]>HEARES3680S0378-5955(01)00265-910.1016/S0378-5955(01)00265-9Elsevier Science B.V.Fig. 1Average difference plots (post-exposure minus pre-exposure DPOAEs) showing changes in DPOAE levels, in response to 55-dB SPL primary tones, for mice (n=8 ears of four mice) after a 10-min OBN exposure centered at 10 kHz, at 105 dB SPL. No change from the pre-exposure control measurements is indicated by the dotted line at ‘0’ on the ordinate. DPOAEs were maximally depressed immediately after the exposure (open squares), but partially recovered, particularly, over the lower frequency range by 1 day post-exposure (open diamonds). By 5 days following exposure (open circles), DPOAEs were essentially recovered to their pre-exposure levels. The control group of stress-only mice (n=4) showed no decreases in DPOAE levels (solid squares), along the dotted line at ‘0’, over the same measurement period. The dashed line around −20 to −30 dB, represents the NF of the combined measurement equipment and animal preparation, and delimits the maximal possible noise-induced loss in DPOAE levels. The shaded line above the abscissa centered at 10 kHz represents the frequency spectrum of the OBN exposure. In this plot and those of Figs. 2 and 3 , the error bars associated with each symbol represent SEM.Fig. 2Average DPOAE difference plots obtained with 55- (top: A,D), 65- (middle: B,E), or 75-dB (bottom: C,F) SPL primaries for mice (n=8 ears from four mice for each group), at 2 (left panels) and 31 days (right panels) following exposure to the OBN, for either 0.5 (open circles), 1 (filled squares), 3 (open diamonds), or 6 (filled circles) h. At 2 days post-exposure, increases in the duration of the OBN exposure produced significant increments in DPOAE loss (A,B,C), for each primary-tone level. However, permanent DPOAE loss essentially plateaued after 3 h of exposure, especially, after 31 days of recovery (D,E,F), for each primary-tone level. Note that, for the 1-h exposure (filled squares), DPOAE losses at 2 days post-exposure were generally less than the permanent reductions measured at 31 days post-exposure, thus suggesting that OHC function worsened during the recovery period.Fig. 3Growth of DPOAE loss as a function of the duration of exposure measured for GM frequencies at 7 kHz (f2=7.786 kHz) (top: A,D), 13.9 kHz (f2=15.6 kHz) (middle: B,E), or 24.2 kHz (f2=27.114 kHz) (bottom: C,F), with L1=L2=55 (asterisks), 65 (open circles), or 75 dB SPL (open squares), at either 2 (left panels) or 31 days (right panels) post-exposure. The 2-day loss data were best approximated by a linear-fit function, while the 31-day findings were more closely matched by a one-phase exponential [Y=Ymax(1−exp)−KX)] function. These differences in growth of loss rate may reflect, in part, different mechanisms involved in temporary versus permanent noise-induced reductions in DPOAE levels. Additionally, it is clear that DPOAEs elicited by 75-dB SPL primaries were generally less affected, particularly, at 31 days post-exposure, than those generated by the lower primary-tone levels. For each plot, the gray dashed lines, in ascending order, indicate the average NFs for the 55-, 65-, and 75-dB SPL stimuli, respectively.Fig. 4Low-power photomicrographs of mouse cochleas from a region of the cochlear duct that was approximately at a 45–60% distance from the apex of the cochlea, equivalent to the 11–18-kHz frequency region, which represents a location expected to be maximally damaged by the 10-kHz OBN exposure. Left panels (A,C,E) represent sections from a control ‘stress-only’ non-exposed mouse, while right panels (B,D,F) are from similar regions for a 6-h noise-exposed cochlea, at 1 month (31 days) post-exposure. In the noise-exposed ear (B), only about 25%, on average, of the OHCs remained in the damaged region of the cochlea, most of which were first-row OHCs. In the control cochlea (A), all three rows of OHCs are clearly visible. In this same region, the efferent innervation visualized using staining for AChE activity remained at 1 month following the OBN exposure (D,F) and was not visibly different from that for a stress-control specimen (C,E). The focus for these sections (C,D) was on the ISB, while the bottom panels (E,F) were focussed on the efferent synapses with the OHCs, in the same region of the cochlea. IHC=inner hair cells, OHC=outer hair cells for rows 1, 2, and 3, ISB=inner spiral bundle, bars=50 μm.Temporary and permanent noise-induced changes in distortion product otoacoustic emissions in CBA/CaJ miceAna E.Vázqueza*avazquez@med.miami.eduAnne E.LuebkeabGlen K.MartinabBrenda L.Lonsbury-MartinabaDepartment of Otolaryngology, University of Miami Ear Institute (M805), P.O. Box 016960, Miami, FL 33101-6960, USAbNeuroscience Program, University of Miami School of Medicine, Miami, FL, USA*Corresponding author. Tel.: +1 (305) 243-4641; Fax: +1 (305) 243-5552AbstractA number of studies have shown that the ear can be protected from sound over-exposure, either by activating the cochlear efferent system, or by sound ‘conditioning’ in which the role of the efferent system is less certain. To study more definitively the molecular basis of deliberately induced cochlear protection from excessive sounds, it is advantageous to determine, for an inbred mouse strain, a range of noise exposure parameters that effectively alter cochlear function. As an initial step towards this goal, young CBA/CaJ mice were exposed to a 105-dB SPL octave-band noise (OBN), centered at 10 kHz, for various lengths of time consisting of 10 min, or 0.5, 1, 3, or 6 h. Distortion product otoacoustic emissions (DPOAEs) at the 2f1−f2 frequency, in response to equilevel primary tones of low to moderate levels, were used to quantify the damaging effects of these sound over-exposures on cochlear function. In addition, staining for acetylcholinesterase (AChE) activity to assess for noise-induced changes in the pattern of efferent-nerve innervation to the cochlea was also performed in a subset of mice that were exposed to the longest-lasting 6-h OBN. The 10-min OBN resulted in only temporary reductions in DPOAE levels, which recovered to pre-exposure values within 5 days. Increasing the exposure to 0.5 h resulted in permanent DPOAE losses that, for low primary-tone levels, were still present at 31 days post-exposure. Additionally, the 1-h and longer exposures caused permanent reductions in DPOAEs for all test levels, which were measurable at 31 days following exposure. Light-microscopic observations restricted to the 11–18-kHz frequency region of the organ of Corti, for a subset of mice exposed to the 6-h OBN, uncovered a significant loss of outer hair cells (OHCs). However, despite the OHC loss in this region, the AChE activity associated with the related pattern of efferent innervation remained largely intact.KeywordsNoise damageCochleaDistortion product otoacoustic emissionMouseAcetylcholinesteraseEfferent innervationAbbreviationsAChE, acetylcholinesteraseAHL, age-related hearing lossCAP, compound action potentialCBA, CBA/CaJDSP, digital signal processorDPOAE, distortion product otoacoustic emissionFFT, fast Fourier transformOBN, octave-band noiseGM, geometric meanISB, inner spiral bundleOHC, outer hair cellNF, noise floorPBS, phosphate-buffered salineIHC, inner hair cellSD, standard deviationSEM, standard error of the mean1IntroductionThere has long been considerable interest in the fundamental processes that protect the ear from sound over-stimulation, especially with respect to the so-called conditioning effect, whereby prior exposure to a typically low-level sound protects the ear from a subsequent more traumatic exposure to an identical, but more intense sound. Although studied extensively, the basic processes underlying such purported protective effects are just beginning to be elucidated. One possibility, proposed by a number of investigators, is the potentially critical role played by the cochlear efferent system (Canlon et al., 1988; Franklin et al., 1991; Miyakita et al., 1992; Subramaniam et al., 1992; Pukkila et al., 1997). Although the significance of the role of the efferent system in the ‘conditioning’ phenomenon is still being debated, the function of the cochlea’s medial efferent pathway in protecting the ear from noise over-exposure is more firmly established (Cody and Johnstone, 1982; Liberman, 1991; Kujawa and Liberman, 1997; Yamasoba and Dolan, 1998). If such sound-conditioning effects could be reliably obtained in mice, a species commonly used as a molecular model of genetically based ear disorders, they would become ideal experimental subjects within which to pursue the fundamental basis of such cochlear protection.The earliest failed attempts (Fowler et al., 1995) to confirm sound conditioning in mice were unexpected given previous demonstrations of such protection in chinchillas (Clark et al., 1987), guinea pigs (Canlon et al., 1988), rabbits (Franklin et al., 1991), humans (Miyakita et al., 1992), and gerbils (Ryan et al., 1994). This unanticipated outcome in mice was likely due to the relatively low frequencies used, which consisted of narrow-band noises centered at 4.5 kHz, that tested only the extreme lower end of the mouse audibility range (see Ehret, 1974; Heffner and Masterton, 1980). More recently, using higher test and conditioning frequencies, ranging from 8 to 16 kHz, Yoshida and Liberman (2000) reported success in obtaining a protective conditioning effect of sound for CBA/CaJ (CBA) mice. In their study, either a 15-min or a 1-week training protocol reduced the extent of subsequent permanent damage, as measured by reductions in both distortion product otoacoustic emissions (DPOAEs) and compound action potentials (CAPs), compared to what would be expected in the absence of ‘conditioning’.Sun and Kim (1999) showed earlier that mice exhibit efferent-induced alterations in DPOAEs, similar to those described for the cat (Puria et al., 1996), guinea pig (Maison and Liberman, 2000), and rabbit (Luebke et al., 2000). In their paradigm, decreased or increased DPOAE levels were observed following the onset of primary tones, when the efferents innervating the outer hair cells (OHCs) were activated by ipsilaterally applied sounds. Thus, in mice, the contributions of the cochlear efferent system to sound conditioning can be assayed in a straightforward manner, and prior studies have even demonstrated that the efferent system contributes to the mouse’s cochlear response to sound over-exposure.Previous studies on the cochlear efferent system in other species have shown that efferent innervation is very robust, and is not easily affected by acoustic trauma, at least, as indicated by the presence of acetylcholinesterase (AChE) activity in efferent fibers, following the application of a traumatizing noise exposure (Rossi and Cortesina, 1965; Kokko-Cunningham and Ades, 1976). Such findings are consistent with the notion that the cochlear efferent system likely remains functional during sound exposure, and therefore, could play an active role during the actual noise over-stimulation process.Recently, Jimenez et al. (1999a) used DPOAEs to measure age-related changes in OHC activity in four inbred mouse strains, including CBA mice, and compared the susceptibility of these same strains to both temporary and permanent noise-induced cochlear dysfunction (Jimenez et al., 1999b, 2001). Their findings suggested that the cellular processes affected by a standard noise-exposure paradigm, that resulted in reversible reductions in DPOAEs, were not necessarily the same as those affected by a protocol that caused more permanent decreases in DPOAEs. In addition, because different mutant mouse strains exhibited unique exposure effects, the susceptibility of cochlear processes to noise over-stimulation appeared to be dependent on each strain’s specific genetic background (Jimenez et al., 1999b, 2001).In combination, the above findings suggest that by utilizing DPOAE measures, an ideal animal model can be established in the mouse to study the fundamental basis of an efferent-related protection and susceptibility to excessive noises at the OHC level. Mice have an advantage over other species, because genetically based variability can be reduced, or other specific defects introduced, by the selection of particular inbred strains as experimental subjects. In addition, because of the well-established existence of a number of inbred mutant strains, the mouse is one of the only laboratory species in which the genes responsible for the ear’s plasticity, in response to acoustic over-stimulation, can efficaciously be explored.Consequently, the present study was designed to explore the effects of various noise-exposure parameters on DPOAEs in the CBA mouse, with the aim of eventually developing an animal model in which the ear’s resistance, and/or susceptibility to over-exposure, could be explored at both the functional and molecular levels. Toward this end, CBA mice were exposed to one of several durations of a standardized octave-band noise (OBN) to identify the noise-exposure parameters that subsequently result in a range of both reversible and non-reversible changes in DPOAE levels. The effects of non-reversible noise damage on the innervation of the mouse’s cochlear efferent system were also studied by staining for AChE activity in a subset of mice with permanently altered ears.2Materials and methods2.1SubjectsSubjects were 2.5-month-old CBA mice, purchased from a commercial supplier (The Jackson Laboratory), and housed under routine vivarium conditions consisting of standard lighting (i.e. 12-h light/dark cycles) and free access to food and water. The mean noise level of the maintenance room in the vivarium, as measured by a noise-logging dosimeter (Quest M-27) over a representative work interval of 2 days, was about 60 dBA for 99.3% of the time, and approximately 70 dBA for the remaining time. During data collection, mice were anesthetized with an initial dose of ketamine hydrochloride (100 mg/kg) and xylazine (4 mg/kg), with maintenance doses (ketamine 50 mg/kg, xylazine 2 mg/kg) administered when needed, as indicated by twitching vibrissae. The primary study consisted of a within-subjects experimental design (i.e. pre- versus post-exposure comparisons), for five experimental exposure groups, each consisting of four mice (i.e. n=8 ears), along with a sixth ‘stress’ control group of four mice (i.e. n=8 ears). The control-group mice were subjected to the same manipulations as endured by the mice of the 10-min exposure group, but without the noise over-exposure episode. That is, they were handled and transported, anesthetized, tested at baseline for documentation of ‘pre-exposure’ DP-grams, placed in the exposure cage for 10 min, and then post-tested immediately following the mock ‘exposure’ session, and at 1 and 5 days ‘post-exposure’.2.2DPOAE measurementsThe primary functional measure was the 2f1−f2 DPOAE. Complete details of the mouse DPOAE-recording procedure have been described elsewhere (Jimenez et al., 1999a). Briefly, the f1 and f2 primary tones were generated by a dual-channel synthesizer (Hewlett-Packard 3326A) and attenuated using customized software that operated a digital signal processor (DSP) on-board a personal microcomputer system. The f1 and f2 primaries (f2/f1=1.25) were then presented over two separate earspeakers (Radio Shack, Realistic, Dual Radial Horn Tweeters), and delivered to the outer ear canal through an acoustic probe, fitted with a soft rubber tip (Etymotic, ER3-34 Infant Silicon Tip), where they were allowed to acoustically mix to avoid artifactual distortion. Ear-canal sound pressure levels, which were measured by an emissions microphone assembly (Etymotic Research, ER-10B+) embedded in the probe, were sampled, synchronously averaged and processed, using a fast Fourier transform (FFT), for geometric mean (GM) frequencies (i.e. (f1×f2)0.5) ranging from 5.6 to 19.7 kHz (i.e. f2=6.3–22.5 kHz) by the computer-based DSP board. Corresponding noise floors (NFs) were computed by averaging the levels of the ear-canal sound pressure for five frequency bins above and below the DPOAE-frequency bin (i.e. ±54 Hz).For test frequencies above 20.1 kHz, a computer-controlled dynamic-signal analyzer (Hewlett-Packard 3561A) was used. The related NFs were estimated by averaging the levels of the ear-canal sound pressure for the two FFT-frequency bins below the DPOAE frequency (i.e. for 3.75 Hz below the DPOAE). No artifactual DPOAEs were ever measured in a hard-walled cavity that approximated the size of the mouse outer ear canal. This receptacle was also used to calibrate the tonal stimuli by fitting a 1/4-in microphone at the end of the cavity that was opposite the emissions-probe assembly. For both stimulus protocols, DPOAEs were considered to be present when they were, at least, 3 dB above the NF.The primary measure in the form of the DP-gram described DPOAE levels as a function of the GM-test frequencies, and overall, these were obtained from 5.6 to 49.7 kHz (i.e. f2=6.3–54.2 kHz), in 0.1-octave increments. In the DPOAE plots below (Figs. 1 and 2), DP-grams were converted to difference functions by subtracting post-exposure functions from their corresponding pre-exposure, baseline counterparts.2.3Noise exposureAfter DPOAE baseline testing, mice were exposed to an OBN, centered at 10 kHz, at a level (RMS) of 105 dB SPL, for various lengths of time consisting of 10 min, or 0.5, 1, 3, or 6 h. Levels of the 2f1−f2 DPOAE were measured immediately after the over-stimulation episode for all mouse-exposure groups, and at 1 and 5 days post-exposure for the 10-min exposure group. For the remaining exposure groups, DPOAEs were measured at 2 and 31 days post-exposure. The OBN level was monitored with a 1/2-in microphone (ACO Pacific, AC07013), in combination with a precision sound-level meter (Quest 155). The OBN itself was generated by filtering (Frequency Devices 9002) a broadband white-noise signal, produced by a custom-made noise generator, and amplifying it with a stereo amplifier (NAD 3225 PE). The resulting OBN was presented by two direct-reflecting loudspeakers (Bose 901), that were controlled by an associated active sound equalizer. The resulting spectrum, when analyzed with the dynamic-signal analyzer in 1/3-octave bands, ranged from 8 to 15 kHz, with the maximum energy of 100 dB SPL, at 10 kHz. During the noise exposure, one awake mouse was placed into each compartment (12 cm wide) of a custom made, wire-mesh cage, that was divided into two subsections. For each free-field exposure session, two cages, containing a total of four mice, in individual compartments, with free access to food, were positioned in the center of a double-walled sound-isolation chamber, fitted with hard-reflecting surfaces, to achieve a homogeneous sound field.2.4Visualization of OHCs and efferent innervation patternsA subset of four mice exposed to the OBN for 6 h (n=8 ears), along with a counter set of four control, non-exposed mice (n=8 ears), were deeply anesthetized and systemically perfused, initially with phosphate buffered saline (PBS), and then with 4% paraformaldehyde in PBS. Next, each cochlea was dissected out and small perfusion holes placed in the apical and basal turns. The cochlea was then further fixed by gentle perfusion with a pipet that contained 4% paraformaldehyde in PBS, followed by a 2-h submersion in the latter solution. After fixation, the cochlea was decalcified by gentle stirring in a solution of 250 mM ethylenediaminetetraacetic acid, in PBS (pH 7.4), for 5 days. Prior to further processing, the decalcified cochlea was rinsed in distilled water three times.To aid in the visualization of OHCs at a high power of magnification under the light microscope, one cochlea from each mouse was lightly osmicated by gentle perfusion, and then immersed for 10 min in a 1% osmium-tetroxide solution in Dalton’s buffer, with 1 ml of 1.65% CaCl2 per 100 ml of buffer. The contralateral cochlea was stained for AChE activity. Visualization of AChE-positive fibers was accomplished using the method of Karnowsky and Roots (1964), in which the incubation solution was modified with 10−4 iso-tetra-isopropylpyrophosphoramide, using 2 mM acetylthiocholine iodide as the substrate. All control incubations carried out in the absence of substrate were negative. For all specimens, the stain was developed for 20 min at room temperature, and terminated when a brown precipitate corresponding to the inner spiral bundle (ISB) was observed under a dissecting microscope. Finally, both cochleas were embedded in plastic (araldite) and manually cut into wedge-like sections representing quarter turns of the cochlear duct following the methodology originally described by Bohne (1972) for the chinchilla, and later modified by Bohne and colleagues for the mouse (Ou et al., 2000a,b).The 11–18-kHz region of the organ of Corti of the osmium-stained cochlea, which is located approximately at a 45–60% distance from the apical extent of the cochlear duct (Ou et al., 2000a), was inspected by differential-interference contrast microscopy to determine the percentage of missing OHCs. The complete absence of the nucleus, while the plane of focus was changed from the base of the hair cell to the reticular lamina, served as the primary criterion for defining an OHC as missing. To document the histopathology identified for OHCs within the 11–18-kHz organ of Corti region, photomicrographic images were captured at a resolution of 1280×1024 pixels using a digital camera (Pixera PVC 100) mounted on a Nikon microscope (Microphot-SA). A similar region of the AChE strained cochlea was examined for amount and location of brown reaction product signifying the presence of AChE activity.2.5Data analysisFollowing collection of the emission data, DPOAE levels and their related NF levels were converted to ASCII text files, and then imported to a commercial database (Microsoft Corp., Excel 98, v. 7.0). A set of descriptive statistics was obtained, which included the computation of means, SD, and SEM. Curve fits to the post-exposure DPOAE data using mathematical functions were performed by commercially available software (GraphPad Software Inc., Prism, v. 2.0). All animal procedures were approved and monitored by the University of Miami’s Institutional Animal Care and Use Committee.3Results3.1Effects of noise over-exposure on DPOAEsThe difference DP-gram plotted in Fig. 1 illustrates the effects of the shortest, 10-min OBN exposure on DPOAE levels, elicited by the lowest-level primaries at L1=L2=55 dB SPL. This difference plot, devised by subtracting the levels of post-exposure DPOAEs from their pre-exposure, baseline counterparts, shows that immediately after the exposure (open squares), DPOAEs were reduced over the majority of the tested frequency range, except for the highest-test frequencies, that were greater than about 40-kHz GM frequency (f2>50.6 kHz). In the frequency region from 14.1 to 30 kHz, where the maximum noise-induced reductions in DPOAE levels were expected, i.e. at about one half to an octave or two above the 10-kHz OBN, emissions were clearly reduced to NF levels (dashed line). At 1 day post-exposure (open diamonds), DPOAE levels had partially recovered over the lower-frequency range, from 5.6 to 11 kHz. Moreover, by 5 days post-exposure (open circles), the emissions had essentially returned to their pre-exposure levels. Thus, after 5 days of recovery, it is apparent that DPOAEs from the mice exposed to the 10-min OBN, completely superimposed those obtained from their stress-control counterparts (filled squares), which received the same manipulations, but without being subjected to the actual OBN exposure. It is interesting to note that, within the frequencies encompassed by the exposure band (shaded bar along the abscissa), low-frequency DPOAEs recovered more rapidly than those at the more vulnerable higher frequencies. In addition, it is clear from Fig. 1 that variability in the after-effects of the OBN on DPOAE levels between mice, as reflected by the S.E.M. bars, were greatest for the frequencies most affected by the over-exposure episode. In summary, a 10-min OBN exposure resulted only in moderate, temporary cochlear dysfunction in CBA mice as measured by DPOAEs.In Fig. 2A–F, average DP-gram difference plots are shown for mice at 2 (A–C) and 31 days (D–F), following OBN exposures lasting either 0.5 (open circles), 1 (solid squares), 3 (open diamonds), or 6 h (solid circles), for each of the three primary-tone levels tested at L1=L2=55 (A, D), 65 (B, E), or 75 dB (C, F) SPL. While the 2 days post-exposure data (left panels) presumably represents a mixture of both the reversible and permanent components of over-exposure effects, DPOAEs obtained at 31 days post-exposure (right panels) reflect only the permanent consequences of the OBN over-stimulation.From the inspection of these functions, it appears that the 0.5-h OBN exposure (open circles), at 31 days post-exposure (Fig. 2D–F), was near the threshold for producing widespread permanent DPOAE dysfunction for the CBA mouse. That is, on average, the 0.5-h exposure resulted in a mild, permanent decrement in both mid- and high-frequency DPOAEs, at all levels of test stimulation. Clearly, for this exposure group, as shown in Fig. 2F, DPOAEs elicited by 75-dB SPL primaries exhibited near-normal levels at 31 days post-exposure, especially for frequencies <40 kHz. Also note, for the 0.5-h OBN exposure at 31 days of recovery, not unexpectedly, DPOAEs elicited by the lowest-level primaries at 55-dB SPL stimuli (Fig. 2D) exhibited the greatest degree of variability in the most affected frequency region (i.e. ∼13–17 kHz) as indicated by the large S.E.M. values.As might be anticipated, incrementing the exposure duration from 0.5 to 1, 3, or 6 h, significantly increased the DPOAE losses observed at 2 days post-exposure. For example, as illustrated in Fig. 2A–C, at 2 days post-exposure, the 6-h exposure (solid circles) always reduced DPOAEs to NF levels, while, depending upon the primary-tone level, the 1- (solid squares) and 3-h (open diamonds) exposures did not. Overall, at 2 days of recovery, as shown in Fig. 2A,B, increasing exposure duration produced an orderly increase in DPOAE loss, especially for those elicited by the 55- and 65-dB SPL primaries, respectively. In contrast, permanent losses at 31 days post-exposure, as shown in Fig. 2D–F, were much less dependent upon exposure duration. That is, although increasing exposure duration from 0.5 to 1 h substantially increased DPOAE losses, particularly for the lower-level primaries, as shown in Fig. 2D,E, further increases in duration produced a much smaller increment in permanent DPOAE loss. Overall, the amount of DPOAE loss at 2 days post-exposure was much more dependent upon exposure duration than was the permanent loss determined at 31 days. Thus, it is clear that the 3- versus 6-h exposures produced considerably different amounts of initial reductions in DPOAE levels, as illustrated, particularly, in Fig. 2B,C. However, at 31 days post-exposure, these unique exposure durations resulted in similar amounts of permanent loss as shown in Fig. 2D–F. Moreover, although considerable variability, as indicated by the SEM bars, is evident across the exposure groups, such intersubject differences were greater for DPOAEs elicited by the lower primary-tone levels (e.g. Fig. 2A,D), and for emissions measured at 31 days post-exposure, when permanent changes were apparent (Fig. 2D–F).Fig. 3 shows more clearly the above-noted differences between recovery functions obtained early during the recovery process, and those measured about 1 month later. Specifically, these plots show, for three representative test frequencies at 7 (A,D), 13.96 (B,E), and 24 kHz (C,F), differences in the growth of the post-exposure DPOAE loss as a function of the exposure durations of 0.5, 1, 3, and 6 h, at 2 (Fig. 3A–C) and 31 days (Fig. 3D–F) of recovery, for each of the three primary-tone levels tested at 55 (asterisks), 65 (open circles), and 75 dB (open squares) SPL. For example, the plots of Fig. 3B,E indicate the DPOAE loss computed for the GM test frequency of 13.96 kHz (f2=15.6 kHz), i.e. at approximately 1/2 an octave above the center frequency of the 10-kHz OBN exposure, where the noise-induced reductions in DPOAE levels were maximum. It is clear that the 2-day post-exposure rate of loss shown in Fig. 3B was approximated by a linear fit to the data (i.e. r2=0.3, 0.7, and 0.8, for the L1=L2=55-, 65-, and 75-dB SPL primaries, respectively). In contrast, the more permanent DPOAE losses at 31 days post-exposure illustrated in Fig. 3E were approximated more closely by an exponential function (i.e. r2=0.4, 0.5, and 0.3, for the L1=L2=55-, 65-, and 75-dB SPL primaries, respectively). These low correlations resulted primarily from the substantial variability of the noise-induced reductions in DPOAEs between the 0.5- and 1-h exposure groups. When measured at 2 days post-exposure, each hour of OBN over-stimulation resulted in approximately 6.1, 5.7, or 3.5 dB of DPOAE loss, for the 75-, 65, and 55-dB SPL primaries, respectively, until the DPOAEs reached their corresponding NF.The growth of permanent DPOAE loss measured at 31 days of recovery as a function of exposure duration for the two lower-level primaries, at 55 and 65 dB SPL, began to asymptote by 3 h of exposure, at the NF of the measurement equipment, which represented a loss of approximately 30 dB from pre-exposure baseline levels. For the three test-stimulus levels, longer exposures at 3 and 6 h produced less variable reductions in DPOAE levels. More specifically, for the higher primary tones tested at 75 dB SPL, the 0.5-h exposure produced no decrement in DPOAE levels, while for the 1-h and longer exposures, DPOAE loss began to asymptote at approximately 10 dB below pre-exposure levels. Consequently, the DPOAEs elicited by the 75-dB SPL primaries were much less affected by the OBN than those elicited by the lower-level primary tones. However, the pattern of DPOAE loss elicited by the 75-dB SPL primaries followed a similar, if not parallel-shifted, exponential function that described the eventual noise-induced effects.From the inspection of the plots of Fig. 3A,D, or C,F, for the GM test frequencies of 7 and 24 kHz, respectively, it is clear that the above-noted observations for the 13.96-kHz GM test frequency generalize to other test frequencies located both below and above the one expected to show the greatest loss at about half an octave above the center frequency of the exposure (i.e. at 13.96 kHz), even if the amount of DPOAE loss was considerably less (e.g. for 7 kHz).3.2Histological observationsIn general, there was little evidence of OHC loss in the apical turn of the cochlea for either the control or 6-h noise-exposed mice. However, in the basal turn, a segment of cochlear duct that was at about a 45–60% distance from the apex, and that clearly contained a region of OHC loss, was selected for more detailed study. As this area corresponded approximately to the 11–18-kHz region of the mouse hearing range, according to the computations of Ou et al. (2000a), the sensory cell damage was considered, most likely, to be due to the 10-kHz OBN exposure. Certainly, this particular cochlear region corresponded to areas of the mouse DP-grams of the present study that exhibited, at least, a 25-dB decrement in DPOAE levels measured with either 55- or 65-dB SPL primaries. For control mice, the three rows of OHCs in this upper basal-turn region were always visible and appeared normal, as shown in Fig. 4A, for a representative non-exposed mouse. In contrast, as illustrated in Fig. 4B for a 6-h exposed mouse, up to 70% of the OHCs were missing in the noise-exposed mice that were sacrificed about 1 month after the over-stimulation episode.To determine the influence of noise exposure on efferent innervation, four mice from each of the control and the 6-h exposed groups were examined histologically for AChE activity at 31 days post-exposure. In these cochleas, the most intense AChE activity was always observed in the nerve fibers of the ISB, with no apparent discontinuity in staining, even in regions expected to be severely damaged by the OBN exposure (i.e. over the 11–18-kHz region). This finding can be appreciated by comparing panels C and E of Fig. 4, showing light photomicrographs of an AChE-stained control mouse, with panels D and F displaying similar cochlear regions for a 6-h noise-treated mouse. Upon comparison of these sections, it is clear that there were no obvious differences in the amount of brown reaction product between the two non-exposure and exposure conditions.However, overall, there was more variability in the surface preparations of the noise-exposed mice stained for AChE activity. For example, less intense staining of the ISB was observed in some of the noise-damaged cochleas, along with a disrupted arrangement of the efferent endings on the OHCs, and less packed tunnel crossing fibers. In other preparations, where the efferent innervation seemed abnormal, the tunnel of Corti appeared narrow as compared to the appearance of this structure in more normal specimens. Importantly, none of the AChE-stained control mice showed these abnormalities. Considering that these histological data were obtained from the mice receiving the most severe 6-h exposures, that resulted in the greatest DPOAE losses, it is unlikely that many significant changes would have been observed in the efferent endings of mice exposed for the shorter time durations.4DiscussionThe present study was designed to examine the effects of a range of noise-exposure durations on the DPOAEs of CBA mice. The major aim was to identify over-stimulation paradigms that produced OBN-induced changes in exposed mice that could eventually be used to study, in detail, the protective effects of sound conditioning, and the potential role of the cochlear efferent system in such processes. The findings revealed that, for the 2.5-month-old CBA mouse, a 10-min exposure was capable of initially reducing, to NF levels, the range of testable DPOAE frequencies, in and above the frequency band of the 105-dB SPL, 10-kHz exposure. However, no permanent effects were observed 5 days later for this briefest exposure duration. Increasing the duration to 0.5 h appeared to be near the threshold for producing permanent DPOAE decrements, while exposures beyond 3 h did not appreciably increase the amount of DPOAE loss observed for the 3-h exposures themselves, especially after 31 days of recovery. Thus, the tested exposures defined a range of durations within which acoustic over-stimulations appropriate for sound conditioning can be selected, as well as ones relevant for subsequently producing permanently damaging effects on OHC functions.The present study used changes in DPOAE levels to evaluate cochlear function. Thus, the findings reflect functional alterations that were restricted to the OHC level of the peripheral auditory system. In contrast, most prior studies of noise-induced damage in CBA mice utilized electrophysiological potentials to estimate subsequent changes in hearing thresholds. For example, Henry (1982) measuring a CAP-like response in adult CBA mice, was first to demonstrate that a 5-min exposure to a 124-dB SPL OBN, ranging from 12 to 24 kHz, produced changes in such cochlear potentials that lessened with increasing age. Specifically, he showed that 60-day-old mice, representing an early post-puberty stage of development, exhibited the greatest noise-induced threshold elevations, whereas a progressively less severe effect was observed for older mice between the ages of 90 and 360 days.In contrast to the CAP measure used by Henry (1982), the majority of the more recent studies of noise-induced modifications of auditory responses in the CBA mouse have used the auditory brainstem response (ABR) to estimate changes in hearing thresholds. For example, in an extensive series of studies, Li and Borg and colleagues (Li and Borg, 1991; Li, 1992; Li et al., 1993) examined noise-induced damage in CBA mice using high-level exposures, at 120 dB SPL, that were lower in their frequency range, at 2–7 kHz, than the ones reported here. These investigators showed that susceptibility to noise damage in the CBA mice was maximal at 1 month of age, and decreased substantially by 3 months of age. Other results by Shone et al. (1991) also indirectly suggested that 6-month-old mice were more resistant to noise over-exposure. These investigators showed that a 45-min, 101-dB SPL exposure produced no permanent ABR threshold shifts. Finally, the findings that older mice are more resistant to noise damage is also supported in a recent study by Davis et al. (1999), in which 3–4-month-old CBA mice showed no permanent ABR threshold shifts following a 1-h exposure to a 110-dB SPL OBN, centered-weighted between 7 and 17 kHz.The DPOAE results reported here also support the notion of an age-dependent susceptibility to excessive sounds. Thus, the present findings of significant noise-exposure effects in 75-day-old mice (i.e. age=2.5 month) can be contrasted to those of Jimenez et al. (1999b, 2001), who showed, in slightly more mature 90-day-old CBA mice (i.e. age=3 months), that the same 10-kHz, 105-dB SPL, 1-h OBN exposure as used here did not result in any detectable permanent DPOAE losses.Recently, two studies by other investigators reported the effects on DPOAEs of exposing CBA mice to a similar OBN. Specifically, for 2.5-month-old CBA mice, Yoshida et al. (1999, 2000) described frequency-dependent permanent threshold shifts for CAPs and DPOAEs following a 2-h exposure to a 100-dB SPL OBN ranging from 8 to 16 kHz. The findings of these investigators are in agreement with the present results, thus, supporting an increased susceptibility to noise for young CBA mice. Moreover, although Yoshida et al. (1999) observed no OBN-induced loss of OHCs, they described a disarrangement of the stereocilia pattern on OHCs for mice that were terminated 7 days after the exposure. In the present study, mice were histologically examined at 1 month after the OBN exposure. Thus, it is possible that the longer survival time, or the longer duration of the 6-h exposure they experienced, accounts for the degeneration of the OHCs that was observed in the region of the noise-exposure band for these subjects. In all, it is important to take into account the combined findings of the present study and those of Jimenez et al. (1999b, 2001), that CBA mice have a relatively abrupt change in their susceptibility to noise between 2 and 3 months of age, when comparing the susceptibility of the CBA strain to other mouse strains, such as the 129/SvEv (Yoshida et al., 2000).The effects of noise over-exposure on the efferent-nerve innervation of the cochlea were studied previously in several experimental models. In an early study in guinea pigs by Rossi and Cortesina (1965), treatment with neomycin sulfate (20 mg/kg/day) combined with an acoustic trauma, in the form of an 8-h, 90-dB SPL pure-tone exposure, at either 0.512, 2.048, 4.096, or 6.144 kHz, for 60 days, completely destroyed the organ of Corti in the majority of subjects. However, at 18 months post-exposure, for guinea pigs in which portions of the organ of Corti remained, histologic examination showed that AChE activity was present. Although the animals exhibiting a complete destruction of the organ of Corti showed no AChE staining in the cochlear duct, AChE activity clearly remained in the intraganglionic spiral bundle, and within the canaliculi of the osseous spiral lamina. Further work by Strominger et al. (1995) in the chinchilla showed a complete disappearance of efferent fibers, but only in cochlear regions that had been entirely destroyed by exposure for 12 h to a 0.5-kHz OBN at 120 dB SPL. Again, in the ‘partial wipeout’ regions described by these authors, where remnants of the organ of Corti remained, AChE-positive fibers were still observed.To our knowledge, observations of AChE-positive fibers in noise-treated mice have not yet been reported. The apparent survival of efferents when the organ of Corti was incompletely damaged may be due to a small percentage of remaining OHC and supporting cells. These cells may help maintain an environment that supports survival of the efferent fibers. Alternatively, if, as discussed by Kokko-Cunningham and Ades (1976), the efferent fibers bifurcate at the level of the tunnel, based on light- and electron-microscopy observations, such branching would allow a fiber to synapse onto more than one OHC, thus, allowing the fiber to survive, even when a large percentage of the target cells were missing. In any case, it appears that maintenance of efferent innervation seems dependent on there being some remaining target cells, or the afferent fibers, onto which a portion of the efferent fibers are known to synapse. Thus, it appears that, during both the temporary and permanently damaging effects of sound exposure, efferents are present and may actively participate in the damage process.There is considerable evidence that different mechanisms are involved in the temporary versus permanent effects of noise damage (Nordmann et al., 2000), regardless of the methods used to assess the damaged ear. The findings at 2 days post-exposure, when temporary influences presumably dominate, that decreases in DPOAEs proceeded linearly as a function of exposure duration, whereas permanent losses developed in an exponential manner and peaked, also argues for significant differences between the temporary and permanent aspects of noise damage as assessed with DPOAEs. Similar conclusions regarding the differences between temporary and permanent DPOAE losses were reached when comparing the temporary effects of a brief 1-min pure-tone exposure to those produced by the 1-h OBN employed here, in normal CBA mice versus three strains of mutant mice exhibiting AHL (Jimenez et al., 1999b).A major observation of the present investigation was that DPOAEs evoked by high- as compared to low-level primaries behaved differently with respect to their sensitivity to noise damage. That is, DPOAEs evoked by high-level primary tones (Fig. 2F) were much less affected than those evoked by lower-level primaries (Fig. 2D). What is especially interesting with respect to this finding is the observation that the temporary effects of noise exposure can completely eliminate DPOAEs evoked by high-level primaries (e.g. Fig. 2C). However, when the driving potential across the OHC is reduced by the temporary effects of loop diuretics on the endocochlear potential, low-level DPOAEs are eliminated, while those evoked by 75-dB SPL primaries are completely unaffected (see Fig. 3 in Whitehead et al., 1992). Thus, the insensitivity of high-level DPOAEs does not necessarily pertain to temporarily damaging noise exposures. Perhaps, in this reversible situation, OHC stereocilia are damaged or detached from the tectorial membrane (Nordmann et al., 2000), so that no amount of acoustic stimulation can activate the non-linear aspects of OHC transduction to result in the production of DPOAEs. Similar results were also obtained when the effects of age-related hearing loss (AHL) were studied and compared in three strains of mice exhibiting AHL (Jimenez et al., 1999a). In these latter experiments, high-level primaries were just as effective as lower-level tones in detecting the progressive OHC losses. Again, it appears as if some aspect of OHC transduction is gradually eliminated during longer durations of noise over-exposure that prevents even high-level stimuli from eliciting DPOAEs.Finally, although noise exposure in humans tends to produce the greatest losses in hearing sensitivity at a half to an octave above the frequency band of the exposure (e.g. Taylor et al., 1965), the after-effects noted here at a half to two octaves above the 10-kHz OBN are considerably more extensive for mice. These latter observations have also been demonstrated by Henry (1982) and Ou et al. (2000b), who showed that mice sustained maximum noise-induced shifts in ABR thresholds for frequencies more than two octaves above the exposure band.In summary, the present results define noise-exposure parameters that produce either temporary or permanent decreases in DPOAE levels, indicative of a loss of OHC function in young 2.5-month-old CBA mice. For purposes of sound conditioning, it appears that exposures should be limited to 10–15 min, if levels as high as 105 dB SPL are employed. For more traumatic exposures, durations in the range of 1–3 h appear adequate. Based on the first reports of successful sound conditioning in mice (Yoshida et al., 1999; Yoshida and Liberman, 2000), the exposure parameters suggested here, at about 105 dB SPL for 10 min, are within the ranges needed to rapidly produce such training effects.Observations in the present work of the effects of noise over-exposure on efferent innervation demonstrated that the efferent fibers remain, even after severe damage to their OHC targets, although a much larger variability in the surface preparations of the noise-exposed mice stained for AChE activity was apparent than for control animals. In addition, less intense staining of the ISB was observed, along with, in some noise-damaged cochleas, a disrupted arrangement of the efferent endings and less packed tunnel crossing fibers. These features of some 6-h exposed cochleas were not observed in any of the control cochleas. Such observations are in agreement with earlier studies in guinea pig and chinchilla which showed that, only in regions where the organ of Corti was completely destroyed, did the efferent innervation degenerate (Rossi and Cortesina, 1965; Kokko-Cunningham and Ades, 1976; Strominger et al., 1995). Based upon the present findings in mice and those in other species, it appears that efferents could be active over a large range of noise-exposure conditions, consistent with a number of studies implicating the cochlear efferent system in protection of the ear during the exposure process. Overall, it appears that the mouse has considerable potential for studies of cochlear protection and susceptibility to noise damage. Thus, DPOAE measurements in the mouse may provide one means of assessing noise conditioning and susceptibility effects as they specifically relate to OHC transduction.AcknowledgementsSupported by NIDCD (DC00613, DC03114, DC03086), NIA (AG17275), and the University of Miami’s Chandler Chair Fund.ReferencesBohne, 1972B.A.BohneLocation of small cochlear lesions by phase contrast microscopy prior to thin sectioningLaryngoscope821972116Canlon et al., 1988B.CanlonE.BorgA.FlockProtection against noise trauma by pre-exposure to a low level acoustic stimulusHear. Res.341988197200Clark et al., 1987W.W.ClarkB.A.BohneF.A.BoettcherEffects of periodic rest on hearing loss and cochlear damage following exposure to noiseJ. Acoust. Soc. Am.82198712531264Cody and Johnstone, 1982A.R.CodyB.M.JohnstoneTemporary threshold shifts modified by binaural acoustic stimulationHear. Res.61982199205Davis et al., 1999R.R.DavisM.L.CheeverE.F.KriegL.C.ErwayQuantitative measure of genetic differences in susceptibility to noise-induced hearing loss in two strains of miceHear. Res.1341999915Ehret, 1974G.EhretAge-dependent hearing loss in normal hearing miceNaturwissenschaften111974506Fowler et al., 1995T.FowlerB.CanlonD.F.DolanJ.M.MillerThe effect of noise trauma following training exposures in the mouseHear. Res.881995113Franklin et al., 1991D.J.FranklinB.L.Lonsbury-MartinB.B.StagnerG.K.MartinAltered susceptibility of 2f1−f2 acoustic-distortion products to the effects of repeated noise exposure in rabbitsHear. Res.5819915762Heffner and Masterton, 1980H.E.HeffnerB.MastertonHearing in Glires: Domestic rabbit, cotton rat, house mouse, and kangaroo ratJ. Acoust. Soc. Am.68198015841599Henry, 1982K.R.HenryAge-related changes in sensitivity of the postpubertal ear to acoustic traumaHear. Res.81982285294Jimenez et al., 1999aA.M.JimenezB.B.StagnerG.K.MartinB.L.Lonsbury-MartinAge related loss of distortion product otoacoustic emissions in four mouse strainsHear. Res.138199991105Jimenez et al., 1999bA.M.JimenezB.B.StagnerG.K.MartinB.L.Lonsbury-MartinEffects of noise exposure on distortion product otoacoustic emissions in four mouse models of age-related hearing lossAssoc. Res. Otolaryngol. Abstr.22199996Jimenez et al., 2001Jimenez, A.M., Stagner, B.B., Martin, G.K., Lonsbury-Martin, B.L., 2001. Susceptibility of DPOAEs to sound over-exposure in inbred mice with AHL. J. Assoc. Res. Otolaryngol. (in press).Karnowsky and Roots, 1964M.J.KarnowskyL.RootsA ‘direct coloring’ thiocholine method for cholinesterasesJ. Histochem.121964219221Kokko-Cunningham and Ades, 1976A.Kokko-CunninghamH.W.AdesAcetylcholinesterase activity in the chinchilla organ of Corti in normal and acoustically overstimulated animalsActa Otolaryngol.8119764856Kujawa and Liberman, 1997S.G.KujawaM.C.LibermanConditioning-related protection from acoustic injury: Effects of chronic deefferentation and sham surgeryJ. Neurophysiol.78199730953106Li, 1992H.LiGenetic influences on susceptibility of the auditory system to aging and environmental factorsScand. Audiol.361992139Li and Borg, 1991H.LiE.BorgAge-related loss of auditory sensitivity in two mouse phenotypesActa Otolaryngol.1111991827834Li et al., 1993H.LiM.HultcrantzE.BorgInfluence of age on noise-induced permanent threshold shifts in CBA/Ca and C57BL/6J miceAudiology321993195204Liberman, 1991M.C.LibermanThe olivocochlear efferent bundle and susceptibility of the inner ear to acoustic injuryJ. Neurophysiol.651991123132Luebke et al., 2000A.E.LuebkeB.B.StagnerG.K.MartinB.L.Lonsbury-MartinAn efferent sum measure predicts susceptibility to sound over-exposure in rabbitAssoc. Res. Otolaryngol. Abstr.232000282Maison and Liberman, 2000S.F.MaisonM.C.LibermanPredicting vulnerability to acoustic injury with a noninvasive assay of olivocochlear reflex strengthJ. Neurosci.20200047014707Miyakita et al., 1992T.MiyakitaP.A.HellstromE.FrimansonA.AxelssonEffect of low level acoustic stimulation on temporary threshold shift in young humansHear. Res.601992149155Nordmann et al., 2000A.S.NordmannB.A.BohneG.W.HardingHistopathological differences between temporary and permanent threshold shiftHear. Res.13920001330Ou et al., 2000aH.C.OuG.W.HardingB.A.BohneAn anatomically based frequency-place map for the mouse cochleaHear. Res.1452000123129Ou et al., 2000bH.C.OuG.W.HardingB.A.BohneNoise damage in the C57BL/CBA mouse cochleaHear. Res.1452000111122Pukkila et al., 1997M.PukkilaS.ZhaiJ.VirkkalaU.PirvolaJ.YlikoskiThe ‘toughening’ phenomenon in rat’s auditory organActa Otolaryngol.52919975962Puria et al., 1996S.PuriaJ.J.GuinanM.C.LibermanOlivocochlear reflex assays: Effects of contralateral sound on compound action potentials versus ear-canal distortion productsJ. Acoust. Soc. Am.991996500507Rossi and Cortesina, 1965G.RossiG.CortesinaAcetylcholinesterase activity in the efferent cochlear fibers after destruction of the organ of Corti and afferent fibresActa Otolaryngol.611965488494Ryan et al., 1994A.F.RyanT.M.BennettN.K.WoolfA.AxelssonProtection from noise-induced hearing loss by prior exposure to an nontraumatic stimulus: Role of the middle ear musclesHear. Res.7219942328Shone et al., 1991G.ShoneR.A.AltschulerJ.M.MillerA.L.NuttallThe effect of noise exposure on the aging earHear. Res.561991173178Strominger et al., 1995R.N.StromingerB.A.BohneG.W.HardingRegenerated nerve fibers in the noise-damaged chinchilla cochlea are not efferentHear. Res.9219955262Subramaniam et al., 1992M.SubramaniamD.HendersonP.CampoV.SpongrThe effect of ‘conditioning’ on hearing loss from a high frequency traumatic exposureHear. Res.5819925762Sun and Kim, 1999X.M.SunD.O.KimAdaptation of 2f1−f2 distortion product otoacoustic emission in young-adult and old CBA and C57 miceJ. Acoust. Soc. Am.105199933993409Taylor et al., 1965W.TaylorJ.PearsonA.MairW.BurnsStudy of noise and hearing in jute weavingJ. Acoust. Soc. Am.381965113120Whitehead et al., 1992M.L.WhiteheadB.L.Lonsbury-MartinG.K.MartinEvidence for two discrete sources of 2f1−f2 distortion-product otoacoustic emission in rabbit. II: Differential physiological vulnerabilityJ. Acoust. Soc. Am.92199226622682Yamasoba and Dolan, 1998T.YamasobaD.F.DolanThe medial cochlear efferent system does not appear to contribute to the development of acquired resistance to acoustic traumaHear. Res.1201998143151Yoshida and Liberman, 2000N.YoshidaM.C.LibermanProtective effect of sound conditioning in CBA/CaJ miceAssoc. Res. Otolaryngol. Abstr.163200079Yoshida et al., 1999N.YoshidaA.KrisiansenM.C.LibermanHeat stress and protection from permanent acoustic injury in miceJ. Neurosci.1919991011610124Yoshida et al., 2000N.YoshidaS.J.HequembourgC.A.AtencioJ.J.RosowskiM.C.LibermanAcoustic injury in mice: 129/SvEv is exceptionally resistant to noise-induced hearing lossHear. Res.141200097106