| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Zoological Institute, University of Frankfurt, 60323 Frankfurt am Main, Germany 2 Institute of Biochemistry and Biology, University of Potsdam, 14471 Potsdam, Germany 3 Department of Animal Physiology, University of Habana, Vedado, La Habana 10400, Cuba 4 School of Biology, University of Sussex, Falmer, Brighton BN19QG, United Kingdom
Submitted 3 February 2003; accepted in final form 16 June 2003
 |
ABSTRACT |
|---|
|
61 kHz This resonance is present immediately after birth in bats that do not yet echolocate. Its frequency is lower (46 kHz) and the corresponding threshold minimum of cochlear microphonic potentials is broader than in adults. Long-lasting ringing of the cochlear microphonic potential after tone stimulus offset that characterizes the adult auditory response close to CF2 is absent in newborns. In the course of the first 5 postnatal weeks, there is a concomitant upward shift of CF2 and the frequency of cochlear threshold minima. Up to the end of the third postnatal week, sensitivity of auditory threshold minima and the Q value of the cochlear resonance increase at a fast rate. Between 2 and 4 wk of age, two cochlear microphonic threshold minima are found consistently in the CF2 range that differ in their level-dependent dynamic growth behavior and are 1.5–5.7 kHz apart from each other. In older animals, there is a single minimum that approaches adult tuning in its sharpness. The data provide evidence to show that during maturation of the cochlea, the frequency and the sensitivity of the threshold minimum associated with CF2 increases and that these increases are associated with the fusion of two resonances that are partly dissociated in developing animals. |
|
INTRODUCTION |
|---|
|
To process CF echolocation calls, the mustached bat employs mechanical specializations of the cochlea that lead to extremely sharp frequency tuning to the dominant second harmonic call frequency (CF2) close to 61 kHz (e.g., Henson et al. 1982
; Pollak et al. 1972; Russell and Kössl 1999; Suga and Jen 1977). Stimulation with tone frequencies close to 61 kHz evokes a strong ringing in the cochlear microphonic (CM) potential, which persists for many milliseconds after the tone offset. This specialized response indicates the presence of a strong cochlear resonance, which is involved in enhancing cochlear frequency tuning (Suga et al. 1975) and in producing pronounced otoacoustic emissions (Henson et al. 1985; Kössl and Vater 1985a). At the exact frequency of the ringing and hence of cochlear resonance, there is a pronounced minimum in CM thresholds. In addition the CF2 frequency range 61 kHz is vastly over-represented in the basal turn of the cochlea (Kössl and Vater 1985b). In the range of this so-called auditory fovea, there are multiple morphological specializations, which are likely to be involved in sharp tuning to 61 kHz. The fovea can be divided into an apical region that is densely innervated by auditory nerve fibers and a more basal, sparsely innervated (SI) region (Henson et al. 1977), where innervation density is low. Most remarkably, the tectorial membrane, the basilar membrane as well as their anchoring structures change their thickness and shape within a few hundred micrometers at the boundary between the apical CF2 region and the SI region (Henson and Henson 1991; Henson et al. 1977; Vater and Kössl 1996). There is evidence from anatomical and physiological data that the SI region contains most of the morphological adaptations that are required for cochlear resonance (Henson and Henson 1991; Kössl and Vater 1996; Vater and Kössl 1996). Accordingly, resonant oscillations at 61 kHz seem to be generated in the basal part of the fovea and then lead to sharp tuning in the more apical part of the fovea where auditory nerve fibers tuned to 61 kHz have their highest density.
The only other mammals with a cochlea of comparable frequency resolving power are the old world CF-bats, the horseshoe bats and hipposiderids. They too have an acoustic fovea tuned to the second harmonic CF frequency of the call, which is associated with frequency tuning as sharp as that of the mustached bat. However, ringing in cochlear responses, including the CM and otoacoustic emissions (OAEs), is considerably reduced or even absent (Henson et al. 1985; Kössl 1994). This indicates that the cochleae of horseshoe bats are more strongly damped than those of mustached bats. During postnatal development of the auditory and audio-vocal systems in the horsehoe bat Rhinolophus rouxi, there is a concomitant upward shift of both the CF frequency of the calls and of the preferred tuning of the neurons in the auditory midbrain that process the foveal frequency range in the adult. In R. rouxi, the CF2 frequency shifts from 60 to 78 kHz within a period of 3 wk, starting at an age of 2 wk. Within the same period, sharply tuned neuronal responses appear in the inferior colliculus correlated with a CF2 frequency of 63 kHz and finally reaching the 78 kHz of adults (Rübsamen 1987; Rübsamen and Schäfer 1990). During the first weeks of postnatal development of the mustached bat, there is a continuous upward shift of the CF2 component from 48 to 61 kHz. In addition, there are profound changes in energy distribution between the different harmonic components of the call (Vater et al. 2003). To gain insight into the developmental emergence of the remarkable resonance of the mustached bat's cochlea and the underlying mechanisms that are responsible for it, we investigated the cochlear properties in young mustached bats in the first postnatal weeks.
|
|
METHODS |
|---|
|
4–5 wk of postnatal development. Newborn bats have a forearm length of 20 mm. At an approximate growth rate of 1 mm/day, the adult forearm length (FAL) of 51–52 mm is attained after 4–5 wk of development (see Vater et al. 2003). Comparable growth rates are reported for R. rouxi (1 mm/day) (Rübsamen 1987) and Myotis lucifugus (0.7–1.4 mm/day) (Kunz and Anthony 1982). In the caves, the bats live in a very hot and humid environment (38°C and 99% humidity), which in this respect is comparable to their mother's womb. The bats are hairless for 
4 wk, and they start to fly around in the cave at an age that corresponds to 46–48 mm forearm length. As described by Silva Taboada (1979) subadult bats start to hunt close to the cave entrance, but their insect diet is at this time still complemented by mother's milk. Young bats were caught in the hot cave and kept for 1–3 days in a metal cage (dimensions: 30 x 20 x 17 cm) before measurement of cochlear potentials. The metal cage was positioned in a plastic container, and the humidity was kept high. During this period of captivity, care was taken that the bats' environment was at a temperature close to 38°C. The bats were hanging together, and therefore they were able to hear the vocalizations of other bats. This situation is comparable to that in the maternity colony in the cave but without exposure to the vocalizations of adult bats.
To record CM potentials, which are interpreted as a measure of receptor currents that are due largely to the activity of outer hair cells, and the compound action potential (CAP) of the auditory nerve, young bats were anesthetized with xylazinhydrochloride (0.01–0.06 ml of 0.2% solution per 10 g body weight) or with a xylazinhydrochloride and Ketavet mixture (100 mg/ml ketamine-HCl, 2% Rompun, water in the proportion 9:1:190; dose 0.005–0.02 ml/10 g body wt). The skull of the bats was fixed by dental acrylic to a metal bar, and an insulated tungsten electrode was introduced through the cerebellum into the cochlear aqueduct (Henson and Pollak 1972). During the penetration, pure tone sound stimuli were presented and the CM response measured. Typically, after reaching the aqueduct, there was a sudden increase in CM magnitude that did not change substantially when the electrode was advanced further. Having reached this position, we stopped the penetration and started the data acquisition. During the recordings, the body temperature was maintained with an infrared lamp. All experimental animals were anesthetized, and they were killed with an overdose of the anesthetic at the end of the experiment.
The animal use in this study was authorized by the Centre for the Inspection and Control of the Environment, Ministry of Science, Technology and Environment, Cuba. The experiments comply with the "Principles of Animal Care," publication 86-23, revised 1985, of the National Institutes of Health and also with the Declaration of Helsinki.
Pure tone sound stimuli of 10-ms duration with 1-ms cosine rise and fall envelopes were used for acoustic stimulation. The stimuli were generated by a Microstar 3000a/212 or a Microstar 4200a data-acquisition board and subsequently attenuated by a computer-controlled attenuator (custom made by J. Hartley, University of Sussex). Either MicroTech Gefell 1-in MK102.1 microphone capsules or Polaroid type 6500 sonar speaker modules were used to generate sound stimuli. In the frequency range between 30 and 70 kHz, where the resonant phenomena occurred, the frequency response of the speakers was flat within ±4 dB. Constant sound pressure levels (SPL in dB re 20 µPa) for different frequencies were guaranteed either by the use of on-line calculation of loudspeaker calibration curves or the SPLs were calculated off-line from the calibration curves. For each data acquisition, 80 responses were averaged. The responses were fed into the Microstar board, and programs written in Testpoint (CEC) were used to control data acquisition. In some cases, the application of loud sound pressure levels >80 dB SPL caused attenuation of the CM response magnitude with a latency of 2–6 ms to the CM response onset, which is typical for the effects of the middle ear muscle reflex. Data showing these characteristics were rejected.
Usually, the resonance frequency of the cochlea was defined as that stimulus frequency where CM ringing lasted more than one millisecond after stimulus offset. In some cases, where there was no significant ringing, the resonance frequency was determined as the stimulus frequency where abrupt CM ON,OFF amplitude changes occurred (see RESULTS).
|
|
RESULTS |
|---|
|
4–5 wk (see Vater et al. 2003). In addition, we measured cochlear potentials from three adult animals with a forearm length of 51 mm. Echolocation calls were recorded on the day that we did the electrophysiological measurements. In the two youngest animals, we were not able to measure calls; in the other animals, we could obtain calls emitted either spontaneously or in response to being positioned on a pillow that was gently moved up and down (see Vater et al. 2003).
Examples of CM and CAP responses during stimulation with tone bursts of 70 dB SPL with the frequency increasing from 55 to 59 kHz in steps of 100 Hz are shown in Fig. 1. The animal was a female bat with a FAL of 40 mm. The custommade recording amplifier was set to a high-pass filter characteristic with a corner frequency of 50 Hz and with a flat throughput response up to 100 kHz. The sampled data were subsequently band-pass filtered with software routines involving inverse fast Fourier transform (FFT) algorithms. To focus on CM responses (Fig. 1, A and C), a software band-pass of 50–70 kHz was applied. To show the CAP response (Fig. 1, B and D), the filter was set to 50–2,000 Hz. A pronounced maximum in CM amplitude appeared at 56.7 kHz (Fig. 1A, presentation 18). In Fig. 2, CM responses from an immature bat are compared with those of an adult bat for two different stimulus levels. For medium and high sound levels, at the frequency of maximum CM amplitude, a long-lasting ringing after cessation of the tone burst is indicative of strong cochlear resonance (Fig. 2). In the adult bat (Fig. 2A), at both stimulus levels, there is beating observable in the response during the ongoing stimulus. In the subadult bat (Fig. 2B), an amplitude maximum and resonant ringing during stimulus presentation can be observed although these are less pronounced than in the adult. Both features, the amplitude maximum and the ringing, were used to define the resonance frequency of the cochlea. In immature and mature bats, the cochlear resonance shifted to lower frequencies when the stimulus level was increased above 70 dB SPL. Therefore we used the lowest possible levels where the response was large enough to define the frequency of cochlear resonance. In addition to resonant ringing that is characterized by a smooth logarithmical decay (CM-after) (Suga and Jen 1977), at frequencies just above and/or below the resonance frequency, there are pronounced amplitude peaks at the onset and offset (Figs. 2 and 1, A and C, e.g., presentation 27) of the tone stimulus (CM ON, CM OFF) (Suga and Jen 1977). Characteristically, below and above the resonance frequency, the responses during sound stimulation were smaller but they showed ON and/or OFF responses (Fig. 2). ON responses were most apparent for frequencies lower than the resonance. OFF responses were most pronounced for frequencies above the resonance. In addition, at higher levels of tone stimulation, there is a frequency band above the resonance where the CM amplitude reaches a sharp minimum (Fig. 1A: between presentations 27 and 35; Fig. 2A, 80 dB SPL, upper traces with CM OFF; Fig. 2B: 93 dB SPL: upper traces with CM OFF). At these frequencies, the responses are often larger at lower sound pressure levels giving evidence for extremely nonlinear behavior of the cochlea or for level-dependent cancellation phenomena. It should be noted that in very immature bats, where the resonance was not yet strong enough to produce a significant amplitude maximum and ringing after the tone offset at moderate sound levels, CM ON/OFF phenomena could be seen at the highest sound pressure levels and were taken as evidence for the presence of resonance (Fig. 3).
|
|
|
By low-pass filtering the tone-evoked, electrical response from the cochlea, it is possible to observe both a tonic positive-going summating potential (SP), and the CAP, which is visible as the first of a series of amplitude maxima and minima close to the onset of the response. The CAP has a slightly longer latency than the onset of the SP due to synaptic delay. As a consequence of the 50-Hz high-pass filtering of the amplifier, the SP appears to have an adaptation-like shift during the period of the tone-burst. When evaluating the CAP, care was taken to use the first down and up going peak and not the SP onset as the cochlear response. As evident from the CAP example in Fig. 1 and from the threshold data (see following text), the highest CAP amplitudes and lowest CAP thresholds occurred not at the resonance frequency where the CM is largest but at a slightly higher frequency. This relationship between the cochlear resonance frequency and the frequency of the CAP threshold minimum is also found in adults (Fig. 4) (Suga and Jen 1977; Suga et al. 1975).
|
Threshold curves
To determine threshold curves of cochlear potentials, the level of 10-ms tone bursts was increased in 5-dB steps. At each level, the responses to 80 stimulus presentations were averaged. The CAP threshold was determined as the lowest stimulus level that evoked a minimum/minimum peak. The CAP peak at threshold was determined visually and had a magnitude of 10–30 µV. The CM threshold was determined using a software routine. An FFT analysis of the first 11.5 ms of the recorded data traces, a span that includes the stimulus response, yielded spectral peaks that grew with increasing sound level.
From these FFT-based CM growth functions, the sound level sufficient to produce a CM response of 2 µV (threshold criterion) was interpolated.
At or a few hundred hertz above the dominant echolocation frequency (CF2) in the 60-kHz range of adult mustached bats, there is a very sharp CM threshold minimum. The frequency of this minimum coincides exactly with extended ringing at stimulus offset and with maximum CM amplitudes (Fig. 4, A and B). in mature bats, the slope of CM growth around the resonance frequency is rather steep with values close to 1 dB/dB over the first 20 dB of growth of the response. The CAP displayed a threshold minimum (tip) that was located between a few hundred hertz and 2 kHz above the CM minimum.
Figure 5A displays threshold data for the youngest animal investigated. This bat, which still had its umbilical cord attached, had a forearm length of 21.5 mm and was probably 1–2 days old. Both the CM and the CAP thresholds were clearly elevated in comparison to those measured in the mature bat. Unlike adults and older juveniles, the CAP threshold was above the CM threshold curve for most frequencies and rather insensitive. This individual did not produce echolocation calls. Nonetheless, there is a sequence of a broad local CM threshold minimum followed by a relatively shallow CAP minimum in the range of 40–50 kHz. The range of this CM threshold minimum coincides with CM OFF responses found for stimulus frequencies between 45.6 and 46.3 kHz. (Fig. 3); this indicates the presence of a resonant cochlear processing stage. In a slightly older individual (Fig. 5B: FAL 24 mm), the CAP-threshold was on average more sensitive than the CM threshold. Most significantly, at this age both the CM and the CAP threshold minima were already clearly visible and at, or below, the frequency of the CM minimum there were resonant off responses at 46 and 48 kHz. The slope of growth of the CM amplitude decreased from 0.9 dB/dB at 46 kHz to 0.42 at 48 kHz.
|
In animals that had forearm lengths between 28 and 34 mm, which is reached in the second postnatal week, there is a general threshold improvement, particularly at low frequencies (Fig. 5, C and D). The CAP threshold curve is more sensitive than the CM threshold at frequencies between 20 and 45 kHz and the frequency range of resonance shifts toward 54 kHz. The shapes of the threshold curves, however, are rather variable in terms of the depth and sharpness of both the CAP and CM minimum. The range of the CM minimum coincides with the CF2 frequency of the calls.
In the third and fourth postnatal week (forearm lengths of 34–48 mm), there is a further slight threshold improvement of CAP and CM (Fig. 6). The CM minimum either extends over a broad frequency range (Fig. 6A, PP12, 3rd week) or there are two distinct threshold minima (Fig. 6, B–D), which are separated by 1.5 up to 5.7 kHz. In PP16 (FAL 40 mm, Fig. 6B), the frequency separation is 4 kHz; in the larger animals shown (Fig. 6, C and D), the minima are separated by 3.6 and 3.2 kHz, respectively. The frequency of longest ringing, and hence of cochlear resonance is associated with the lower frequency threshold minimum, which is therefore defined as the primary threshold minimum. The secondary higher frequency minimum is close to the CAP threshold minimum. In the same frequency region, CM OFF responses that occur at higher stimulus levels are most pronounced (Figs. 1 and 2). In some of the animals with broadened or two peaked threshold minima, the CF2 frequency of the calls varies over a frequency range of >1 kHz (Fig. 6, A–C).
|
For age stages >40 mm FAL, pronounced off responses of the CAP potential could be measured in most animals. The range of CAP OFF responses coincides with the range of resonance, and CAP OFF responses have lowest thresholds at the resonance frequency as it is reported for adults (Suga and Jen 1977). In near mature animals with forearm lengths close to 50 mm, the double CM threshold minima are reduced or absent (Fig. 7). The threshold curves are similar to those of mature bats and the frequencies of the CF2 component of the call and the CM and CAP threshold minima are in the range of mature bats.
|
The age-dependent upward shift of the frequencies of CM and CAP threshold minima is illustrated in Fig. 8. Often, the CAP threshold reaches a distinct maximum close to the frequency of the CM threshold minimum. In bats that are >1 wk, the CAP threshold curve is more sensitive than the CM threshold for frequencies below the CAP threshold maximum. In most cases, the CAP threshold is also more sensitive than the CM threshold for frequencies well above the CAP threshold maximum and the CM resonance. In bats with FALs of 37–50 mm, the CAP threshold maximum coincides with the CM resonance and the primary or only CM threshold minimum. In the two adults, the CAP threshold maximum is slightly below the CM threshold minimum. The double-peaked CM threshold minima are most pronounced in an intermediate age group centered around 40-mm forearm length. The shape and depth of the CF2-related minima of the CAP and the CM threshold curve change with age. A continuous decrease of the minimum threshold values of both CAP and CM can be observed up to an age that corresponds to a forearm length of 40 mm (Fig. 9). For older animals, both threshold minima seem to have reached a steady state, the CAP threshold varies between 25 and 41 dB SPL and, at its most sensitive, is close in sensitivity to the 19- and 24-dB SPL CAP thresholds of the two mature bats that we have used for comparison. The CM thresholds of bats with forearm lengths 40 mm ranges between 13 and 33 dB SPL, and the most sensitive immature CM threshold minima are similar to those of the mature bats (13 and 15 dB SPL).
|
|
Resonant CM potentials
We assessed the strength of the resonance in animals of different ages by comparing the iso-level CM responses elicited during the ongoing sound stimulation with those measured after the offset of the stimulus. The later response is correlated with the strength of ringing of the CM. In Fig. 10, the frequency dependence of CM magnitude for different stimulus levels is displayed during and after sound stimulation for seven young bats and one adult bat. The magnitude was calculated from FFT amplitude maxima at the stimulus frequency. A time window of 11.5 ms, which covered the duration of the response to the ongoing stimulus, was used to obtain FFT calculations for the CM during sound stimulation. A similar time window starting at 11.5 ms was used to assess CM ringing after the offset. The limit value of 11.5 ms includes acoustic delays and was defined after analysis of CM responses elicited by tone frequencies much higher or lower than the resonance frequency. In such responses, there was no FFT magnitude maximum at the stimulus frequency for FFTs that started at 11.5 ms or later.
|
In the youngest bat (Fig. 10A; 21.5 mm FAL), the CM reaches a broad magnitude maximum in the region of 46 kHz at moderate sound levels between 55 and 75 dB SPL. At 48.5 kHz, there is a pronounced notch. Similar level-dependent notches are dominant features in iso-level CM responses of bats with forearm lengths of 27 and 30 mm (Fig. 10, B and C). In these very young bats, CM activity after stimulus offset is brief (<3 ms), very small in magnitude, and restricted to the highest sound levels. The CM maxima during sound stimulation and the response after offset are both more pronounced in bats that have reached 2.5 weeks of age (FAL: 37 mm). In the bats with 37 and 45 mm FAL (Fig. 10, D and E), a broadening of the CM peak or a double peak can be observed. These same bats had CM threshold minima that were either broad or had two tips (see preceding text). The CM magnitudes of the oldest immature bat are comparable to those of the adult. In general, the magnitude data of the oldest immature bat are slightly more sensitive than in younger bats, i.e., magnitude maxima of comparable level can be elicited by lower sound pressure levels of the stimulus. In the adult, during sound stimulation with 15 dB SPL, there is already a small CM maximum that exceeds the measurement noise floor (Fig. 10H, top). At frequencies just above this maximum a notch or minimum is evident.
An example of a double-peaked CM magnitude maximum is shown Fig. 11A. In the same animal, a pronounced double-tipped CM threshold minimum can be measured (Fig. 11B) with long-lasting ringing associated with the primary threshold minimum at 59.59 kHz. At this frequency, the slope of growth of the CM potentials is close to 1 (Fig. 11C) and is clearly steeper than at the frequency of the secondary CM threshold minimum at 61.19 kHz (slope of 0.3). Comparable differences in the slope of CM growth at the primary resonance and slightly above were often observed in immature animals.
|
The Q value of a resonant oscillation was used to provide a measure of the sharpness of tuning. The Q values of the resonance were calculated from the duration of resonant ringing. For this purpose, the envelope of the CM afterresponse was determined by software routines that connected and smoothed the maximum values of the positive peaks of the sinusoid oscillations. From these envelopes, the decay time that corresponds to a decrease of the signal to 36% of its original value was calculated and the inverse taken as the Q value of the resonance. The Q values of the resonance increase steeply up to an age that corresponds to FAL values of 35–40 mm (Fig. 12). In older juvenile bats, the rate of growth of the Q value is slightly smaller. At their most advanced stage of development, preadult bats have Q values close to 300, which are within the lower boundaries of Q values found in adults. In one adult, the resonance was close to instability, and the ringing decayed over a very long time course that resulted in extremely high Q values >1,000.
|
The frequencies of the auditory threshold minima, and the CF2 component of the call frequency are plotted as a function of FAL (Fig. 13). The frequencies of the primary and secondary (higher frequency, if present) CM threshold minima, and the CAP minimum, and the CF2 component of the calls from immature bats of different age groups (FAL), are more loosely correlated and more variable than those recorded from mature bats. In the two adults, the frequency of the CF2 component of the call is below and within 390–500 Hz of the frequency of the CM threshold minimum. This is within the range found in adult populations from Jamaica (e.g., Kössl and Vater 1985a). However, in immature bats, even those that are close to maturity, the frequencies of the calls and CM threshold minima vary over a much larger frequency range. Often the CF2 frequency of the young animals is higher than the frequency of the primary CM threshold minimum and the resonance frequency. In some of the young animals, the CF2 frequency seems to be associated with, or adjusted to, the secondary CM threshold minimum or CAP minimum. There were two immature bats where the CF2 requency was much higher than in normal adults. In one of those indviduals with a CF2 frequency of 64 kHz, we measured CM potentials and found that the cochlear resonance frequency was at 58 kHz. However, when sound stimuli of 64 kHz were applied, the CM potential in response to these tones had a strong 58-kHz component (Fig. 14). It appears that the 64-kHz stimuli were well suited to elicit 58-kHz resonant oscillations and, when echolocating, the higher frequency call caused the ear of this bat to ring.
|
|
|
|
DISCUSSION |
|---|
|
) and the frequency of the CM and CAP threshold minima. At the age of 2–4 wk, we find consistently two CM threshold minima close to each other with different level-dependent dynamic growth behavior and with CM ringing occurring at the frequency of the primary lower frequency threshold minimum. This could indicate the presence of two cochlear resonators or resonance modes with separate characteristic frequencies at this age. In older animals, there is only one CM threshold minimum that, in its sharpness, approaches adult tuning. We interpret the data to show that development of the cochlea involves a progressive increase in frequency, strength, sensitivity, and bringing together of two resonances that remain partially dissociated in the developing animals.
In carrying out this study, it was not technically possible to register the moment of birth of the young bats that were the subject of our observations and experiments, and thus we judged their postnatal age on the basis of FAL. This measure has proven to be a reliable indicator of age in other bat species (Kunz and Anthony 1982; Rübsamen 1987), and for this reason, it was adopted here.
According to earlier studies of the mustached bat's cochlea (Kössl and Vater 1985a), high levels of anesthesia can cause a downward shift in the frequency of the cochlear resonance. Therefore it is possible that the frequency of the resonance will be shifted below that of the calls recorded in unanasthetized bats. This might, in part, explain the large divergence between the frequencies of the calls that were higher than the cochlear resonances measured in the young bats. However, this cannot be the whole explanation because the resonance frequencies of the two adult bats, which were treated similarly in terms of anesthesia to the young bats, were very close to their echolocation call frequencies as in nonanesthetized animals. Either the effects of anesthesia differ between young versus adult bats or the differences between calls and resonance in the young bats represent a true discrepancy that disappears when the cochlea and the audio-vocal system are fully developed. Another possible systematic error could be produced by the way in which calls were elicited and collected. Some of the young bats only produced calls when they were passively displaced up and down. However, under these circumstances, the calls are usually emitted at frequencies lower than those emitted spontaneously (Vater et al. 2003). Therefore it is unlikely that the method used to elicit the calls is responsible for the observed differences between the frequencies of the resonance and call in the young bats.
Developmental changes in auditory sensitivity
It is possible to measure both CM and CAP from the ears of newborn bats, although neither response is very sensitive to sound (Fig. 5). Sound stimulation evokes receptor and neural potentials from the cochlea, and newborn bats occasionally respond to sounds by moving their head and/or ears. However, they do not echolocate and hence do not use their ears for active orientation. In the very youngest bats, the CM threshold curves are more sensitive than the CAP threshold curves. Later in development, this difference in sensitivity reverses so that in older bats the CAP threshold curves are more sensitive than those of the CM. At 4–5 wk of age, the general threshold characteristics of the bats are similar to those of mature animals. Thus in newborn bats, cochlear processes associated with sensory transduction appear to be more advanced in their development than are neural mechanisms associated with the synaptic transmission and/or conduction of sound evoked signals in the auditory nerve. In rodent development, hair cell transduction also precedes neural potentials. In cochlear preparations of 1–3 day postnatal rats (Glowatzki and Fuchs 2000), receptor potentials, but not hair-cell-evoked neural activity, can be recorded. Postsynaptic potentials can be recorded from the axon terminals of inner hair cell afferents by day 7 of development (Glowatzki and Fuchs 2002). In rats, however, these events occur before the onset of hearing around postnatal day 14, and therefore caution is required when comparing these data with auditory development in bats whose onset of hearing occurs much earlier.
Developmental changes in the cochlear resonance
In mustached bats, cochlear resonance is present at birth, although the corresponding threshold minima of the CM and CAP are not very sensitive and are relatively broadly tuned with frequencies that are substantially below that of the resonance in the adult cochlea. The establishment of resonant oscillations is therefore due to the inherent mechanical properties of the cochlea and does not require either auditory, or active audio-vocal experience for its establishment. However, such experience may well have a role to play in establishing and shaping the fine frequency tuning that is eventually manifested in the adult cochlea.
In accordance with frequency tuning of CM and CAP responses in the cochleae of adult mustached bats, resonant CM responses, even in the youngest bats, are tuned to frequencies slightly below the CAP threshold minimum. In adult bats, there is a direct correspondence between the CM threshold minimum and the neural threshold minimum based on single neuron responses, which are both precisely and narrowly tuned to the resonance frequency (Kössl and Vater 1995). The magnitude of CAP responses, however, depends on precise synchronization of onset action potentials in the auditory nerve, and therefore a CAP threshold minimum should not be expected exactly at the resonance frequency where due to slow build up of receptor potentials a poor spike synchronization has to be expected. This lack of precise timing has been implicated as the basis for the discrepancy between the tuning frequency of the resonance, as measured in the CM, and the peak of the CAP, which is shifted to frequencies several hundreds of hertz above the CM peak in adult bats (Suga et al. 1975) (see also Fig. 8).
This explanation is unlikely to account for the large discrepancies that exist between the threshold minima of the CM and CAP in developing bats where the tuning of the resonance is still broad and therefore onset timing in afferent fibers that innervate the CF2 region would be expected to be precise and not smeared by strongly resonant properties as in the cochlea of adults. A possible alternative explanation for the discrepancies between the threshold minima of the CM and CAP is that the recording electrode that is introduced through the cochlear aqueduct will preferentially pick up CAP potentials from the SI region that is located right at the aqueduct and not so much from the more distant CF2 region. In the basal part of the fovea that corresponds to the SI region, the auditory nerve fibers are tuned to frequencies slightly above the resonance frequency (Kössl and Vater 1985b). The cochlear resonance that is evident in the CM recordings, on the other hand, may prevail over any high-frequency CM response. In addition there is evidence that the resonance indeed originates in the SI region (Kössl and Vater 1995, 1996; Russell and Kössl 1999).
The CM recorded from bats at an intermediate stage of development often has two threshold minima with the primary minimum at the lower frequency coincident to CAP threshold maximum and the secondary upper minimum close to or coincident with the CAP threshold minimum. The dynamic growth behavior of the CM associated with each peak is quite different from each other (Fig. 11). This observation points to the possibility that two resonators and associated mechanisms are responsible for enhancing both frequency tuning and sensitivity in the mustached bat's cochlea. The primary lower frequency resonance is associated with the long CM ringing and with OAEs that have been recorded in adult mustached bats (Henson et al. 1985; Kössl and Vater 1985a) and is characterized by relatively steep CM growth. This primary resonance has been attributed to mechanical processing associated with passive properties of the TM (Henson and Henson 1991; Kössl and Vater 1996). Indeed, measured as OAE, this resonance is temporarily enhanced after administration of aminoglycoside antibiotics and the destruction of OHCs (Kössl and Vater 2000). The rather shallow CM growth curve at the second, higher frequency resonance, may be due to BM resonance that is amplified by feedback from the OHCs. However, the small amplitudes and the low values of CM growth may indicate that at this early stage of development, the OHCs may not be able to exert sufficient force to overcome the mass and stiffness loading of the BM, even at the resonance.
The upward shift of the cochlear resonance frequency in young mustached bats does not imply that different cochlear regions are producing the resonance at different age stages. Instead, the same cochlear region may be responsible for resonance at all age stages. This conclusion is supported by the fact that the massive local adaptations of cochlear anatomy that are associated with the resonance are present in neonates (Vater, unpublished data). The postnatal shift in resonance frequency of the mustached bat cochlea most likely reflects the ontogenetic upward shift in cochlear frequency representation as a basic principle of cochlear functional development (Echteler et al. 1989; Harris and Dallos 1984; Lippe and Rubel 1983; Rubel and Ryals 1983; Rübsamen and Lippe 1998). This all supports the conclusion that ontogenetic changes in the cellular and acellular structure of the organ of Corti contribute to a shift in the general micromechanical response of mammalian cochleas as well as in specific cochlear resonance in bats.
Comparison to cochlear development in other bat species
In young horseshoe bats (Vater and Rübsamen 1988), a developmental shift of the cochlear frequency place map results in a change of the frequency tuning of the first cochlear turn (fovea) from a broad and rather insensitive tuning to low frequencies in the youngest bat (8 days old) to sharply enhanced tuning to 78 kHz in adults. When the horsehoe bats reached an age of 2 wk and a CF2 frequency of 60 kHz, enhanced tuning first appears as measured in the properties of cochlear nucleus neurons (Vater and Rübsamen 1988). In the following weeks, a shift of the frequency of sharpest tuning from 60 to 78 kHz is matched by a shift of the CF2 frequencies. The range of this shift is about half of an octave and thus comparable to the shift of resonance in the mustached bat from 48 to 61 kHz.
As a measure of the sharpness of cochlear tuning, we used the sharpness of CM threshold minima and the Q values of CM ringing instead of the Q10 dB (tip frequency/bandwidth 10 dB above tip) values from neuronal tuning curves obtained in the cochlear nucleus or inferior colliculus (Rübsamen and Schäfer 1990; Vater and Rübsamen 1988). In the horseshoe bat, enhanced neuronal tuning is first recorded in 2-wk-old bats from cochlear nucleus neurons tuned to frequencies between 60 and 70 kHz. Similar enhanced tuning can also be recorded in the inferior colliculus when the bats are 3 wk of age. In mustached bats, sharp CM threshold minima, which are often double tipped, first appear at a comparable age of 2–3 wk (34–40 mm FAL). In the mustached bat during the first 3 wk of age, the Q values of the resonance increase rapidly with developmental age. By the end of the third week, the Q values increase more slowly with age, and the sensitivity at the tip of the CM threshold minimum is almost adultlike. The resonance frequency continues its upward shift until week 5 of development when it reaches adult values. This time course of development is similar to that for the upward shift of the acoustic fovea center frequency, which has been demonstrated in neuronal recordings in the horseshoe bat (Vater and Rübsamen 1988).
Postnatal maturation of cochlear structure
In Pteronotus parnellii as in other bats there is no postnatal growth of the inner ear. In addition, the distribution of auditory nerve fibers that innervate the organ of Corti along the length of the BM in the juvenile cochlea is similar to that seen in the adult cochlea (M. Vater, unpublished data). Thus changes in the physiology of the cochlea during development must be due to developmental changes at the cellular and molecular level rather than in the general morphology and organization of the cochlea. This would be in accordance with anatomical data from immature horseshoe bats where the general organization of cochlea and organ of Corti is adult-like at birth as well as the basic cytoarchitecture of IHCs and OHCs (Vater et al. 1997). In the first weeks of postnatal development of the horsehoe bat organ of Corti, the most pronounced morphological postnatal changes concern the cytoskeleton of supporting cells, in particular the distinct Deiter cell cups, which are, specialized structures holding the basal end of the OHCs. They are formed postnatally and associated stiffening within the organ of Corti may be linked to postnatal shifts of the frequencies of sharpest tuning in this bat species (Vater et al. 1997). In addition, changes in the fiber distribution within the extracellur matrices of the TM and the BM may also increase their stiffness and change the longitudinal interactions along the organ of Corti. It has been shown that between the second and fifth postnatal week actin-filaments are incorporated in tension fibroblasts of the spiral ligament of horseshoe bats (Henson and Rübsamen 1996). This may be a prerequisite for creating radial tension of the BM.
In bats it is still unclear if there are postnatal changes in OHC function. As pointed out in the preceding text, in horseshoe bats, the general morphology of OHCs and of IHCs does not change during postnatal development. Active OHC motility, however, should be critical for functional cochlear maturation and could also be an important factor for developmental changes in the resonance of the mustached bat's cochlea. Prestin, a motor protein of mammalian outer hair cells (review: Dallos and Fakler 2002), is present in OHCs from adult mustached bats (Vater et al. 2002). Its emergence during cochlear maturation still has to be investigated in bats. In gerbils, OHC motility was demonstrated one week after birth and 5 days before the onset of hearing (He et al. 1994) In the rat, nonlinear capacitance in postnatal OHCs as indicator of the presence of Prestin is not present at birth but slowly increases afterward and reaches a steady state at the onset of hearing at day 12 (Oliver and Fakler 1999).
As a consequence of structural maturation of the cochlea, the changing frequency filtering properties of the cochlea will shape a young bat's perception of its acoustic environment and its own echolocation calls. Accordingly, it might be expected that the maturation and refining of vocalization, and the frequency content of echolocation calls in particular, depends to a large degree on the maturation of the cochlea. This is consistent with the observation that Doppler-shift compensation behavior, whereby the bat adjusts the CF2 frequency of the call to the range of cochlear resonance and thus maximizes cochlear responses, first appears in animals with FALs >40 mm (Vater et al. companion paper). At this age stage, the first, rapid, phase of development of the cochlear resonance has already been completed.
|
|
DISCLOSURES |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: M. Kössl, Zoologisches Institut, Siesmayerstr. 70, 60323, Frankfurt/M., Germany (E-mail: koessl{at}zoology.uni-frankfurt.de).
|
|
REFERENCES |
|---|
|
Echteler SM, Arjmand E, and Dallos P. Developmental alterations in the frequency map of the mammalian cochlea. Nature 341: 147–149, 1989.[Medline]
Glowatzki E and Fuchs PA. Cholinergic synaptic inhibition of immature inner hair cekks in the neonatal mammalian cochlea. Science 288: 2366–1268, 2000.
Glowatzki E and Fuchs PA. Transmitter release at the hair cell ribbon synapse. Nat Neurosci 5: 147–154, 2002.[Web of Science][Medline]
Harris DM and Dallos P. Ontogenetic changes in frequency mapping of a mammalian ear. Science 225: 741–743, 1984.
He DZ, Evans BN, and Dallos P. First appearance and development of electromotility in neonatal gerbil outer hair cells. Hear Res 78: 77–90, 1994.[Web of Science][Medline]
Henson MM and Henson OW Jr. Specializations for sharp tuning in the mustached bat: the tectorial membrane and spiral limbus. Hear Res 56: 122–132, 1991.[Web of Science][Medline]
Henson MM, Henson OW Jr, and Goldman LJ. The perilymphatic spaces in the cochlea of the bat, Pteronotus p. parnellii (Gray). Anat Rec 187: 767, 1977.
Henson MM and Rübsamen R. The postnatal development of tension fibroblasts in the spiral ligament of the horseshoe bat, Rhinolophus rouxi. Aud Neurosci 2: 3–13, 1996.
Henson OW Jr and Pollak GD. A technique for chronic implantation of electrodes in the cochleae of bats. Physiol Behav 8: 1185–1188, 1972.[Medline]
Henson OW Jr, Pollak GD, Kobler JB, Henson MM, and Goldman LJ. Cochlear microphonic potentials elicited by biosonar signals in flying bats, Pteronotus parnellii. Hear Res 7: 127–147, 1982.[Web of Science][Medline]
Henson OW Jr, Schuller G and Vater M. A comparative study of the physiological properties of the inner ear in Doppler-effect compensating bats (Rhinolophus rouxi, Pteronotus parnellii). J Comp Physiol [A] 157: 587–597, 1985.[Medline]
Kössl M. Otoacoustic emissions from the cochlea of the "constant frequency" bats Pteronotus parnellii and Rhinolophus rouxi. Hear Res 72: 59–72, 1994.[Web of Science][Medline]
Kössl M and Vater M. Evoked acoustic emissions and cochlear microphonics in the mustache bat, Pteronotus parnellii. Hear Res 19: 157–170, 1985a.[Web of Science][Medline]
Kössl M and Vater M. The cochlear frequency map of the mustache bat, Pteronotus parnellii. J Comp Physiol [A] 157: 687–697, 1985b.[Medline]
Kössl M and Vater M. Cochlear structure and function in bats. In: Handbook of Auditory Research. Hearing By Bats, edited by Popper A and Fay R). New York: Springer, 1995, vol. 11, p. 191–234.
Kössl M and Vater M. Further studies on the mechanics of the cochlear partition in the mustached bat. II. A second cochlear frequency map derived from measurement of acoustic distortion products. Hear Res 94: 78–86, 1996.[Web of Science][Medline]
Kössl M and Vater M. Consequences of outer hair cell damage for otoacoustic emissions and audio-vocal feedback in the mustached bat. J Assoc Res Otolaryngol 01: 300–314, 2000.[Medline]
Kunz TH and Anthony ELP. Age estimation and postnatal growth in the bat Myotis lucifugus. J Mammal 63: 23–32, 1982.
Lippe W and Rubel EW. Development of the place principle: tonotopic organization. Science 219: 514–515, 1983.
Oliver D and Fakler B. Expression density and functional characteristics of the outer hair cell motor protein are regulated during postnatal development in the rat. J Physiol 519: 791–800, 1999.
Pollak GD, Henson OW Jr, and Novick A. Cochlear microphonic audiograms in the "pure tone" bat Chilonycteris parnellii parnellii. Science 176: 66–68, 1972.
Rubel EW and Ryals BM. Development of the place principle: acoustic trauma. Science 219: 512–514, 1983.
Rübsamen R. Ontogenesis of the echolocation system in the rufous horseshoe bat, Rhinolophus rouxi (Audition and vocalization in early postnatal development). J Comp Physiol [A] 161: 899–913, 1987.[Medline]
Rübsamen R, and Lippe W. Functional development of the cochlea. In: Handbook of Auditory Research: Development of the Mammalian Auditory System, edited by Rubel EW, Popper AN, and Fay RR. New York: Springer, 1998, p. 193–270.
Rübsamen R and Schäfer M. Ontogenesis of auditory fovea representation in the inferior colliculus of the Sri Lankan rufous horseshoe bat, Rhinolophus rouxi. J Comp Physiol [A] 167: 757–769, 1990.[Medline]
Russell IJ and Kössl M. Micromechanical responses to echolocation signals in the cochlear fovea of the mustached bat. J Neurophysiol 82: 676–686, 1999.
Silva Taboada G. Los Murcielagos de Cuba. La Habana, Cuba: Editorial Academia, 1979.
Suga N and Jen PHS. Further studies on the peripheral auditory system of CF-FM bats specialized for fine frequency analysis of Doppler-shifted echoes. J Exp Biol 69: 207–232, 1977.
Suga N, Simmons JA, and Jen PHS. Peripheral specializations for fine analysis of Doppler shifted echoes in the auditory system of the CF-FM bat Pteronotus parnellii. J Exp Biol 63: 161–192, 1975.
Vater M. Adaptation of the auditory periphery of bats for echolocation. In: Bat Biology and Conservation, edited by Kunz TH and Racey PA. Washington, DC: Smithonian Institution Press, 1998, p. 231–245.
Vater M and Kössl M. Further studies on the mechanics of the cochlear partition in the mustached bat. I. Ultrastructural observations on the tectorial membrane and hair cells. Hear Res 94: 63–77, 1996.[Web of Science][Medline]
Vater M, Kössl M, Foeller E, Coro F, Mora E, and Russell IJ. Development of echolocation calls in the mustached bat, Pteronotus parnellii. J Neurophysiol 90: 2274–2290, 2003.
Vater M, Lenoir M, and Pujol R. Development of the organ of Corti in horseshoe bats: scanning and transmission electron microscopy. J Comp Neurol 377: 520–534, 1997.[Web of Science][Medline]
Vater M and Rübsamen R. Postnatal development of the cochlea in horseshoe bats. In: Cochlear Mechanisms—Structure, Function, and Models, edited by Wilson JP and Kemp DT. New York: Plenum, 1988, p. 217–223.
Vater M, Weber T, and Knipper M. Prestin-like immunoreactivity in outer hair cells of the mustached bat (Pteronotus parnellii). Acta Otolaryngol Belgica 56: 295, 2002.
This article has been cited by other articles:
|
|
 |
|
  I. J. Russell, M. Drexl, E. Foeller, M. Vater, and M. Kossl The Development of a Single Frequency Place in the Mammalian Cochlea: The Cochlear Resonance in the Mustached Bat Pteronotus parnellii J. Neurosci., November 26, 2003; 23(34): 10971 - 10981. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Vater, M. Kossl, E. Foeller, F. Coro, E. Mora, and I. J. Russell Development of Echolocation Calls in the Mustached Bat, Pteronotus parnellii J Neurophysiol, October 1, 2003; 90(4): 2274 - 2290. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
) and CAP (
) threshold curves in 2 adult bats. The CF2 frequency is given by 
. The frequency of the cochlear resonance, defined by maximum ringing and maximum CM magnitude during the stimulus presentation, is given by 
. Note that in both types of threshold curves, a narrow and steep minimum is evident either directly at the resonance frequency (CM) or slightly above (CAP). As CM threshold, the sound pressure level that was sufficient to produce a CM response of 2 µV (threshold criterion) was interpolated from CM amplitude growth functions. The CAP threshold curves were determined visually from the low-pass filtered waveform data (see 
the frequency of cochlear resonance (for further details, see text).
