.5 ± 31.3 pA (n = 17) and were completely blocked by CNQX (20 µM), indicating that they were mediated by AMPA/kainate receptors (Fig. 3, A and B, Table 1). The EPSC decay at –70 mV was fitted by a double-exponential function with time constants of 1.0 ± 0.4 ms (n = 16) and 14.2 ± 7.9 ms (n = 15; Table 1). As shown in Fig. 3, C and D, EPSCs evoked at positive holding potentials included a prominent, slow NMDA component (Table 1), blocked by D-AP5 (50 µM; data not shown). Consistent with the presence of both AMPA and NMDA components of the EPSC, I-V relations for peak current were roughly linear in control solutions but exhibited outward rectification in the presence of NBQX (Fig. 3C). Table 1 uses the reversal potential determined from these IVs (+1.38 mV) to estimate the synaptic conductance. During high-frequency stimulation at –70 mV, a summated current appeared as a slow decay after the termination of stimulation (Fig. 3E). This slow current was quantified as the current level at the base of the last EPSC. For 100-Hz stimuli, the current was –40 ± 35 pA (n = 7) and had a decay time constant of 77 ± 18 ms, similar to that of NMDA EPSCs. However, because the slow decay was observed at potentials at which NMDA receptors should be largely blocked by Mg2+, it is possible that it represents activation of AMPA receptors by residual glutamate, as observed for cerebellar granule cells (Overstreet et al. 1999).
|
|
To determine whether a single EPSC could elicit an action potential, we compared the amplitude of EPSCs and excitatory postsynaptic potentials (EPSPs) while varying the stimulus voltage. With weak stimuli, short-latency, monosynaptic inputs were apparent; these showed one or more incremental increases with stimulus strength (Fig. 4A). Further increase in the stimulus voltage often triggered additional, longer-latency responses. Such delayed firing of inputs, observed in 12 of 18 cells, indicates the presence of polysynaptic excitation of granule cells. To compare the evoked EPSC to EPSP, we used the same stimulation condition to elicit EPSPs in current clamp. In four of five cells, the initial EPSPs, generated as stimulus strength was gradually increased, were subthreshold. In the remaining cell, weak stimulation produced spikes in only 3% of the responses (Fig. 4B, 15-V response). With a brief train (100 Hz) of weak stimuli, the EPSPs summated and produced action potentials in all seven cells that were analyzed (Fig. 4B). With high stimulation, each stimulus elicited one or more action potentials as shown in Fig. 4B. Recruitment of polysynaptic EPSPs was noted in three of eight cells of which 33% elicited spikes.
|
A cocktail of DNQX (20 µM) and D-AP5 (50 µM) was used to block excitatory events and thereby isolate inhibitory synaptic responses. As shown in Fig. 5A, synaptically evoked IPSCs at a holding potential of –70 mV were mainly blocked by strychnine (0.5 µM) with the remaining current blocked by bicuculline (10 µM). Synaptic currents recorded in the presence of glutamate receptor blockers plus bicuculline were considered glycinergic. The average glycinergic IPSC amplitude was –528 ± 304 pA (n = 20), and its decay time was fitted by double-exponential function with average time constants of 16.7 ± 15.1 and 73.3 ± 67.9 ms (n = 20). Table 1 shows these average values and compares synaptic conductances for the different transmitters. As shown in Fig. 5B, glycinergic IPSCs were able to follow high-frequency stimulation. At 100 Hz, the summated current was –238 ± 92 pA (n = 11). Similar glycinergic responses were observed in seven granule cells of the granule cell region in between the DCN and VCN (data not shown).
|
). The average GABAergic IPSC was –241 ± 128 pA (n = 8), and its decay time was fitted by double-exponential function with an average fast decay constant of 4.5 ± 1.5 ms and slow decay constant of 32 ± 14 ms (Table 1). Thus GABAergic inputs were both weaker and faster than glycinergic synapses. As with glycinergic currents, GABAergic currents were able to follow high-frequency stimulation, as shown in Fig. 6C. At 100 Hz, the summated current (current at foot of last IPSC) was –136 ± 59 pA, n = 4.
|
In cerebellar granule cells, ambient GABA produces a prominent tonic inhibition mediated by GABAA receptors containing 
6 and 
subunits (Farrant and Nusser 2005; Santhakumar et al. 2006). Because DCN granule cells contain 6 but lack subunits (Campos et al. 2001), we predicted that they would also lack tonic GABA current. To compare the background GABA sensitivity of DCN granule cells to cerebellar granule cells quantitatively, we made recordings from both cell types and examined the ability of local application of bicuculline to block tonic current activated by 2 µM GABA (Fig. 7A). All-points histograms were obtained to assess the effect of the blocker on the holding current (Fig. 7B) as opposed to the occasional IPSC (Santhakumar et al. 2006). Gaussian curves were fit to current samples before and after bicuculline application and the differences in their mean values compared in the two cell types (cerebellar granule cells, –21.5 ± 10.8 pA; DCN granule cells, –2.8 ± 1.7 pA; Fig. 7C). The results confirm the presence of a tonic GABA response in cerebellar granule cells and further indicate that DCN granule cells are almost completely insensitive to proposed levels of ambient GABA (Farrant and Nusser 2005; Santhakumar et al. 2006).
|
6- containing GABA receptors, it seems unlikely that there might be a tonic glycine current. Indeed local, puffer application of 50 µM strychnine blocked background current by 0.00 ± 0.04% (n = 6, P = 0.91). Numbers of glycine inputs and receptors
Because glycinergic inputs seemed particularly powerful in granule cells, we asked how many synapses and how many receptors would be needed to generate such relatively large currents. The strength of a single input was estimated by first recording IPSCs in normal bath Ca2+ (Fig. 8Ai) and then dropping Ca2+ to 0.75 mM (Fig. 8A). At this [Ca2+], IPSCs fluctuated widely in amplitude, often failing to occur. The apparent modal amplitude in this cell, excluding failures, was –40 pA; among five cells examined, the mean excluding failures was 86 ± 77 pA. This probably overestimates the strength of a single input, as some events must be due to more than one quantal release. We tried to plot these as amplitude histograms to determine a modal unitary value over the population, but the scatter was too great to resolve a peak. Moreover, miniature IPSCs in TTX (0.5 µM) were rarely observed in these cells and so could not be used to estimate unitary IPSC size.
|
). These were recorded in the presence of CNQX (20 µM), D-AP5 (50 µM), and SR-95531 (10 µM) at a holding potential of –70 mV. The current during the decay phase of the IPSCs were represented as histograms which showed peaks corresponding to the presence of channel openings (Fig. 8C). These peaks were fit with Gaussian curves and the difference between adjacent peaks taken as the amplitude of individual channel openings. (Fig. 8C). Single-channel conductance was calculated from such step currents in nine cells (Fig. 8D), assuming a reversal potential of –2.6 mV (see Fig. S11 ); these gave an average conductance of 48.2 ± 9.3 pS (n = 9; average current 3.39 ± 0.65 pA, holding potential corrected for junction potential), indicating the presence of heteromeric synaptic glycine receptors in these cells (Legendre 2001).
This estimate was based on the largest single-channel currents in the IPSC decay, which presumably have the biggest effect on the shape of histograms, as in Fig. 8C. However, glycine receptors exhibit several subconductance states (Bormann et al. 1987), and it is possible that these contribute significant current to the IPSC. We therefore used peak-scaled nonstationary fluctuation analysis to assess the weighted mean of the channel currents during IPSCs (see METHODS). Figure 9A shows an example of individual traces and their mean selected for this analysis from one cell. After scaling the mean to each trace and subtracting them, we obtained the average variance during the decay phase of the IPSCs and plotted this against the mean current, as shown in Fig. 9B. A line fitted to the initial slope for this curve yielded a mean single-channel current of 2.86 pA. Similar measurements from a total of six neurons gave a mean current of 3.27 ± 1.08 pA and a mean conductance of 46.4 ± 15.3 pS (Fig. 9C), not significantly different from our single-channel recordings described above. Thus subconductance states probably do not contribute significantly to the single-channel currents underlying glycinergic IPSCs of granule cells. Given this estimate of single-channel current, we can evaluate the number of channels at each synapse. With a minimal synaptic strength of 86 pA and a mean of 528 pA for the evoked response, a minimum of 6 synapses each with 25 receptors could account for the evoked IPSC. These values represent underestimates since the probabilities of transmitter release and of channel opening are most likely less than one.
|
Given that glycine may be excitatory in some neurons of the DCN (Golding and Oertel 1996), we monitored spikes extracellularly in granule cells using a cell-attached pipette and then applied exogenous glycine to determine its effects on firing, using application of glutamate as a control. On glutamate application (100 µM) for 10 ms, the activity of the granule cell increased by >75% in all three cells examined (Fig. 10, A and C). By contrast, on glycine application (100 µM) spikes were inhibited by >80% in all four cells examined (Fig. 10, B and C). Thus we infer from these results that IPSCs are strongly inhibitory and that the reversal potential for glycine is probably negative to the resting membrane potential.
|
|
|
DISCUSSION |
|---|
|
Anatomical and histochemical studies have demonstrated both glutamatergic mossy and small bouton endings on granule cells originating from diverse brain regions (Ryugo et al. 2003; Schofield and Coomes 2005; Wright and Ryugo 1996; Zhou and Shore 2004). With the use of pharmacological blockers, we were able to explore the physiological properties of these inputs. Glutamatergic transmission to granule cells utilized both AMPA and NMDA subtypes of glutamate receptors, consistent with antibody staining for glutamate receptors in granule cells (Petralia et al. 2000). Together these inputs generated excitatory signals, the magnitude and timing of which depended on the strength and frequency of stimulation. A single, weak stimulus produced EPSPs that were rarely suprathreshold. Stronger shocks elicited larger, suprathreshold EPSPs but also recruited delayed EPSPs that we interpret as polysynaptic. The short-latency events presumably reflect activation of different mossy or bouton inputs to the granule cell's modest dendritic tree. The longer-latency events might reflect mossy/bouton activation of local excitatory neurons, such as unipolar brush cells. Little is known about such interneurons in the cochlear nucleus, but their counterparts in the cerebellar cortex make presumptive excitatory contact with granule cells and may serve to amplify extrinsic excitatory signals (Alibardi 2004; Mugnaini et al. 1997; Nunzi et al. 2001; Weedman et al. 1996). Given that single inputs appear to have little power to elicit spikes, these data suggest that granule cells are most effectively driven by integration of different inputs. Moreover, given the diverse sources of input to the granule cell system, these cells may integrate signals from different brain regions and sensory modalities.
Our data suggest that both the long-lasting excitatory drive from NMDA receptors and the presence of polysynaptic activity should promote lengthy depolarizing currents during bursts of somatosensory inputs. Although brief current injections could fire well-timed spikes (Fig. 2), the timing of spikes generated by prolonged excitation would likely be determined by the granule cell's intrinsic properties. Both our work, and that of Rusznak et al. (1997), shows that granule cell firing rate reflects the DC level of excitatory current. This firing rate is tightly regulated by A-type (inactivating) K+ currents in these cells (Rusznak et al. 1997). Moreover, because these channels are reprimed by hyperpolarization, as might be experienced during inhibitory signals, it is likely that they could contribute to the length of time strong GABAergic/glycinergic signals halt firing or alter the timing of firing during background levels of inhibitory activity (Street and Manis 2007).
Inhibitory synaptic inputs to granule cells
IPSCs in granule cells were generated by both glycine and GABA. We were not able to determine if these were released from the same axons, but this seems possible based on previous immunohistochemical work (Alibardi 2003a). By determining the reversal potentials for excitatory and inhibitory currents, we could compare the relative strength of synaptic conductances mediated by the three transmitters (Table 1). From these it was clear that in response to low-frequency stimuli the GABA conductance was nearly four times larger than the AMPA-R (glutamate)-mediated response, and the glycine conductance was nearly nine times larger. Although glutamatergic currents summated to produce a slower decaying current, a similar observation was made for IPSCs with current magnitudes consistent with the larger single-stimulus response. Thus GABA- and glycinergic inputs to granule cells synapses are well suited to abruptly overwhelm excitatory synaptic activity. Differences were observed however between GABA- and glycine-mediated transmission with GABAergic currents having amplitudes and decay times that were less than half those of glycinergic currents. Glycinergic IPSC decay times were also much slower than those reported for other auditory neurons, such as the VCN bushy cells (Lim et al. 1999), MNTB (Awatramani et al. 2005), LSO (Nabekura et al. 2004), and MSO (Magnusson et al. 2005). Thus although glycine is the major fast inhibitory transmitter of the auditory brain stem, its effects on granule cells are unique and seem better suited to produces slow changes in the level of excitation than interacting in a rapid, phasic way with well-timed excitatory signals.
The sources of inhibitory inputs to granule cells are not known. Within the DCN, it is possible that stellate, tuberculoventral, cartwheel, or Golgi neurons could make inhibitory inputs on granule cells (Alibardi 1999, 2003a,b; Manis et al. 1994), whereas D-stellate/radiate cells might provide input from the ipsi- or contralateral VCN (Doucet et al. 1999; Schofield and Cant 1996; Wenthold 1987). Because the granule cell pathway is essential in the convergence of nonauditory input to the DCN (Shore 2005), the inhibition we have described could play an essential role in sculpting the response to somatosensory stimuli. If inhibition is triggered by such somatosensory input, then it could alter the temporal profile of granule cell firing, suppress granule cell activation by different sensory modalities, or possibly permit higher brain regions to control the convergence of sensory streams. It is interesting to consider that granule cells might be inhibited by neurons that are primarily activated by auditory input (e.g., D-stellate cells); this would suggest a powerful means for gating multimodal input to the DCN by acoustic signals. Recent studies have indicated that parallel fiber synapses to cartwheel and fusiform cells undergo use-dependent synaptic plasticity (Fujino and Oertel 2003; Tzounopoulos et al. 2004, 2007), which might function to adjust the strength of multimodal signals in the DCN. It will be of interest to determine whether the strength of inhibitory inputs to granule cells also exhibit plasticity. Further studies are needed to determine whether the inhibition is generated from cells activated by parallel fibers (feedback inhibition) or from cells activated by afferent fibers.
Comparison to cerebellar granule cells
The close similarities of auditory granule cells to those of the cerebellar cortex have been described in detail for different species at the anatomical (Hackney et al. 1990; Mugnaini et al. 1980a,b), embryological (Ivanova and Yuasa 1998), and molecular levels (Fünfschilling and Reichardt 2002; Juiz et al. 1996). Our study suggests similarities also in the physiology of excitatory but not in inhibitory transmission. DCN granule cell EPSCs were comparable with respect to amplitude and decay time with those of the cerebellum (D'Angelo et al. 1993; Silver et al. 1992). The presence of NMDA receptor-mediated currents could enhance the summation of EPSPs and is comparable to cerebellar granule cells (D'Angelo et al. 1995). We also observed a slow decay of synaptic current after stimulus trains, as did Overstreet et al. (1999), that may reflect slow clearance of glutamate from glomeruli. Consistent with this idea is the observation that glutamate transporters are excluded from glomeruli and synaptic nests in cochlear nucleus (Josephson and Morest 2003). One major difference is in the prominence of glycinergic transmission in cochlear granule cells because inhibition of cerebellar granule cells appears to be dominated by strong GABAergic input (Farrant and Cull-Candy 1993; Kaneda et al. 1995). Another striking difference is in the sensitivity to background GABA levels. 6 subunit is a key component of GABAA receptors generating tonic current and is expressed in both cerebellar and cochlear granule cells (Funfschilling and Reichardt 2002; Laurie et al. 1992; Varecka et al. 1994), However, subunits, apparently also essential for tonic current (Farrant and Nusser 2005), are absent in the granule cells of the DCN (Campos et al. 2001); presumably this difference accounts for why the auditory granule cells are not sensitive to low concentrations of GABA.
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
1 The online version of this article contains supplemental data. 
Address for reprint requests and other correspondence: L. O. Trussell, Oregon Hearing Research Center and Vollum Institute, L335A, 3181 S. W. Sam Jackson Park Rd., Portland, OR 97239 (E-mail: trussell{at}ohsu.edu)
|
|
REFERENCES |
|---|
|
Alibardi L. Ultrastructural distribution of glycinergic and GABAergic neurons and axon terminals in the rat dorsal cochlear nucleus, with emphasis on granule cell areas. J Anat 203: 31–56, 2003a.[CrossRef][Web of Science][Medline]
Alibardi L. Ultrastructural immunocytochemistry for glycine in neurons of the dorsal cochlear nucleus of the guinea pig. J Submicrosc Cytol Pathol 35: 373–387, 2003b.[Medline]
Alibardi L. Mossy fibers in granule cell areas of the rat dorsal cochlear nucleus from intrinsic and extrinsic origin innervate unipolar brush cell glomeruli. J Submicrosc Cytol Pathol 36: 193–210, 2004.[Medline]
Awatramani GB, Turecek R, Trussell LO. Staggered development of GABAergic and glycinergic transmission in the MNTB. J Neurophysiol 93: 819–828, 2005.
Benson TE, Brown MC. Synapses formed by olivocochlear axon branches in the mouse cochlear nucleus. J Comp Neurol 295: 52–70, 1990.[CrossRef][Web of Science][Medline]
Bormann J, Hamill OP, Sakmann B. Mechanism of anion permeation through channels gated by glycine and gamma-aminobutyric acid in mouse cultured spinal neurons. J Physiol 385: 243–286, 1987.
Brown MC, Liberman MC, Benson TE, Ryugo DK. Brainstem branches from olivocochlear axons in cats and rodents. J Comp Neurol 278: 591–603, 1988.[CrossRef][Web of Science][Medline]
Bukowska D. Morphological evidence for secondary vestibular afferent connections to the dorsal cochlear nucleus in the rabbit. Cells Tissues Organs 170: 61–68, 2002.[CrossRef][Web of Science][Medline]
Burian M, Gstoettner W. Projection of primary vestibular afferent fibers to the cochlear nucleus in the guinea pig. Neurosci Lett 84: 13–17, 1988.[CrossRef][Web of Science][Medline]
Caicedo A, Herbert H. Topography of descending projections from the inferior colliculus to auditory brain stem nuclei in the rat. J Comp Neurol 328: 377–392, 1993.[CrossRef][Web of Science][Medline]
Campos ML, de Cabo C, Wisden W, Juiz JM, Merlo D. Expression of GABA(A) receptor subunits in rat brainstem auditory pathways: cochlear nuclei, superior olivary complex and nucleus of the lateral lemniscus. Neuroscience 102: 625–638, 2001.[CrossRef][Web of Science][Medline]
D'Angelo E, De Filippi G, Rossi P, Taglietti V. Synaptic excitation of individual rat cerebellar granule cells in situ: evidence for the role of NMDA receptors. J Physiol 484: 397–413, 1995.
D'Angelo E, Rossi P, Taglietti V. Different proportions of N-methyl-D-aspartate and non-N-methyl-D-aspartate receptor currents at the mossy fiber-granule cell synapse of developing rat cerebellum. Neuroscience 53: 121–130, 1993.[CrossRef][Web of Science][Medline]
Doucet JR, Ross AT, Gillespie MB, Ryugo DK. Glycine immunoreactivity of multipolar neurons in the ventral cochlear nucleus which project to the dorsal cochlear nucleus. J Comp Neurol 408: 515–531, 1999.[CrossRef][Web of Science][Medline]
Farrant M, Cull-Candy S. GABA receptors, granule cells and genes. Nature 361: 302–303, 1993.[CrossRef][Medline]
Farrant M, Nusser Z. Variations on an inhibitory theme: phasic and tonic activation of GABA(A) receptors. Nat Rev Neurosci 6: 215–229, 2005.[CrossRef][Web of Science][Medline]
Fujino K, Oertel D. Bidirectional synaptic plasticity in the cerebellum-like mammalian dorsal cochlear nucleus. Proc Natl Acad Sci USA 100: 265–270, 2003.
Fünfschilling U, Reichardt LF. Cre-mediated recombination in rhombic lip derivatives. Genesis 33: 160–169, 2002.[CrossRef][Web of Science][Medline]
Golding NL, Oertel D. Context-dependent synaptic action of glycinergic and GABAergic inputs in the dorsal cochlear nucleus. J Neurosci 16: 2208–2219, 1996.
Hackney CM, Osen KK, Kolston J. Anatomy of the cochlear nuclear complex of guinea pig. Anat Embryol 182: 123–149, 1990.[Medline]
Haenggeli CA, Pongstaporn T, Doucet JR, Ryugo DK. Projections from the spinal trigeminal nucleus to the cochlear nucleus in the rat. J Comp Neurol 484: 191–205, 2005.[CrossRef][Web of Science][Medline]
Itoh K, Kamiya H, Mitani A, Yasui Y, Takada M, Mizuno N. Direct projections from the dorsal column nuclei and the spinal trigeminal nuclei to the cochlear nuclei in the cat. Brain Res 400: 145–150, 1987.[CrossRef][Web of Science][Medline]
Ivanova A, Yuasa S. Neuronal migration and differentiation in the development of the mouse dorsal cochlear nucleus. Dev Neurosci 20: 495–511, 1998.[CrossRef][Web of Science][Medline]
Josephson EM, Morest DK. Synaptic nests lack glutamate transporters in the cochlear nucleus of the mouse. Synapse 49: 29–46, 2003.[CrossRef][Web of Science][Medline]
Juiz JM, Helfert RH, Bonneau JM, Campos ML, Altschuler RA. Distribution of glycine and GABA immunoreactivities in the cochlear nucleus: quantitative patterns of putative inhibitory inputs on three cell types. J Hirnforsch 37: 561–574, 1996.[Web of Science][Medline]
Kaneda M, Farrant M, Cull-Candy SG. Whole-cell and single-channel currents activated by GABA and glycine in granule cells of the rat cerebellum. J Physiol 485: 419–435, 1995.
Kanold PO, Young ED. Proprioceptive information from the pinna provides somatosensory input to cat dorsal cochlear nucleus. J Neurosci 21: 7848–7858, 2001.
Kevetter GA, Perachio AA. Projections from the sacculus to the cochlear nuclei in the Mongolian gerbil. Brain Behav Evol 34: 193–200, 1989.[Web of Science][Medline]
Laurie DJ, Seeburg PH, Wisden W. The distribution of 13 GABAA receptor subunit mRNAs in the rat brain. II. Olfactory bulb and cerebellum. J Neurosci 12: 1063–1076, 1992.[Abstract]
Legendre P. The glycinergic inhibitory synapse. Cell Mol Life Sci 58: 760–793, 2001.[CrossRef][Web of Science][Medline]
Lim R, Alvarez FJ, Walmsley B. Quantal size is correlated with receptor cluster area at glycinergic synapses in the rat brainstem. J Physiol 516: 505–512, 1999.
Magnusson AK, Kapfer C, Grothe B, Koch U. Maturation of glycinergic inhibition in the gerbil medial superior olive after hearing onset. J Physiol 568: 497–512, 2005.
Manis PB, Spirou GA, Wright DD, Paydar S, Ryugo DK. Physiology and morphology of complex spiking neurons in the guinea pig dorsal cochlear nucleus. J Comp Neurol 348: 261–276, 1994.[CrossRef][Web of Science][Medline]
May BJ. Role of the dorsal cochlear nucleus in the sound localization behavior of cats. Hear Res 148: 74–87, 2000.[CrossRef][Web of Science][Medline]
Momiyama A, Silver RA, Hausser M, Notomi T, Wu Y, Shigemoto R, Cull-Candy SG. The density of AMPA receptors activated by a transmitter quantum at the climbing fiber-Purkinje cell synapse in immature rats. J Physiol 549: 75–92, 2003.
Mugnaini E, Dino MR, Jaarsma D. The unipolar brush cells of the mammalian cerebellum and cochlear nucleus: cytology and microcircuitry. Prog Brain Res 114: 131–150, 1997.[Web of Science][Medline]
Mugnaini E, Osen KK, Dahl AL, Friedrich VL, Jr, Korte G. Fine structure of granule cells and related interneurons (termed Golgi cells) in the cochlear nuclear complex of cat, rat and mouse. J Neurocytol 9: 537–570, 1980a.[CrossRef][Web of Science][Medline]
Mugnaini E, Warr WB, Osen KK. Distribution and light microscopic features of granule cells in the cochlear nuclei of cat, rat, and mouse. J Comp Neurol 191: 581–606, 1980b.[CrossRef][Web of Science][Medline]
Nabekura J, Katsurabayashi S, Kakazu Y, Shibata S, Matsubara A, Jinno S, Mizoguchi Y, Sasaki A, Ishibashi H. Developmental switch from GABA to glycine release in single central synaptic terminals. Nat Neurosci 7: 17–23, 2004.[CrossRef][Web of Science][Medline]
Nunzi MG, Birnstiel S, Bhattacharyya BJ, Slater NT, Mugnaini E. Unipolar brush cells form a glutamatergic projection system within the mouse cerebellar cortex. J Comp Neurol 434: 329–341, 2001.[CrossRef][Web of Science][Medline]
Oertel D, Young ED. What's a cerebellar circuit doing in the auditory system? Trends Neurosci 27: 104–110, 2004.[CrossRef][Web of Science][Medline]
Ottersen OP, Laake JH, Storm-Mathisen J. Demonstration of a releasable pool of glutamate in cerebellar mossy and parallel fibre terminals by means of light and electron microscopic immunocytochemistry. Arch Ital Biol 128: 111–125, 1990.[Web of Science][Medline]
Overstreet LS, Kinney GA, Liu YB, Billups D, Slater NT. Glutamate transporters contribute to the time course of synaptic transmission in cerebellar granule cells. J Neurosci 19: 9663–9673, 1999.
Petralia RS, Rubio ME, Wang YX, Wenthold RJ. Differential distribution of glutamate receptors in the cochlear nuclei. Hear Res 147: 59–69, 2000.[CrossRef][Web of Science][Medline]
Rusznak Z, Forsythe ID, Brew HM, Stanfield PR. Membrane currents influencing action potential latency in granule neurons of the rat cochlear nucleus. Eur J Neurosci 9: 2348–2358, 1997.[CrossRef][Web of Science][Medline]
Ryugo DK, Haenggeli CA, Doucet JR. Multimodal inputs to the granule cell domain of the cochlear nucleus. Exp Brain Res 153: 477–485, 2003.[CrossRef][Web of Science][Medline]
Santhakumar V, Hanchar HJ, Wallner M, Olsen RW, Otis TS. Contributions of the GABAA receptor alpha6 subunit to phasic and tonic inhibition revealed by a naturally occurring polymorphism in the alpha6 gene. J Neurosci 26: 3357–3364, 2006.
Schofield BR, Cant NB. Origins and targets of commissural connections between the cochlear nuclei in guinea pigs. J Comp Neurol 375: 128–146, 1996.[CrossRef][Web of Science][Medline]
Schofield BR, Coomes DL. Auditory cortical projections to the cochlear nucleus in guinea pigs. Hear Res 199: 89–102, 2005.[CrossRef][Web of Science][Medline]
Shore SE. Multisensory integration in the dorsal cochlear nucleus: unit responses to acoustic and trigeminal ganglion stimulation. Eur J Neurosci 21: 3334–3348, 2005.[CrossRef][Web of Science][Medline]
Shore SE, Vass Z, Wys NL, Altschuler RA. Trigeminal ganglion innervates the auditory brain stem. J Comp Neurol 419: 271–285, 2000.[CrossRef][Web of Science][Medline]
Silver RA, Traynelis SF, Cull-Candy SG. Rapid-time-course miniature and evoked excitatory currents at cerebellar synapses in situ. Nature 355: 163–166, 1992.[CrossRef][Medline]
Street SE, Manis PB. Action potential timing precision in dorsal cochlear nucleus pyramidal cells. J Neurophysiol 97: 4162–4172, 2007.
Sutherland DP, Glendenning KK, Masterton RB. Role of acoustic striae in hearing: discrimination of sound-source elevation. Hear Res 120: 86–108, 1998.[CrossRef][Web of Science][Medline]
Takahashi T, Momiyama A. Single-channel currents underlying glycinergic inhibitory postsynaptic responses in spinal neurons. Neuron 7: 965–969, 1991.[CrossRef][Web of Science][Medline]
Traynelis SF, Silver RA, Cull-Candy SG. Estimated conductance of glutamate receptor channels activated during EPSCs at the cerebellar mossy fiber-granule cell synapse. Neuron 11: 279–289, 1993.[CrossRef][Web of Science][Medline]
Tzounopoulos T, Kim Y, Oertel D, Trussell LO. Cell-specific, spike timing-dependent plasticities in the dorsal cochlear nucleus. Nat Neurosci 7: 719–725, 2004.[CrossRef][Web of Science][Medline]
Tzounopoulos T, Rubio ME, Keen JE, Trussell LO. Coactivation of pre- and postsynaptic signaling mechanisms determines cell-specific spike-timing-dependent plasticity. Neuron 54: 291–301, 2007.[CrossRef][Web of Science][Medline]
Varecka L, Wu CH, Rotter A, Frostholm A. GABAA/benzodiazepine receptor alpha 6 subunit mRNA in granule cells of the cerebellar cortex and cochlear nuclei: expression in developing and mutant mice. J Comp Neurol 339: 341–352, 1994.[CrossRef][Web of Science][Medline]
Weedman DL, Pongstaporn T, Ryugo DK. Ultrastructural study of the granule cell domain of the cochlear nucleus in rats: mossy fiber endings and their targets. J Comp Neurol 369: 345–360, 1996.[CrossRef][Web of Science][Medline]
Weedman DL, Ryugo DK. Pyramidal cells in primary auditory cortex project to cochlear nucleus in rat. Brain Res 706: 97–102, 1996.[CrossRef][Web of Science][Medline]
Weinberg RJ, Rustioni A. A cuneocochlear pathway in the rat. Neuroscience 20: 209–219, 1987.[CrossRef][Web of Science][Medline]
Wenthold RJ. Evidence for a glycinergic pathway connecting the two cochlear nuclei: an immunocytochemical and retrograde transport study. Brain Res 415: 183–187, 1987.[CrossRef][Web of Science][Medline]
Wickesberg RE, Oertel D. Tonotopic projection from the dorsal to the anteroventral cochlear nucleus of mice. J Comp Neurol 268: 389–399, 1988.[CrossRef][Web of Science][Medline]
Wright DD, Ryugo DK. Mossy fiber projections from the cuneate nucleus to the cochlear nucleus in the rat. J Comp Neurol 365: 159–172, 1996.[CrossRef][Web of Science][Medline]
Young ED, Davis KA. Circuitry and function of the dorsal cochlear nucleus. In: Integrative Functions of the Mammalian Auditory Pathway, edited by Oertel D, Popper AN, Fay RR. New York: Springer-Verlag, 2002, p. 160–206.
Zhou J, Nannapaneni N, Shore S. Vessicular glutamate transporters 1 and 2 are differentially associated with auditory nerve and spinal trigeminal inputs to the cochlear nucleus. J Comp Neurol 500: 777–787, 2007.[CrossRef][Web of Science][Medline]
Zhou J, Shore S. Projections from the trigeminal nuclear complex to the cochlear nuclei: a retrograde and anterograde tracing study in the guinea pig. J Neurosci Res 78: 901–907, 2004.[CrossRef][Web of Science][Medline]
Zhou J, Shore S. Convergence of spinal trigeminal and cochlear nucleus projections in the inferior colliculus of the guinea pig. J Comp Neurol 495: 100–112, 2006.[CrossRef][Web of Science][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
, axon boutons. Granule cells typically displayed small cell body ranging from 6 to 8 µm, with 2–3 short dendrites, with claws or small excrescences along their length. Z-stacks were typically made from 30 to 60 images at 1-µm sectioning. Scale bar = 5 µm.
and 
, amplitudes of EPSCs at 4 and 15 ms after the stimuli (n = 8 cells), as indicated in C. E: evoked EPSCs at different frequencies (20, 50, and 100 Hz). The summated current (
) at the foot of the last EPSC was –28 pA at 100 Hz.
