In the chinchilla, systemic administration of the anti-neoplastic drug, carboplatin, damages inner hair cells (IHCs), leaving the outer hair cells (OHCs) morphologically intact (Wake et at, 1993; 1994), An animal model with total and selective IHC loss could be a particularly useful preparation if the OHCs are not functionally compromised by the drug treatment, The present experiments were undertaken to address this issue, Distortion product otoacoustic emissions (DPOAEs; 2f(1)-f(2), f(2)-f(1)) and compound action potentials (CAPs) of the auditory nerve were recorded at a number of stimulus frequencies from carboplatin-treated and control chinchillas, Cochleas were subsequently retrieved and processed for light microscopic evaluation of hair cell status, Results of the histologic evaluation confirmed previous reports of selective loss of IHCs in carboplatin-treated animals, DPOAEs at 2f(1)-f(2) were indistinguishable from control, even when IHC loss was essentially complete, whereas DPOAEs at f(2)-f(1) were often diminished in magnitude, CAP "thresholds" were elevated when the IHC loss exceeded roughly 50% and were eliminated with total IHC loss, Results suggest that OHC function remains essentially normal in the carboplatin-treated chinchilla with extensive IHC loss, and are consistent with the view that IHCs are not required to generate a normal 2f(1)-f(2) DPOAE, although they may play a role in the generation of f(2)-f(1).
Many species of bats, including the little brown bat (Myotis lucifugus), use echolocation to forage for flying insect prey. Insect wingbeat frequency is a salient cue for detection and recognition, and bats can discriminate insects with different wingbeat frequencies and patterns. To determine insect wingbeat frequency these bats might have to integrate such information across multiple, sequential echoes as the frequency-modulated (FM) biosonar pulses emitted by these bats, and thus the resulting echoes, are too short (e.g., 0.5-5 ms) to register the period of the wingbeat frequencies of most insects. Thus, neurophysiological recordings were undertaken to determine if single neurons in the auditory cortex (AC) of M. lucifugus could extract amplitude-modulations (AM) in acoustic signals that mimicked echoes from fluttering insect targets. For this, trains of simulated "echoes" were sinusoidally amplitude modulated (5-110 Hz) across sequential sound pulses. The modulated trains of pulses were presented at four different base repetition rates [25, 50, 100 and 200 pulses per second (pps)] encompassing the range of biosonar emission rates commonly emitted at different stages of target-directed flight and the neurons' response selectivities measured. We found that over half the neurons studied in the auditory cortex, at all repetition rates tested, showed a preference for a specific frequency of across-pulse AM that corresponds to the range of insect wingbeat frequencies that little brown bats encounter in nature. Importantly, the repetition rate of the signals was found to influence the ability of neurons to extract the across-pulse AMs. Specifically, neurons were increasingly selective for particular across-pulse AM frequencies at base repetition rates corresponding to phases of echolocation when bats identify their insect prey. These data are compared to results reported earlier for midbrain neurons and a mechanism through which neurons in the AC mediate ranging (via delay tuning) and discrimination (via AM coding) of their fluttering insect targets is discussed.
Afferent fibres in the auditory nerve of the barn owl were counted and their diameters measured, using a semi-automated image analysis of light-microscopical sections. Since fibres of both the basilar papilla and the lagenar macula run in the auditory nerve, but cannot be sharply distinguished, an independent evaluation of the lagenar fibres was obtained from sections near the apical end of the cochlear duct. Average numbers of 31,142 afferents from the basilar papilla and 1,342 lagenar fibres were found. Papillar axons were, on average, considerably larger than lagenar axons. Analysis of serial sections at regular intervals along the cochlea showed that less than 20% of all papillar afferents derive from regions of the basilar papilla corresponding to frequencies below 2 kHz. Above 2 kHz, about equal numbers of afferents were counted per octave. While this reflects an unusually heavy emphasis on high frequencies among birds, the afferent fibre supply is less focused on a narrow frequency band than the cochlear space map implies. Axon diameters increased systematically with frequency up to approximately 7 kHz and then decreased again towards the base of the papilla. This pattern would be suited to exaggerate latency differences between frequencies in the cochlear nucleus. However, larger axons could also be an adaptation to the increasing demands for temporal accuracy in phase locking at high frequencies. Myelination of papillar afferents was studied in ultrathin sections and found to be very uniform, with an average sheath thickness of 0.58 mu m, regardless of axon diameter.
Auditory nerve fibers in the bullfrog (Rana catesbeiana) are sensitive to shifts in the phase spectra of multi-harmonic stimuli. Fibers of both auditory organs, the amphibian papilla (AP) and basilar papilla (BP), exhibit changes in their vector strength of synchronization to the fundamental period (VS(f)1) and changes in the shapes of their fundamental period histograms with shifts in the phase spectrum of a multi-harmonic stimulus. In addition, AP fibers also exhibit changes in their average spike rates. In this study, we assess the contributions of several factors upon the phase sensitivity of auditory afferents: changes in the stimulus envelope, the stimulus fine-temporal structure and the magnitude of peripheral nonlinearities. The phase dependent changes in the VSf1 of all BP fibers and 35% of AP fibers are correlated with the changes in the peak amplitude of the stimulus waveform that occur with shifts in the phase spectrum of a signal. For all fibers with best excitatory frequencies (BEFs) > 600 Hz, changes in the shapes of the stimulus envelope and a fiber's period histograms were highly correlated. Thus, the phase sensitivity of BP, and some AP, fibers reflect their sensitivity to changes in the stimulus envelope. For AP fibers with BEFs < 600 Hz there was a relatively high correlation between changes in a fiber's period histogram and changes in the fine-temporal structure of the stimulus. In addition, many low frequency AP fibers exhibit changes in their VSf1 and average spike rate when the relative phase angle of a component-outside a fiber's tuning curve was varied from 0-360 degrees. This indicates that nonlinearities in the peripheral auditory system also contribute to the phase sensitivity of low frequency AP fibers.
The oto-and nephrotoxic effects of aminoglycoside antibiotics are well known, However, the exact molecular mechanisms of the toxicity are not fully understood, A recent observation that, in the kidney, the major aminoglycoside-receptor is megalin, explains nephrotoxicity, since megalin is an abundant membrane protein at the luminal surface of the epithelial cells of renal proximal tubules, We demonstrate that megalin mRNA and protein are distinctly expressed in the inner ear epithelia responsible for the homeostasis of inner ear fluids. These include stria vascularis and Reissner's membrane in the cochlea, dark cells in vestibular organs and the epithelium of the endolymphatic duct and sac. Megalin is not expressed in the cochlear or vestibular hair cells which, according to earlier studies, are the primary targets of the ototoxic action of aminoglycosides.
Voltage recorded in awake guinea pigs from an electrode implanted at the round window was evaluated using fast-Fourier transformation, Acoustical stimulation was performed with a loudspeaker mounted in an aluminium housing and coupled to the contralateral outer ear canal with a silastic tube, In the absence of acoustical stimulation, the major part of the power spectrum of the round-window activity was localized to the frequency range of 500-2000 Hz, and peaked around 1 kHz. This peak reflects the ensemble background activity (EBA) of the auditory nerve. During the presentation of the white noise (WN) bursts (duration 300 msec, 2.5 msec rise/fall times) to the contralateral ear, the EBA was reduced, The first detectable change was observed with the sound pressure level (SPL) of the WN as low as 20 dB; the maximal suppressive effect (up to 50% reduction of the original power spectrum) occurred with the WN at 50-60 dB SPL, The suppressive effect of contralateral white noise on the EBA was reversibly eliminated by a single injection of gentamicin at a dose of 150 mg/kg, However, the basal level of the EBA was not significantly changed after gentamicin injection, The changes of contralateral suppression of the EBA after gentamicin injection were similar to those observed with transient evoked otoacoustic emissions, distortion-product otoacoustic emissions or click-evoked compound action potential of the auditory nerve.
Temporal Modulation Transfer Functions (TMTFs) were obtained for tonal carriers with frequencies of 500, 1000, and 4000 Hz, The carrier was either on continuously or gated, In the continuous-carrier condition the duration of the modulation was 125 or 500 msec, and in a gated-carrier condition the entire stimulus duration was 125 or 500 msec, Data were collected from six listeners using an adaptive, two-alternative, forced-choice procedure, The lowest thresholds were obtained for modulation rates between 4 and 8 Hz in the continuous-carrier condition and between 16 and 64 Hz in the gated-carrier condition, Several different processing strategies were used to try to account for the shape of the tonal TMTFs, The threshold changes with carrier frequency in the tonal TMTFs at high rates of modulation were best accounted for by assuming that listeners can detect the sidebands which fall outside of the critical band centered at the carrier frequency, The low-and mid-rate regions of the tonal TMTF were best described in a qualitative manner by a process that accumulates information based on the number of cycles of envelope fluctuations, However, this process could not quantitatively explain the tonal TMTF data in the low-to mid-rate regions nor could it account for the differences between tonal and noise TMTFs.
Changes in the size and shape of the isolated tectorial membrane (TM) of the mouse were measured in response to changes in the pH of the surrounding artificial endolymph solution, Altering the pH from 7 to values from 5 to 11 caused changes of both the thickness and volume of the TM that were highly correlated; there was little change in surface area, Changes in thickness for pH between 6 and 9 were small: on the order of 1%, In basic solutions, swelling increased with pH: median values across experiments increased from <1% at pH 9 to 9% at pH 10 to 79% at pH 11, In mildly acidic solutions, thickness decreased to a local minimum of -1% near pH 6 which suggests that this is an isoelectric point of the TM, Thickness swelled in more acidic solutions, reaching a local maximum of 14% near pH 5.25, In solutions with very low pH (<4), the radial dimension and area of the TM decreased rapidly and monotonically while the thickness of the TM increased transiently and then decreased, Time courses for all of these responses were on the order of 1 to 4 minutes, These results are similar to results that have been reported for other gels, Although more complicated theories are possible, these results are consistent with a simple theory in which the pH of the bath alters the state of ionization of charge groups that are known to be present in TM macromolecules.
We describe the equilibrium behavior of an isotropic polyelectrolyte gel model that is intended to help us interpret chemical, electrical, mechanical and osmotic properties of the tectorial membrane (TM). The gel is homogeneous, isotropic, and contains water, fixed ionizable charge groups and mobile ions. The gel is in contact with an aqueous ionic solution (bath). At equilibrium, the gel characteristics are derived from the following physical principles: macroscopic electroneutrality, electrodiffusive equilibrium, osmotic equilibrium and Hooke's law. These physical principles lead to a pair of coupled algebraic equations that are solved analytically for certain special cases and numerically in general to yield the concentrations of all mobile ions, the concentration of fixed charge, the osmotic pressure difference between the gel and the bath, the volume of the gel, and the electric potential difference between the gel and the bath. The gel model indicates that fixed charges play a key role in such properties of the TM as the capacity to concentrate ions, the occurrence of a difference in electric potential between the TM and the bath solution, and the capacity to imbibe water and to swell. Using the gel model to fit measurements of the electric potential of the TM (Steel, 1983a), we estimate that the concentration of fixed charge in the TM at neutral pH is in the range -6.4 to -8.4 mmol/L. Using the measured biochemical composition of the TM (Thalmann et al., 1993), we estimate that the fixed charge concentration of the TM, due to its glycosaminoglycan constituent is -18 mmol/L.
Projections of octopus cells, which are located within the posteroventral cochlear nucleus, to globular cells within the contralateral ventral nucleus of the lateral lemniscus were studied in cat by means of anterograde filling of axons, by immunostaining cells, axons and nerve endings for calretinin, and by single unit electrophysiological recordings within the lateral lemniscus. Injections of biocytin into cut axons of the intermediate acoustic stria resulted in labeling of octopus cells ipsilateral to the injection and of large, complex nerve terminals (endbulbs) within the ventral division of the contralateral ventral nucleus of the lateral lemniscus. Immunostaining for calretinin showed that octopus cells, their axons and endbulbs within the lateral lemniscus were positive. Single unit recordings within the portion of the lemniscal nucleus that contains endbulbs showed the characteristic electrical signature of cells that receive endbulbs (prepotentials) was associated with cells whose activity in response to acoustic stimulation indicated that their inputs were from octopus cells. Immunostaining of human tissue indicated that octopus cells, their axons and complex endings within the ventral nucleus of the lateral lemniscus were positive for calretinin, which suggests that the projection of octopus cells found in cat is also present in human. Counts of globular cells, the probable recipients of endbulbs, in the ventral nucleus of the lateral lemniscus in human tissue indicated that these cells comprise 38% of the nucleus, as contrasted with 4% in cat. The results suggest that functions subserved by globular cells in this nucleus may play a dominant role in processing acoustic stimuli in the human, whereas cells specialized for processing high frequency sounds are more dominant in cat.
We examined the effect of temperature on information transmission in the auditory system of the frog, In particular, we determined how the signal-to-noise ratio (SNR) of the encoded stimulus at the level of the eighth cranial nerve changed with body temperature, Core temperature shifts were induced experimentally and the resulting changes in SNR were quantified, Single fiber recordings were made from the eighth nerve of 24 anesthetized adult northern leopard frogs, while the body temperature was shifted within a range of 10-29 degrees C, A Fast Fourier Transform (FFT) was computed for the spike train evoked by a pure tone to obtain a power spectrum; ten such spectra were averaged, Both signal strength and noise level were obtained directly from the power spectrum, The resulting SNR in dB was plotted against temperature and internal noise level, but neither plot exhibited a discrete maximum as would be expected in a system demonstrating stochastic resonance, However, we found that the SNR of stimulus encoding in both saccular fibers and in mid-frequency (caudal) amphibian papilla (AP) fibers increased with temperature, whereas the SNR from the majority of low-frequency (rostral) AP fibers did not. Linear regressions of SNR on noise level using the method of least squares were significant for 15 of 19 saccular fibers, 9 of 15 rostral AP fibers and 8 of 15 caudal AP fibers, Increasing the temperature resulted in nominal increases in internal noise intensities and these, in turn, were accompanied by increasing SNR, These results cannot easily be explained on the basis of linear filter theory.
In this paper a line of thought is outlined that can aid in understanding the fundamental relation between cochlear nonlinearity and frequency selectivity, The treatment is based on a very wide class of models of the cochlea, in which certain elements (for instance, outer hair cells) act to amplify the cochlear waves in a place- and frequency-specific way, and are also the sole sites of cochlear nonlinearity, In particular, it is shown why input-output cross-correlation functions are the most useful tools of investigation, The principal result of the analysis is the 'EQ-NL theorem', which states: to each nonlinear cochlear model-within the given class of models - there exists a linearized model of the same structure that has the same input-output cross-correlation function for wide - band noise stimuli. In the linearized model the wave-amplifying elements are linear and are found to operate with a reduced efficiency; the reduction is larger for stronger stimulus signals, On the one hand, this theorem provides for an interpretation of experimental results obtained in terms of input-output correlation functions for various levels of stimulation. On the other hand, application of the theorem might indicate ways of testing whether our current notions on nonlinearity and cochlear amplification are justified or not.
We previously described a model for loudness perception for people with cochlear hearing loss. However, that model is incompatible with our most recent and most satisfactory model of loudness for normal hearing. Here, we describe a loudness model that is applicable to both normal and impaired hearing. In contrast to our earlier model for impaired hearing, the new model correctly predicts: (1) that a sound at absolute threshold has a small but finite loudness; (2) that, for levels very close to the absolute threshold, the rate of growth of loudness is similar for normal ears and ears with cochlear hearing loss; (3) the relation between monaural and binaural threshold and loudness; (4) recent measures of equal-loudness contours. Like the earlier model, the new model can account for the loudness recruitment and reduced loudness summation that are typically associated with cochlear hearing loss.
Strial marginal cells (SMC) and vestibular dark cells (VDC) are known to secrete K(+) into endolymph. Slowly-activating, voltage-dependent K(+) channels (KCNQ1/KCNE1; IsK; min K) have been identified in the apical membrane of these cells. Several experimental maneuvers known to increase or decrease transepithelial K(+) secretion have been found in VDC to change the current through these channels in the same ways. In both SMC and VDC the kinetics of activation and deactivation resemble those of the I(sK) channel exogenously expressed in Xenopus oocytes and endogenous to heart myocytes. The present study sought evidence that this current is indeed carried by I(sK) channels and that this current is the basis for transepithelial K(+) secretion. Both on-cell macro-patch recordings of the apical membrane and perforated-patch whole-cell recordings were made on SMC from gerbil in order to measure macroscopic cell currents. The on-cell current was found to 1) be K(+)-selective, 2) have a cation permeability sequence of K(+) ~ Rb(+) > Cs(+) >> Li(+) = Na(+), 3) be activated with a time constant of 1764 ± 413 ms by voltage steps from 0 to +40 mV, 4) be deactivated with a time constant of 324 ± 57 ms by voltage steps from 0 to -40 mV and 5) be reduced 84 ± 5% by bumetanide (10(-5) M), an inhibitor of K(+) secretion. The single-channel conductance of the apical currents in the homologous VDC was estimated by fluctuation analysis to be 1.6 pS. The potent inhibitor of I(sK) channels, chromanol 293B (10(-5) M), reduced the whole-cell current in SMC by 72 ± 10 %. Clofilium (10(-4) M), a putative I(sK) channel inhibitor known to have additional non-specific effects, led to a stimulation of both on-cell (by 598 ± 177%) and whole-cell (by 162 ± 18%) currents in gerbil SMC but to a decrease of whole-cell currents (by 39 ± 12%) in rat SMC. Taken together with other findings reviewed here, these results strongly argue that the slowly-activating, voltage-dependent conductance in the apical membrane of SMC is the I(sK) channel and provide additional evidence for the poor specificity of clofilium.
We have previously shown that inhibitory afferents regulate dendrite size in the gerbil lateral superior olive (LSO) during the first 3 postnatal weeks. To determine how rapidly this effect occurs, we performed unilateral cochlear ablations at one week postnatal, and LSO neurons were filled with biocytin-containing whole cell electrodes in the acute brain slice preparation. The manipulation leads to excitatory denervation in the ipsilateral LSO, and inhibitory denervation in the contralateral LSO. Filled neurons were analyzed with a computer-based morphometric system at 1 to 6 days after the manipulation. For all contralateral neurons examined there was a significant increase in the total number of dendritic branches and the total dendritic length, as compared to control neurons. A time-course analysis of the change in branch points showed that inhibition exerted its effect with a latency of 2-3 days. Excitatory denervation led to a significant reduction in soma size, although dendrites neither shrank nor expanded. Since cochlea removal was performed 5 days before sound-evoked activity began, the hypertrophy of the contralateral LSO neurons indicates that trophic modulation may be attributed to spontaneous inhibitory synaptic transmission.
Enolase is a glycolytic enzyme active as a dimer. In the adult central nervous system of vertebrates, three isoforms alpha alpha, alpha gamma and gamma gamma have been described. During neuronal development there is a switch from the ubiquitous alpha alpha isoform to the specific gamma gamma, isoform. In the adult rat brain, although the alpha gamma hybrid is responsible for 30% of total enolase activity, immunocytochemical investigations have shown only few neurons containing both the alpha and gamma forms. Recent in situ hybridization studies have identified numerous neurons that coexpress the alpha and gamma genes. We report an immunocytochemical investigation of a enolase in the rat spiral ganglion which contains two populations of neurons, type I and type II. Type II neurons were also characterized by neurofilament immunostaining. Investigations were performed in the adult and between postnatal days (P) 5 and 15. Immunocytolabeling with anti-alpha enolase antibodies in the adult spiral ganglion revealed two neuronal subpopulations: a small number of alpha enolase-positive neurons and a large population of alpha enolase-negative neurons. The location, number, cell size and neurofilament immunoreactivity of the alpha enolase-positive neurons was consistent with their being type II neurons. These neurons represented 6.15% of the total population of neurons and their mean diameter (8.86 +/- 1.30 mu m) was significantly smaller (P < 0.001) than that of the non-reactive neurons (12.17 +/- 1.56 mu m). During the first postnatal week, alpha enolase immunostaining was present in all neurons and, between P12 and P15, disappeared from the majority of neurons except type II neurons. Thus alpha enolase is only present in adult type II neurons. The disappearance of alpha enolase immunoreactivity from type I neurons during development, between P12 and P18, is discussed.
Hearing begins 9-12 days after birth in the rat. Genes activated immediately before this period may play a role in cochlear function, including neurotransmission, mechano-electrical transduction, the production of endolymph and generation of the endocochlear potential. With a differential display-polymerase chain reaction (DD-PCR) method, we identified several differentially expressed sequence tags in the post-natal day 8 (P8) cochlea compared with those in the neonatal (PO) cochlea. Total RNA purified from whole cochleas of either PO or P8 rats was reverse-transcribed using 9 different oligo-dT anchored primers. The first strand cDNA was amplified by PCR with P-33-labeled dATP and pairs of primers consisting of the anchored oligo-dTs and 24 different arbitrary sequence primers, After separation of PCR products on polyacrylamide gels, twenty-one bands showed increased density on films at P8 from three separate DD-PCR runs. These bands were excised, re-amplified, subcloned and sequenced. With screening by in situ hybridization, we verified that three of those clones represented genes that were up-regulated in the P8 cochlea. One clone showed up-regulation in the stria vascularis and spiral limbus. Strong expression was also observed in adult outer hair cells. The second clone was up-regulated in cells of the developing cochlear capsule. In the adult, only bone marrow showed expression of this clone. The third clone was up-regulated in the spiral ligament and spiral limbus. This clone was not expressed in the adult cochlea.
Tension fibroblasts in the spiral ligament contain parallel arrays of actin filaments which likely function to create radial tension in the basilar membrane-spiral ligament complex (Henson and Henson, 1988; Henson et al., 1985). Their prominence and density in horseshoe bats makes the cochleae of these animals well-suited for the study of potential changes in tension as a function of postnatal development. In this study we examined the ears of horseshoe bats previously used to establish the development of hearing and vocalization (Rubsamen and Schafer, 1990). TEM micrographs were obtained for specimens of postnatal weeks one, two, three and five. Ultrastructure of the cells was studied in the acoustic fovea and in areas basal and apical to this specialized region. In the one week old animal the fibroblasts along the entire ligamentotic capsule junction are aligned in one or two rows. The cells are highly vacuolated and devoid of actin filaments and seem to be involved with protein synthesis, probably collagen (procollagen). By the second week, some adhesion plaques, which represent surface membrane specializations for the attachment of actin filaments and extracellular matrix, are present in the basal turn of the cochlea but there are only a few organized bundles of actin filaments. Between the third and fifth weeks there is a dramatic increase in the number of plaques and filaments. By the time the adult condition is reached (end of the 5th week) the surfaces of the cells that face the otic capsule and spiral ligament are almost one continuous assembly of conical adhesion plaques and the cytoplasm is almost entirely filled with parallel bundles of actin filaments. These data suggest significant changes in radial tension in the basilar membrane-spiral ligament complex during critical stages of postnatal development when hearing and vocalization are being established.
A single-channel envelope detector model was used to simulate detection of silent temporal gaps bounded by sinusoidal markers (Formby et al., 1996). The model consisted of three serial stages: an initial rounded-exponential (roex) bandpass filter, followed by a non-linearity and a single-pole lowpass filter. The ratio of the maximum-to-minimum of the waveform envelope at the output of the final filter was used as a decision rule in a two-alternative, forced-choice detection task. Temporal gap detection (TGD) thresholds were simulated as a function of the frequency separation between the pre- and post-gap markers, F-1 and F-2, at F-1=500 and 4000 Hz. When the model was implemented with an optimal first-stage filter centered between F-1 and F-2, the simulated TGD thresholds showed some qualitative similarities with the human TGD thresholds (Formby et at, 1996). The simulated thresholds for F-1=500 Hz closely matched those of humans listeners; however, simulated thresholds were smaller than those of human listeners given the same TGD task at F-1=4000 Hz. Like the human results, the simulated TGD thresholds decreased as the frequency separation between the two markers decreased. Simulated TGD thresholds also became smaller at higher E/N-0. The simulated TGD thresholds much more closely matched those of human listeners when the model was implemented with a compressive non-linearity, which limited the dynamic range of the second-stage element to 35 dB. The simulated TGD functions mirrored the roex attenuation characteristics of the first-stage filter, and increasing the bandwidth of the first-stage filter generally broadened the TGD functions. The threshold functions also reflected the same dynamic-range asymmetry found for the human TGD thresholds (i.e., a greater range of TGD thresholds for F-1 greater than or equal to F-2 than for F-1 less than or equal to F-2). TGD measured as a function of marker frequency separation may be a useful paradigm for estimating frequency selectivity and temporal acuity simultaneously, and thus may provide an understanding of the interaction between frequency resolution and temporal acuity in auditory processing.
If the electromotility of outer hair cells (OHC) is to reduce the mechanical impedance of the cochlear partition, it must not only deliver a force of sufficient magnitude, but the force must be exerted at the correct moment in the stimulus cycle. The amplitude and phase of the receptor potential of OHCs in response to direct mechanical stimulation of the stereocilia bundle were measured for cells isolated from along the entire length of the adult guinea-pig cochlea. Irrespective of their place of origin, the frequency response of the receptor potential, for 20-nm stereociliary displacement per spectral point, was governed by the OHC electrical input impedance measured near the resting potential. The response was a low-pass filter with amplitude that decreased by 6 dB/oct and phase that lagged stereocilia displacement by 90 degrees. The corner frequency of the response decreased exponentially from 546 Hz for the shortest OHC (20 mu m) with an exponential length constant of 25 mu m, or equivalently 0.58 oct per 10 mu m increase of cell length. The tonotopy was achieved by an exponential decrease in the total specific conductance with increasing cell length, beginning at 66 pS/mu m(2) for the shortest OHC; the specific capacitance was constant (2.0 mu F/cm(2)). The corner frequency was 3-6 oct below the presumed place-frequency.
