| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Communication Sciences and Disorders, School of Allied Health Sciences, East Carolina University, Greenville, North Carolina; and 2Epstein Laboratory, Department of Otolaryngology-Head and Neck Surgery, University of California, San Francisco, California
Submitted 25 April 2007; accepted in final form 5 August 2007
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
). A similar cochleotopic organization of auditory projections is formed and preserved at essentially all levels of the neuraxis including the auditory cortex. These projections ultimately relay information about sound frequency, which is encoded spatially in the cochlea and thus provide the basis for tonotopic mapping through out the central auditory system. In auditory nuclei receiving binaural input, the projections of each ear are segregated and preserved during ontogeny, and these connections provide a means to locate sounds in the environment through neural computations. Understanding the mechanisms involved in the refinement, segregation, and preservation of these maps is of great interest.
The initial topographical adjustments and final refinement of the cochlear nerve terminal fields in cochlear nuclei of the cat are essentially completed before hearing begins (Cant 1998; Leake et al. 1992, 2002; Snyder and Leake 1997; Snyder et al. 1997). However, the mechanisms underlying these prehearing developmental processes are not yet clear. In the visual system, segregation and refinement of central retinotopic projections occurs in neonatal mammals before onset of vision (McLaughlin and O'Leary 2005). The detailed refinements of retinotopic projections are thought to depend critically on both molecular guidance and patterned neural activity in retinal ganglion cells (Cang et al. 2005; Demas et al. 2006; Hindges et al. 2002; McLaughlin et al. 2003a-c; Mrsic-Flogel et al. 2005; O'Leary and McLaughlin 2005; Ruthazer and Cline 2004; Ruthazer et al. 2003; Stellwagen and Shatz 2002; Yates et al. 2004).
Of particular interest is the observation that spontaneous waves of correlated rhythmic neural activity in retinal ganglion cells provide a signal that is critical for the establishment of fully refined retinotopic maps at all levels of the neuraxis (Cang et al. 2005; Demas et al. 2006; McLaughlin et al. 2003c; Mrsic-Flogel et al. 2005; O'Leary and McLaughlin 2005; Ruthazer et al. 2003; Stellwagen and Shatz 2002). Spontaneous rhythmic discharge has been reported in brain stem auditory pathways of prehearing birds and mammals (Gummer and Mark 1994; Kotak and Sanes 1995; Lippe 1994; Rubsamen and Schafer 1990). The presence of rhythmic bursting among cochlear ganglion cells in the embryonic chicken has also been established recently and is reminiscent of rhythmic discharge patterns in the retina (Jones et al. 2001). The idea that rhythmic discharge patterns play a role in guiding refinements in central cochleotopic fields has been suggested by a number of investigators (Gummer and Mark 1994; Jones et al. 2001, 2006; Lippe 1994, 1995). Not firmly established in the mammal is the question of whether spiral ganglion cells themselves exhibit such spontaneous activity patterns. Neurons of the auditory nerve in cats are known to be spontaneously active as early as 2 days postnatal (P2) (Carlier et al. 1975; Romand 1984; Walsh and McGee 1987). However, it is not clear whether spiral ganglion cells (SGCs) of neonatal kittens exhibit the necessary repetitive spontaneous bursting discharge patterns. We hypothesize that spontaneous rhythmic bursting discharge patterns are present in mammalian SGCs before hearing begins and that such discharge patterns serve to guide activity-dependent central refinements in mammals. To critically address the first of these assertions, we recorded spontaneous activity of SGCs in neonatal kittens ranging in age from P3 to P9.
|
|
METHODS |
|---|
|
The care and use of animals in this study were approved by the Institutional Animal Care and Use Committee at the University of California at San Francisco (UCSF) and conformed to all National Institutes of Health guidelines. All animals included in this study were bred in a closed colony maintained at UCSF. Queens were bred for periods of 
24 h, so the gestation period for each litter was known within ±12 h. Total number of days postconception (dpc, number of days of gestation plus number of days postnatal) is used to define age, because this value correlates best with development of the organ of Corti (Sato et al. 1999) and development of electrophysiological response properties in the auditory nerve (Fitzakerley et al. 1998). This report includes recordings made in eight young kittens ranging in age from 69 to 75 dpc (corresponding to postnatal ages P3–P9) and in one older cat studied at 102 dpc (P36) as summarized in Table 1.
|
and was measured in situ using an electrometer (model WPI 767-B, World Precision Instruments). Electrophysiological recording
The discharge activity of SGCs, ongoing ECG, and ventilation were recorded digitally (12-bit conversion, 44,100-Hz sampling). Electrophysiological activity was recorded for up to 
14 min. Figure 1A shows a typical spiral ganglion neuron recording and signal-to-noise ratio generally achieved in the neonate. Analysis of discharge patterns was accomplished off-line using the digital records. Throughout this report, mean values will be expressed as the mean ± SD (n), where n = sample size. Useful data were obtained from cellular recordings having 1) a signal-to-noise ratio (spike amplitude/ background noise) that permitted unambiguous spike detection and 2) a recording duration that was sufficient to glean useful information about cell discharge. We used the product of the number of spikes multiplied by the time between the first and last spike of a record as an objective metric (spikes-seconds) to exclude cells. Cells not providing 
50 spike-seconds of well resolved spike activity were excluded from analysis.
|
; Goldberg and Fernandez 1971; Jones and Jones 2000). The CVi for spontaneously discharging auditory ganglion cells ranges between 0.6 and 1.0, with a mean of 0.8 in the mature mammal (cat; Walsh et al. 1972) and 1.0 for the mature bird (Jones and Jones 2000). CVi's ranging between 0.7 and 1.0 reflect the near stochastic discharge of mature auditory neurons, which is often described as a quasi-Poisson excitation or renewal process. CVi was reported for all neurons in which recording times were sufficient to determine whether bursting was present and adequate numbers of spikes occurred to permit the calculation of the burst index (below). Burst rate was calculated as a simple count of the total number of bursts divided by the total recording time. Bursts were defined in the context of a preceding period of low activity, and this was signaled by long spike intervals. For the purpose of counting bursts, we defined a minimum preburst period (PBp, ms). This was used as a threshold to define the onset spike of bursts. Here PBp was set equal to three times the mean spike interval for the entire discharge record. This is the same criterion used by Gummer and Mark, (1994). The onset of a prospective burst was signaled by a spike having a spike interval equal to or greater than the PBp. A burst was counted only if the onset spike was followed by a group of two or more spikes having spike intervals less than PBp. The word "spontaneous" in this study means "in the absence of sound stimulation." Determination of bursting versus entrainment
Entrainment to cardiac or ventilation rhythms, which could be misinterpreted as bursting, was determined by 1) visual inspection of the time records; 2) cross-correlation; and 3) probability density functions. Entrained cells were excluded from summary spontaneous discharge data. Bursting was evaluated using two computational methods: 1) probability density functions and 2) burst index (BI), a dimensionless metric useful for identifying and ranking bursting patterns [originally referred to as burst factor (BF)] (Jones and Jones 2000; Jones et al. 2001).
Visual inspection of time records and cross-correlation.
The original spike records were displayed and recording quality confirmed. Spike onset times were scored and evaluated for obvious signs of entrainment (e.g., inspected for regular bursting at rates equal to ECG or ventilation rates). Spike rate and onset time were plotted. Plots of cardiac and ventilation cycle onset times were added to spike time plots. These combined plots were examined visually in detail for evidence of SGC entrainment. Neurons found to discharge in a manner that was linked in time to either cycle were designated as "entrained." An example of such a plot and a detailed description of the graphical method used for inspection is provided in Supplemental Fig. 1.1
Cross-correlation of spike times with cardiac or ventilation cycle onset times involved constructing two separate period histograms for the recorded spike onset times (Perkel et al. 1967). One period histogram was based on cardiac cycle onsets and the other on ventilation cycle onsets. The term "period histogram" will be used herein to indicate the final result of a cross-correlation. The duration of the time window for a cross-correlation and resultant period histogram was equal to the mean cardiac or ventilator period, and we used bins that were 10 ms long. The time window was centered on the onset of each cardiac or ventilation cycle. Theoretically, for a neuron exhibiting discharge activity that is randomly generated and independent of the cardiac and/or ventilation cycle, the distribution of the period histogram amplitudes should be relatively flat, reflecting a uniform probability distribution across the time window (Perkel et al. 1967).
PROBABILITY DENSITY FUNCTIONS. A simple spike time histogram was created to represent discharge probability as a function of time. The number of spikes occurring in the space of each time bin was taken as the discharge amplitude (spike count), and this was plotted as a function of time, where each time bin corresponded to one cycle period of ECG or ventilation. The time represented by the total number of bins was equal to the total time of the recording. If spike discharge was strictly linked to the ECG or ventilation cycle, each cycle (bin) would have a comparable number of counts, and the spike discharge amplitudes would distribute evenly across all bins throughout the time histogram. Similarly, if spike discharge was randomly generated and independent of ECG and ventilation events, spike discharge amplitude would also be randomly distributed throughout the time histogram. However, if spike discharges were nonstochastic and occurred in recurrent bursts interspersed with silent periods, the distribution of spike counts would not be uniform and would not follow a Poisson distribution as described below.
USING THE VC TO IDENTIFY NONUNIFORM PROBABILITY DISTRIBUTIONS.
To construct frequency distribution functions for period histograms and spike time histograms, the amplitude of a given time bin i (measured in spikes) was denoted g(i), i = 1, 2,..., m, where m was the total number of bins in the period or time histogram window. Values of the frequency distribution array, p(k), were calculated by counting the number of times that g(i) was exactly equal to k (k = 0, 1, 2, ...) Thus p(2) was the number of bins of the period or time histogram that contained exactly two spikes. The weighted mean (Mw, also widely referred to as expected value) and variance (
w2) of the frequency distribution function p(k) were calculated as follows
 |
 |
 |
The VCs of sampled SGC spike discharge data were contrasted with the VCs of Monte Carlo simulations having an equal number of spikes. Spike onset times in simulations were randomly assigned over the same recording period as sampled data, and these "simulated spike trains" were cross-correlated with the mechanical event cycles (cardiac and ventilation) present in the sampled data records. This process produced a period histogram for simulated data. In a similar fashion, the simulated spike trains were accumulated in appropriate time bins of the spike time histogram and the results were compared with those for sampled spike data. Frequency distribution functions were constructed for the simulations, and VC values were calculated from the weighted mean and variance as described above.
INTERPRETING VCS.
Theoretically, if SGC spike discharges are independent of cardiac and ventilation cycles, the distribution of period histogram amplitudes, [g(i)], should approach that produced by a stochastic Poisson excitation process and approximate those produced by Monte Carlo simulations. The same is true for amplitudes (spike counts) of the spike time histograms. The resulting frequency distribution [p(k)] would in those cases follow or approach the theoretical Poisson probability distribution P(k), which is given by P(k) = (ck/k!)e–c,, where c is a constant equal to the mean counts per bin, k is the spike count, and k! is k factorial (Batschelet 1979). The value P(k) for any given k is the fraction of the total number of bins that will contain k spike counts if each spike event is equally likely to enter any of the bins and must enter one. For a true Poisson frequency distribution, the resulting VC is 1.0 (VC = 1.0), because in that case, the weighted mean and variance are equal (w2 = Mw). Values of VC substantially larger or smaller than 1.0 were taken to indicate substantial deviation from a Poisson distribution and thus were indicative of spike event times that were linked to the ventilation or cardiac cycle in period histograms or that were produced by nonstochastic bursting activity represented in spike time histograms. Values of VCs obtained from simulated and sampled data were compared and statistical contrasts made using the Wilcoxon signed-rank test.
NOMENCLATURE FOR VCS. As noted above, general reference to the VC will be made using the acronym VC. VCs derived from period histograms (PHs) will be referred to as VCPH and those derived from time histograms (THs) as VCTH. To distinguish VCs based on cardiac (ECG) and ventilation (vent) data, the corresponding subscripts will be added (i.e., VCPH ECG, VCPH vent, VCTH ECG, VCTH vent). Finally, there are occasions where VCs are pooled for ECG and ventilation data, and these are referred to as VCPH ECGandvent or VCTH ECGandvent.
BI.
BI was calculated only for those records with 11 or more spikes. The intervals in each spike record were ranked from longest to shortest. The longest intervals were used to calculate BI. The four longest intervals were used for records containing 11–80 spikes. If >80 spikes were recorded, the number of longest intervals used was calculated as 5% of the total number of spikes (truncated to an integer). The BI was calculated as follows
 |
 |
; Jones et al. 2001). |
|
RESULTS |
|---|
|
; Sanes and Walsh 1998). The discharge patterns of 17 of these 66 useful cells (25%) were clearly seen graphically to be in part synchronized with the mechanical action of the heart beat or with the respiratory ventilation cycle. We have designated all 17 of these cells as entrained cells. Data from these entrained cells were not used to generate metrics for spontaneous discharge activity, although data from six of the cells were used to show the effects of entrainment on period histograms and VCs. The discharge activity of the remaining well-characterized 49 SGCs (48 neurons in neonates) was designated as "spontaneous" (nonentrained), and these data served as the basis for the quantitative description of SGC spontaneous discharge activity. Cross-correlation: period histograms and entrainment
An example of a typical period histogram of an entrained neuron is shown in Supplemental Fig. 2. Spike counts in the PHs of entrained cells showed considerable bias in spike distribution toward some phase of the ECG or ventilation cycle. VCs for such PHs (VCPH ECG and VCPH vent) were considerably >1.0 and ranged 27. The mean value for the pooled VCs was 8.6 (see VCPH ECGandvent; Table 2), and this served as an illustration of the values expected from spike discharge driven in part by cardiac or ventilation cycles.
|
|
) and simulated spike data (
) showed considerable overlap, indicating that spike discharge in sampled neurons generally approximated the distribution for random spike discharge for both ECG and ventilation cycles. The range of VCs for both ECG and ventilation PHs was slightly larger for sampled data. Only a few cells (5) had VCs >2.0, which was the upper limit of VCs obtained from simulated data.
|
8.6; Table 2). Together, these results showed that entrainment by ECG and ventilation played a negligible role in forming the activity patterns of the spontaneous group. Spike rates
As reported by many investigators, the spontaneous discharge rate of SGCs in the neonatal kitten was considerably lower than rates observed for most cells in older kittens and adults (Walsh and Romand 1992). Figure 3 summarizes the distribution of spontaneous spike discharge rates for neonates at various ages postconception. Spike rates ranged from 0.06 to 56 spikes/s. Only two ganglion cells had spike rates >10 spikes/s. The mean spontaneous discharge rate across all 48 SGCs in neonates was 3.09 ± 8.2 (48) spikes/s. Neonatal spike rates reported here are similar to those reported by Romand (1984) for 6-day-old kittens (i.e., 4 spikes/s, figure line labeled C). This contrasts markedly with the rate of 88.6 spikes/s for an SGC recorded in the 1-mo-old kitten; a spontaneous discharge rate that falls within the upper ranges reported for mature cats (60–100 spikes/s) (Kiang 1965; Liberman 1978; Walsh and McGee 1987; Walsh et al. 1972). Romand (1984) reported the mean discharge rates for the adult as 45 spikes/s and for a 1-mo-old kitten as 27 spikes/s, and these values are represented on Fig. 3 (lines labeled A and B, respectively).
Bursting patterns
The pattern of spike discharge over time was also remarkably different for neonates in comparison to mature animals as shown in Figs. 4 and 5. Figure 4 shows examples of the original recordings from SGCs of a mature animal (top trace) and a neonate (bottom trace). The discharge in the mature animal reflected relatively continuous, irregular spike discharges that contained no appreciable long periods of inactivity, whereas the discharge pattern in the neonate occurred in a series of bursts interrupted by long silent periods. The variation in discharge rate over much longer periods of time is better appreciated in discharge rate plots as shown in Fig. 5. The panels of Fig. 5 show results from the same two cells shown in Fig. 4. Each point in the plots of Fig. 5 represents the onset time and the reciprocal of the spike interval. The reciprocal of the time interval gives the corresponding spike rate in spikes per second for the interval, which is sometimes referred to as the ongoing "instantaneous spike rate." Figure 5A clearly shows the steady, near-stochastic spontaneous discharge of an SGC in a 1-mo-old animal (g001D). The spontaneous discharge pattern shown in Fig. 5A gave a CVi = 0.88 and a BI = 0.65. A CVi approaching 1.0 and a BI <1.1 are consistent with an exponential distribution of interspike intervals, which is characteristic of a stochastic discharge pattern and the absence of a bursting process (Jones and Jones 2000; Jones et al. 2001; Kiang 1965; Walsh et al. 1972). In contrast, Fig. 5B shows a typical pattern of discharge for the neonatal animal (sg15014D). This pattern of discharge gave a CVi = 4.11 and a BI = 8.8. These large values of CVi and BI are indicative of a nonstochastic discharge process. Note the striking repetitive bursting periods in the record obtained from the neonate (burst rate was 4.1 bursts/min).
|
|
151 s. Bursting discharge patterns were characteristic of all but one (k001bD) neuron in neonates. Only three neurons had BI values <2.0, and only the aforementioned single cell showed the absence of bursting based on the graphical discharge record and on the values for CVi and BI (k001bD: CVi = 1.08; BI = 0.78). CVi ranged from 1.08 (1 nonbursting cell) up to 10.4, and the mean CVi was 2.9 for all nonentrained neonatal neurons. The mean BI for the group was 5.2 and BI ranged from 0.78 (1 nonbursting cell) to 16.5. In all but the one cell, CVi and BI values were >1.1, reflecting dominant nonstochastic discharge patterns associated with the basic spontaneous spike activity and the presence of prominent periods of both silence and high activity (bursting). The overall distributions of values for both CVi and BI were independent of the duration of the recorded activity and CVi was independent of the mean discharge rate. These distributions of CVi and BI are shown in Figs. 6 and 7 . Salient quantitative features of spontaneous activity in the neonatal and 1-mo-old kittens are summarized in Table 3.
|
|
|
) and simulated () data showed striking differences in distributions. More pronounced bursting is associated with larger VCTH values. Those SGC neurons with the lowest VC values for sampled data were cells with very low discharge rates and relatively few spikes. The mean values for sampled and simulated data were substantially different whether the analysis time bins were based on the ECG period or ventilation period (see VCTH ECG and VCTH vent, P < 0.001; Table 2). The means for the pooled VCTHs (ECG and ventilation) for sampled data were VCTH ECGandvent = 8.2 ± 8.9(49) and for simulated data were VCTH ECGandvent = 1.2 ± 0.4 (49). These remarkable differences further showed the nonstochastic nature of spontaneous SGC bursting discharge patterns for the group of neurons as a whole.
|
Generally, the period of time between recurring bursts was not constant for a given cell, but rather, it varied considerably throughout long records. The ganglion cell sg15014D shown in Fig. 5B provides a good example of such variation. Burst rate ranged from 0.3 to 23.4 bursts/min for the cells of the spontaneous group. The mean rate was 4.3 ± 5.0 (36). The distribution of burst rates across cells is shown in Fig. 9. For most cells (92%), burst rates were <11 bursts/min, and the majority of these were <5 bursts /min. There was no evidence of a systematic increase in burst rate with age in the kitten.
|
|
|
DISCUSSION |
|---|
|
). Burst rates for prehearing chicken embryos ranged from 1 to 54 bursts/min, with only one cell producing a rate >30 (means: through stage 39 = >9.8 burst/min, through stage 43 = >35 bursts/min) Based on the average interburst interval reported, Gummer and Mark (1994) apparently observed burst rates on the order of 7.5–95 bursts/min. The range of burst rates in the kitten (i.e., 0.3–23) overlapped the lower end of those reported for the chicken and wallaby. We did not see a systematic increase in burst rate with age in the kitten as seen in the bird (Jones et al. 2001).
The spontaneous discharge rates of SGCs of early neonates reported here and elsewhere (P2; Carlier et al. 1975; Romand 1984; Walsh and McGee 1987) were very low, and it is reasonable to question whether such low discharge rates would be effective in influencing synaptic refinements in higher-order auditory projections. There is evidence in neonates, however, that central postsynaptic neurons of auditory projections exhibit unusual substantial and prolonged responses to single presynaptic spike discharges, and respond even more dramatically to bursts of such activity (Kotak and Sanes 1995). Thus neonatal central circuits may be predisposed to respond to isolated spikes and bursting patterns present in neonatal SGC neurons.
Distinguishing the roles of spontaneous and driven rhythmic activity in neonatal kittens
Stimulus-driven rhythmic responses of auditory nerve fibers have been reported during the first postnatal week in kittens (Carlier et al. 1975; Pujol 1972; Walsh and Mcgee 1986, 1988). This sound evoked activity should be distinguished from the spontaneous rhythmic patterns shown here for two reasons. First, given the high thresholds of auditory fibers to sound (>90 dB SPL), it is unlikely that external acoustic stimuli play any significant role in driving activity patterns before P10. Second, even if sound levels were sufficient to activate cells during this period, one might expect external stimuli to interfere with the normal endogenous spontaneous rhythms. Hypothetically this could degrade rather than contribute to activity-dependent processes underlying refinements and segregation centrally.
Implications of these findings for auditory development
There are at least two important questions to address. First, what mechanisms give rise to the bursting rhythms in SGCs, and second, what possible role does such activity play in development?
BURSTING MECHANISM.
We have argued previously that, in the absence of sound stimuli, the discharge patterns of SGCs in prehearing animals as well as in the mature animal likely arise endogenously from within the cochlear neuroepithelium (Jones and Jones 2000; Jones et al. 2001, 2006). Our hypothesis is that SGC discharge patterns depend on the release of excitatory neurotransmitter from hair cells at afferent synapses. Therefore the hair cell is seen as the primary source of excitation for spontaneous SGC bursting activity. This hypothesis is consistent with long held ideas about spontaneous ganglion cell discharge in mature animals and is supported by considerable evidence (Adrian 1943; Annoni et al. 1984; Flock and Russel 1976; Furukawa and Ishii 1967; Furukawa et al. 1972; Glowatzki and Fuchs 2002; Harris and Flock 1967; Hudspeth 1986; Ishii et al. 1971; Katz 1969; Rossi et al. 1977; Schessel et al. 1991; Siegel 1992; Siegel and Dallos 1986). However in the mature animal, excitation is a relatively steady stochastic process. Given the nonstochastic bursting characteristics of SGC activity in the prehearing animal, we presume that a corresponding nonstochastic excitatory process must exist to generate such activity. Indeed, there is evidence suggesting that the inner hair cell (IHC) itself could provide the mechanism. During prehearing stages, IHCs are capable of generating spontaneous recurrent action potentials [mammals: (Kros et al. 1998; Marcotti et al. 2003a,b, 2004); birds: (Fuchs and Evans 1990; Fuchs and Sokolowski 1990; Sokolowski and Cunningham 1999)]. Here we propose that spike discharge in the IHC leads to the discharge of SGCs and recurrent IHC spikes form the basis for the bursting patterns observed in this study. The cochlear neuroepithelium is likely competent to mediate such activity at these ages in that the synaptic apparatus is present, postsynaptic glutamate receptors are expressed, and they are functional (Glowatzki and Fuchs 2002; Knipper et al. 1997; Luo et al. 1995; Sobkowicz et al. 1982). Moreover, IHC discharge has been shown to trigger exocytosis (Beutner and Moser 2001) and excite SGC terminal boutons (Glowatzki and Fuchs 2002). The ability of IHCs to generate spikes is transient, but it exists precisely during neonatal prehearing periods and is lost with the onset of hearing.
ACTIVITY-DEPENDENT MECHANISMS, COCHLEOTOPIC REFINEMENT, AND SPONTANEOUS BURSTING DISCHARGE PATTERNS.
As outlined in the Introduction, waves of spontaneous rhythmic discharge appear in the retina before the onset of vision. These patterns of activity are thought to serve Hebbian processes (Hebb 1949) operating centrally during the refinement of retinotopic projections (Katz and Shatz 1996; Ruthazer and Cline 2004). Retinal waves serve to synchronize the discharge of ganglion cells of small focal retinal regions, and this process depends critically on local retinal networks (see Feller 2002). The spatiotemporal pattern of activity is thought to be more important than the mere presence of ganglion cell activity. Indeed, the evidence suggests that central refinement and segregation requires correlated activity arising from spatially localized receptive areas on the retina (McLaughlin et al. 2003c; Mrsic-Flogel et al. 2005; Ruthazer and Cline 2004).
The rhythmic bursting patterns in SGCs of the kitten are reminiscent of retinal discharges. If this SGC activity serves Hebbian cochleotopic refinement processes centrally, there must be a mechanism that can coordinate the activation of neighboring SGC to produce a correlated discharge. To our knowledge, circuitry like the cellular networks generating retinal waves does not exist in the cochlea. However, there are other unique features of cochlear afferent innervation that can lead to a correlated pattern of discharge in small groups of adjacent SGCs.
Generally, in the mature cochlea, each radial SGC innervates only one IHC, although each IHC is innervated by a group of 20–40 SGCs (Liberman 1980, 1982; Spoendlin 1969). This anatomical fact ensures that the activity of all SGCs innervating one IHC is strictly dependent on and thus correlated with activity in the common presynaptic receptor cell. A somewhat less precise but similar configuration is already present at birth in the mammal (Perkins and Morest 1975; Simmons et al. 1991). According to these reports, during the first 2 neonatal days, the vast majority of SGC dendrites branch to innervate two or three adjacent IHCs, although occasionally more contacts are seen (8). Thereafter, dendritic branching and multiple IHC contacts decrease abruptly, and by P3, >60% of SGCs innervate a single IHC. By the end of the first neonatal week, the adult configuration is in place.
Given the common termination of scores of ganglion cells on a single IHC, each IHC must coordinate the activity of a group of SGCs. At the earliest ages, some ganglion cells might be activated by two or more IHCs, but this likely resolves quickly to an exclusive functional relationship with one IHC. Because IHCs are capable of regenerative spike discharge and excitation of SGC terminal boutons at these stages (as discussed above), it follows that such discharge would lead to the correlated activity of scores of SGCs. We hypothesize that for each individual SGC recorded in this study, there were scores of companion SGCs that discharged in a linked coordinated manner. In the presence of repetitive IHC spike discharge, SGCs would exhibit correlated recurrent bursting discharge activity. Ultimately, such correlated activity would be linked to the position occupied by a single IHC in the organ of Corti. This hypothesis simultaneously accounts for the required correlated activity and the strict local cochleotopic organization necessary to drive central Hebbian refinement processes.
An additional mechanism capable of orchestrating correlated activity among IHCs within a small region of the organ of Corti may be required to further refine the spatial organization of higher-order cochleotopic groups. Cochlear efferent terminals are known to form transiently on IHCs in the neonate and one possible mechanism is that activity in these efferent terminations could serve to modulate and thus correlate activity among IHCs locally (Glowatzki and Fuchs 2000; Goutman et al. 2005; Walsh and Mcgee 1997). A complete picture of factors contributing to the patterns of activity in IHCs and SGCs has yet to emerge. It is conceivable that correlated activity of IHCs and in turn SGCs also contributes to ear-specific segregation and refinement at levels of the auditory pathway where binaural projections converge. The working hypotheses outlined above remain to be critically tested.
In addition to guiding central refinements, the recurrent IHC spike discharge may be responsible in part for refinement of SGC terminal contacts in the organ of Corti itself. Although, IHCs are clearly capable of sustaining recurrent spike discharges in the neonate, the regulation of such discharge in vivo remains an exciting open question.
The spontaneous bursting patterns reported here for the kitten were similar to patterns found in cochlear ganglion cells of the early prehearing chicken embryo (Jones et al. 2001) and both patterns remind us of spontaneous retinal discharges. Despite many differences in the cochleas of birds and mammals, the discharge similarities during prehearing periods may reflect shared developmental processes. Whether the working hypothesis proposed here for the mammal applies to the bird is an interesting question and remains to be explored.
These results are consistent with the hypothesis that spontaneous rhythmic bursting activity in SGC of prehearing mammals plays a critical role in the refinement of central cochleotopic projections. Clear from the present study is the fact that these spontaneous discharge patterns are present during the period of final refinement of cochleotopic projections from the cochlea to the cochlear nucleus (Leake et al. 2002). Furthermore, the rhythmic bursting discharge of ganglion cells occurs in vivo precisely during the transient prehearing period following the onset of repetitive IHC spike discharge in vitro (Kros et al. 1998; Marcotti et al. 2003a,b, 2004), the appearance of synaptic ribbons and onset of IHC exocytosis (Beutner and Moser 2001; Glowatzki and Fuchs 2002; Sobkowicz et al. 1982), and during the stabilization of terminal contacts of individual SGCs onto single IHCs and reorganization of efferent terminals (Glowatzki and Fuchs 2000; Goutman et al. 2005; Perkins and Morest 1975; Simmons et al. 1991). Finally, given the putative critical role for the IHC in the hypothesis above, it is worth noting that cochleotopic refinements in the cochlear nucleus fail to occur in neonates when IHCs are destroyed at birth by ototoxic drugs (Leake et al. 2006). Such a lesion would, among other things, eliminate normal spontaneous activity in SGCs. Together, these observations suggest both a mechanism for stimulus-independent rhythmic bursting of ganglion cells and a possible role for such discharge patterns in guiding the prehearing activity-dependent refinement in central auditory relays.
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
1 The online version of this article contains Supplementary Material. 
Address for reprint requests and other correspondence: T. A. Jones, Communication Sciences and Disorders, Allied Hlth. Sci., East Carolina Univ., Health Sciences Bldg., Rm. 3310P, Greenville, NC 27858-4353 (E-mail: jonesti{at}ecu.edu)
|
|
REFERENCES |
|---|
|
Anastasio TJ, Correia MJ, Perachio AA. Spontaneous and driven responses of semicircular canal primary afferents in the unanesthetized pigeon. J Neurophysiol 54: 335–347, 1985.
Annoni JM, Cochran SL, Precht W. Pharmacology of the vestibular hair cell-afferent fiber synapse in the frog. J Neurosci 4: 2106–2116, 1984.[Abstract]
Batschelet E. Introduction to Mathematics for Life Scientists. Berlin: Springer-Verlag, 1979.
Beutner D, Moser T. The presynaptic function of mouse cochlear inner hair cells during development of hearing. J Neurosci 21: 4593–4599, 2001.
Cang J, Renteria RC, Kaneko M, Liu X, Copenhagen DR, Stryker MP. Development of precise maps in visual cortex requires patterned spontaneous activity in the retina. Neuron 48: 797–809, 2005.[CrossRef][Web of Science][Medline]
Cant N. Structural development of the mammalian auditory pathways. In: Development of the Auditory System, edited by Rubel EW, Popper AN, and Fay RR. New York: Springer, 1998, p. 315–413.
Carlier E, Abonnenc M, Pujol R. Maturation des responses unitaires a la stimulation tonale dans le nerf cochleaire du chaton. J Physiol Paris 70: 129–138, 1975.[Medline]
Demas J, Sagdullaev BT, Green E, Jaubert-Miazza L, McCall MA, Gregg RG, Wong ROL, Guido W. Failure to maintain eye-specific segregation in nob, a mutant with abnormally patterned retinal activity. Neuron 50: 247–259, 2006.[CrossRef][Web of Science][Medline]
Feller MB. The role of nAChR-mediated spontaneous retinal activity in visual system development. J Neurobiol 53: 556–567, 2002.[CrossRef][Web of Science][Medline]
Fitzakerley JL, Mcgee J, Walsh EJ. Paradoxical relationship between frequency selectivity and threshold sensitivity during auditory-nerve fiber development. J Acoust Soc Am 103: 3464–3477, 1998.[CrossRef][Web of Science][Medline]
Flock A, Russel I. Inhibition by efferent nerve fibres: action on hair cells and afferent synaptic transmission in the lateral line canal organ of the burbot lota lota. J Physiol 257: 45–62, 1976.
Fuchs PA, Evans MG. Potassium currents in hair cells isolated from the cochlea of the chick. J Physiol 429: 529–551, 1990.
Fuchs PA, Sokolowski BHA. The acquisition during development of Ca-activated potassium currents by cochlear hair cells of the chick. Proc R Soc Lond B Biol Sci 241: 122–126, 1990.[Medline]
Furukawa T, Ishii Y. Neurophysiological studies on hearing in goldfish. J Neurophysiol 30: 1377–1403, 1967.
Furukawa T, Ishii Y, Matsuura S. Synaptic delay and time course of post-synaptic potentials at the junction between hair cells and eighth nerve fibers in the goldfish. Jpn J Physiol 22: 617–635, 1972.[Web of Science][Medline]
Glowatzki E, Fuchs PA. Cholinergic synaptic inhibition of inner hair cells in the neonatal mammalian cochlea. Science 288: 2366–2368, 2000.
Glowatzki E, Fuchs PA. Transmitter release at the hair cell ribbon synapse. Nat Neurosci 5: 147–154, 2002.[CrossRef][Web of Science][Medline]
Goldberg JM, Fernandez C. Physiology of peripheral neurons innervating semicircular canals of the squirrel monkey. III. Variations among units in their discharge properties. J Neurophysiol 34: 676–684, 1971.[Web of Science][Medline]
Goutman JD, Fuchs PA, Glowatzki E. Facilitating efferent inhibition of inner hair cells in the cochlea of the neonatal rat. J Physiol 566: 49–59, 2005.
Gummer AW, Mark RF. Patterned neural activity in brain stem auditory areas of a prehearing mammal, the tammar wallaby Macropus eugenii. Neuroreport 5: 685–688, 1994.[Web of Science][Medline]
Harris GG, Flock A. Spontaneous and evoked activity from the xenopus laevis lateral line. In: Lateral Line Detectors, edited by Cahn PH. Bloomington, IN: Indiana University Press, 1967, p. 135–161.
Hebb DO. The Organization of Behavior. New York: J. Wiley, 1949.
Hindges R, McLaughlin T, Genoud N, Henkemeyer M, O'Leary DD. EphB forward signaling controls directional branch extension and arborization required for dorsal-ventral retinotopic mapping. Neuron 35: 475–487, 2002.[CrossRef][Web of Science][Medline]
Hudspeth AJ. The ionic channels of a vertbrate hair cell. Hear Res 22: 21–27, 1986.[CrossRef][Web of Science][Medline]
Ishii Y, Matsuura S, Furukawa T. Quantal nature of transmission at the synapse between hair cells and eighth nerve fibers. Jpn J Physiol 21: 79–89, 1971.[Web of Science][Medline]
Jones TA, Jones SM. Spontaneous activity in the statoacoustic ganglion of the chicken embryo. J Neurophysiol 83: 1452–1468, 2000.
Jones TA, Jones SM, Paggett KC. Primordial rhythmic bursting in embryonic cochlear ganglion cells. J Neurosci 21: 8129–8135, 2001.
Jones TA, Jones SM, Paggett KC. Emergence of hearing in the chicken embryo. J Neurophysiol 96: 128–141, 2006.
Katz B. The Release of Neural Transmitter Substances. Liverpool, UK: University Press, 1969.
Katz LC, Shatz CJ. Synaptic activity and the construction of cortical circuits. Science 274: 1133–1138, 1996.
Kiang NYS. Discharge Patterns of Single Fibers in the Cat's Auditory Nerve. Cambridge, MA: MIT Press, 19654.
Knipper M, Kopschall I, Rohbock K, Kopke AKE, Bonk I, Zenner H-P. Transient expression of NMDA receptors during rearrangement of AMPA-receptor-expressing fibers in the developing inner ear. Cell Tissue Res 287: 23–41, 1997.
Kotak VC, Sanes DH. Synaptically evoked prolonged depolarizations in the developing auditory system. J Neurophysiol 74: 1611–1620, 1995.
Kros CJ, Ruppersberg JP, Rusch A. Expression of a potassium current in inner hair cells during development of hearing in mice. Nature 394: 284, 1998.[CrossRef][Medline]
Leake PA, Hradek GT, Chair L, Snyder RL. Neonatal deafness results in degraded topographic specificity of auditory nerve projections to the cochlear nucleus in cats. J Comp Neurol 497: 13–31, 2006.[CrossRef][Web of Science][Medline]
Leake PA, Snyder RL, Hradek GT. Postnatal refinement of auditory nerve projections to the cochlear nucleus in cats. J Comp Neurol 448: 6–27, 2002.[CrossRef][Web of Science][Medline]
Leake PA, Snyder RL, Merzenich MM. Topographic organization of the cochlear spiral ganglion demonstrated by restricted lesions of the anteroventral cochlear nucleus. J Comp Neurol 320: 468–478, 1992.[CrossRef][Web of Science][Medline]
Liberman MC. Auditory-nerve response from cats raised in a low-noise chamber. J Acoust Soc Am 63: 442–455, 1978.[CrossRef][Web of Science][Medline]
Liberman MC. Morphological differences among radial afferent fibers in the cat cochlea: an electon-microscopic study of serial sections. Hear Res 3: 45–63, 1980.[CrossRef][Web of Science][Medline]
Liberman MC. Single-neuron labeling in the cat auditory nerve. Science 216: 1239–1241, 1982.
Lippe WR. Rhythmic spontaneous activity in the developing avian auditory system. J Neurosci 14: 1486–1495, 1994.[Abstract]
Lippe WR. Relationship between frequency of spontaneous bursting and tonotopic position in the developing avian auditory system. Brain Res 703: 205–213, 1995.[CrossRef][Web of Science][Medline]
Luo L, Brumm D, Ryan AF. Distribution of non-NMDA glutamate receptor mRNAs in the developing rat cochlea. J Comp Neurol 361: 372–382, 1995.[CrossRef][Web of Science][Medline]
Marcotti W, Johnson SL, Holley MC, Kros CJ. Developmental changes in the expression of potassium currents of embryonic, neonatal and mature mouse inner hair cells. J Physiol 548: 383–400, 2003a.
Marcotti W, Johnson SL, Kros CJ. A transiently expressed SK current sustains and modulates action potential activity in immature mouse inner hair cells. J Physiol 560: 691–708, 2004.
Marcotti W, Johnson SL, Rusch A, Kros CJ. Sodium and calcium currents shape action potentials in immature mouse inner hair cells. J Physiol 552: 743–761, 2003b.
McLaughlin T, Hindges R, O'Leary DD. Regulation of axial patterning of the retina and its topographic mapping in the brain. Curr Opin Neurobiol 13: 57–69, 2003a.[CrossRef][Web of Science][Medline]
McLaughlin T, Hindges R, Yates PA, O'Leary DD. Bifunctional action of ephrin-B1 as a repellent and attractant to control bidirectional branch extension in dorsal-ventral retinotopic mapping. Development 130: 2407–2418, 2003b.
McLaughlin T, O'leary DDM. Molecular gradients and development of retinotopic maps. Annu Rev Neurosci 28: 327–355, 2005.[CrossRef][Web of Science][Medline]
McLaughlin T, Torborg CL, Feller MB, O'Leary DD. Retinotopic map refinement requires spontaneous retinal waves during a brief critical period of development. Neuron 40: 1147–1160, 2003c.[CrossRef][Web of Science][Medline]
Mrsic-Flogel TD, Hofer SB, Creutzfeldt C, Cloez-Tayarani I, Changeux JP, Bonhoeffer T, Hubener M. Altered map of visual space in the superior colliculus of mice lacking early retinal waves. J Neurosci 25: 6921–6928, 2005.
O'Leary DD, McLaughlin T. Mechanisms of retinotopic map development: ephs, ephrins, and spontaneous correlated retinal activity. Prog Brain Res 147: 43–65, 2005.[Web of Science][Medline]
Perkel DH, Gernstein GL, Moore GP. Neuronal spike trains and stochastic point processes I. The single spike train. Biophys J 7: 391–418, 1967.[Web of Science][Medline]
Perkins RE, Morest DK. A study of cochlear innervation patterns in cats and rats with the Golgi method and Nomarski optics. J Comp Neurol 163: 129–158, 1975.[CrossRef][Web of Science][Medline]
Pujol R. Development of tone-burst responses along the auditory pathway in the cat. Acta Otolaryngol (Stockh) 74: 383–391, 1972.[Medline]
Romand R. Functional properties of auditory-nerve fibers during postnatal development in the kitten. Exp Brain Res 56: 395–402, 1984.[Web of Science][Medline]
Rossi ML, Valli P, Casella C. Post-synaptic potentials recorded from afferent nerve fibres of the posterior semicircular canal in the frog. Brain Res 135: 67–75, 1977.[CrossRef][Web of Science][Medline]
Rubsamen R, Schafer M. Ontogenesis of auditory fovea representation in the inferior colliculus of the Sri Lankin rufous horseshoe bat, Rhinolophus rouxi. J Comp Physiol [A] 167: 757–769, 1990.
Ruthazer ES, Akerman CJ, Cline HT. Control of axon branch dynamics by correlated activity in vivo. Science 301: 66–70, 2003.
Ruthazer ES, Cline HT. Insights into activity-dependent map formation from the retinotectal system: a middle-of-the-brain perspective. J Neurobiol 59: 134–146, 2004.[CrossRef][Web of Science][Medline]
Sanes DH, Walsh EJ. The development of central auditory processing. In: Development of the Auditory System, edited by Rubel EW, Popper AN, and Fay RR. New York: Springer-Verlag, 1998, p. 271–314.
Sato M, Leake PA, Hradek GT. Postnatal development of the organ of corti in cats: a light microscopic morphometric study. Hear Res 127: 1–13, 1999.[CrossRef][Web of Science][Medline]
Schessel DA, Ginzberg R, Highstein SM. Morphophysiology of synaptic transmission between type I hair cells and vestibular primary afferents. An intracellular study employing horseradish peroxidase in the lizard, Calotes versicolor. Brain Res 544: 1–16, 1991.[CrossRef][Web of Science][Medline]
Siegel JH. Spontaneous synaptic potentials from afferents terminals in the guinea pig cochlea. Hear Res 59: 85–92, 1992.[CrossRef][Web of Science][Medline]
Siegel JH, Dallos P. Spike activity recorded from the organ of Corti. Hear Res 22: 245–248, 1986.[CrossRef][Web of Science][Medline]
Simmons DD, Manson-Gieseke L, Hendrix TW, Morris K, Williams SJ. Postnatal maturation of spiral ganglion neurons: a horseradish peroxidase study. Hear Res 55: 81–91, 1991.[CrossRef][Web of Science][Medline]
Snyder RL, Leake PA. Topography of spiral ganglion projections to cochlear nucleus during postnatal development in cats. J Comp Neurol 384: 293–311, 1997.[CrossRef][Web of Science][Medline]
Snyder RL, Leake PA, Hradek GT. Quantitative analysis of spiral ganglion projections to the cat cochlear nucleus. J Comp Neurol 379: 133–149, 1997.[CrossRef][Web of Science][Medline]
Sobkowicz HM, Rose JE, Scott GE, Slapnick SM. Ribbon synapses in the developing intact and cultured organ of corti in the mouse. J Neurosci 2: 942–957, 1982.[Abstract]
Sokolowski BHA, Cunningham AM. Patterns of synaptophysin expression during development of the inner ear in the chick. J Neurobiol 38: 46–64, 1999.[CrossRef][Web of Science][Medline]
Spoendlin H. Innervation patterns in the organ of Corti of the cat. Acta Otolaryngol (Stockh) 67: 239–254, 1969.[Medline]
Stellwagen D, Shatz CJ. An instructive role for retinal waves in the development of retinogeniculate connectivity. Neuron 31: 357–367, 2002.
Walsh BT, Miller JB, Gacek RR, Kiang NYS. Spontaneous activity in the eighth cranial nerve of the cat. Int J Neurosci 3: 221–236, 1972.
Walsh EJ, Mcgee J. The development of function in the auditory periphery. In: Neurobiology of Hearing: The Cochlea, edited by Altschuler R, Bobbin R, and Hoffman D. New York: Raven Press, 1986, p. 247–269.
Walsh EJ, Mcgee J. Does activity in the olivocochlear bundle affect development of the auditory periphery? In: Diversity in Auditory Mechanics, edited by Lewis ER, Long GR, Lyon RF, Narins PM, Steele CR, and Hecht-Poinar E. Singapore: World Scientific, 1997, p. 376–385.
Walsh EJ, McGee JA. Postnatal development of auditory nerve and cochlear nucleus neuronal responses in kittens. Hear Res 28: 97–116, 1987.[CrossRef][Web of Science][Medline]
Walsh EJ, McGee JA. Rhythmic discharge properties of caudal cochlear nucleus neurons during postnatal development in cats. Hear Res 36: 233–248, 1988.[CrossRef][Web of Science][Medline]
Walsh EJ, Romand R. Functional development of the cochlea and the cochlear nerve. In: Development of Auditory and Vestibular Systems 2, edited by Romand R. New York: Elsevier Science, 1992, p. 161–219.
Yates PA, Holub AD, McLaughlin T, Sejnowski TJ, O'Leary DD. Computational modeling of retinotopic map development to define contributions of EphA-ephrinA gradients, axon-axon interactions, and patterned activity. J Neurobiol 59: 95–113, 2004.[CrossRef][Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  I. Milenkovic, I. Rinke, M. Witte, B. Dietz, and R. Rubsamen P2 Receptor-Mediated Signaling in Spherical Bushy Cells of the Mammalian Cochlear Nucleus J Neurophysiol, September 1, 2009; 102(3): 1821 - 1833. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sonntag, B. Englitz, C. Kopp-Scheinpflug, and R. Rubsamen Early Postnatal Development of Spontaneous and Acoustically Evoked Discharge Activity of Principal Cells of the Medial Nucleus of the Trapezoid Body: An In Vivo Study in Mice J. Neurosci., July 29, 2009; 29(30): 9510 - 9520. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Eatock, J. Xue, and R. Kalluri Ion channels in mammalian vestibular afferents may set regularity of firing J. Exp. Biol., June 1, 2008; 211(11): 1764 - 1774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Beurg, S. Safieddine, I. Roux, Y. Bouleau, C. Petit, and D. Dulon Calcium- and Otoferlin-Dependent Exocytosis by Immature Outer Hair Cells J. Neurosci., February 20, 2008; 28(8): 1798 - 1803. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Koppl Spontaneous Generation in Early Sensory Development. Focus on "Spontaneous Discharge Patterns in Cochlear Spiral Ganglion Cells Before the Onset of Hearing in Cats" J Neurophysiol, October 1, 2007; 98(4): 1843 - 1844. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
