The medial olivocochlear system attenuates the developmental impact of early noise exposure

Abstract

The early onset of peripheral deafness profoundly alters the functional maturation of the central auditory system. A prolonged exposure to an artificial acoustic environment has a similar disruptive influence. These observations establish the importance of normal patterns of sound-driven activity during the initial stages of auditory development. The present study was designed to address the role of cochlear gain control during these activity-dependent developmental processes. It was hypothesized that the regulation of auditory nerve activity by the medial olivocochlear system (MOCS) would preserve normal development when the immature auditory system was challenged by continuous background noise. To test this hypothesis, knock-out mice lacking MOCS feedback were reared in noisy or quiet environments and then evaluated with behavioral paradigms for auditory processing deficits. Relative to wild-type controls, noise-reared knock-out mice showed a decreased ability to process rapid acoustic events. Additional anatomical and physiological assessments linked these perceptual deficits to synaptic defects in the auditory brainstem that shared important features with human auditory neuropathy. Our findings offer a new perspective on the potentially damaging effects of environmental noise and how these risks are ameliorated by the protective role of MOCS feedback.

FIG. 1

One third octave band sound levels measured over an 8-h period in a noisy heavy-traffic vivarium (A) and quiet restricted-access vivarium (B). Acoustic transients were more frequent in the noisy room. Differences in the average noise spectra of the two rooms were concentrated at frequencies below 10 kHz (C).

FIG. 2

Stimulus paradigms for the three experiments. A The GIASR task measured the subject’s ability to process rapid acoustic events. On baseline trials, a startle-eliciting stimulus was presented with continuous background noise. On gap trials, the background noise was turned off with varying lead times before the presentation of the startle-eliciting stimulus. B Conventional ABR paradigm for measuring the integrity of sound processing in the auditory nerve and brainstem. Clicks or 5-ms tone bursts were presented at a rate of 10 stimuli/s to allow 95–100 ms of recovery between stimuli. C Modified ABR paradigm for measuring the effects of presentation rate on the magnitude and latency of evoked potentials. Tones were presented with longer durations (20 ms) and shorter recovery periods (1–15 ms).

FIG. 3

Individual lead time functions for the GIASR task. A Baseline controls for quiet-reared (gray) and noise-reared WT mice (black). Relative ASR is plotted in relation to the lead time of noise offset. A relative ASR of less than 1.0 (dotted line) indicates GIASR. B, C Lead time functions for quiet-reared and noise-reared α9KO mice.

FIG. 4

A statistical analysis of the effects of treatment group on GIASR. ABoxplots of the distribution of ASR measures for each lead time. See text for explanation of plotting conventions. B 95% confidence intervals (CIs) for the global rank scores of each distribution. CIs that fall entirely outside the baseline responses of quiet-reared WT mice (shaded areas) indicate statistically significant treatment effects (P < 0.05, asterisks).

FIG. 5

A statistical analysis of the effects of treatment group on hearing sensitivity. ABR thresholds were obtained with clicks and tones (8, 16, and 32 kHz). Threshold distributions and the 95% confidence intervals for global rank scores are shown in A and B, as described in Figure. 4. Thresholds did not vary significantly across treatment groups.

FIG. 6

Correlations between ABR thresholds and relative ASR measures. Results are shown for noise-reared α9KO mice at the four lead times that produced statistically significant decreases in GIASR (Fig. 4B). At each lead time, the ASR is plotted twice in relation to the 8- and 16-kHz thresholds of the same subject. The same thresholds are repeated at all four lead times. Linear fits (dashed and solid lines) revealed statistically significant correlations (P < 0.05) between ASR and 8-kHz thresholds at lead times of 1 and 2 ms.

FIG. 7.

Comparison of ABRs obtained with conventional procedures. To minimize the effects of one stimulus presentation on the next, 8-kHz tone bursts were presented with a short duration (5 ms) and a long ISI (95 ms). A, B Waveforms from a representative mouse in each of the four treatment groups. Numerical labels indicate presentation level. C, DBoxplots and 95% CIs for the magnitude of wave 1 and the interpeak latency separating P1 and P2 for all mice that were tested with 8-kHz tones at 80 dB SPL. The features did not show statistically significant treatment effects.

FIG. 8.

Effects of presentation rate on ABR magnitude and latency. To challenge stimulus recovery beyond the conventional ABR procedure, the duration of 8-kHz tone bursts was increased to 20 ms and ISI was varied parametrically. A, B Waveforms from a representative mouse in each of the four treatment groups. Numerical labels indicate ISI. C–E Effects of ISI on wave 1 magnitude, interpeak latency (P1 to P2), and wave 2 magnitude. Error bars indicate the inter-quartile range of the distributions. α9KO mice showed statistically smaller wave 1 and 2 magnitudes than quiet-reared WT mice (P < 0.05, asterisks). Noise-reared α9KO mice also showed longer interpeak latencies.

FIG. 9

Normal structure and function of outer hair cells in a noise-reared α9KO mouse. A Middle turn of the cochlea showing a normal complement of outer hair cells and MOCS terminals. Hair cells were immunolabeled with antibodies for DAPI (blue) and myo7a (green). Efferent terminals were immunolabeled with SV2 (red). B Distortion product otoacoustic emissions produced by primary tones (F1 and F2) at frequencies near 8 and 16 kHz. A cubic difference distortion product (DP) was generated at both frequencies.

FIG. 10

Abnormal endbulb ultrastructure in a noise-reared α9KO mouse. A The soma of a globular bushy cell (color fill) projects a large somatic spine into a nest of nerve terminals. The endbulb termination of an auditory nerve fiber forms a long, wave-shaped postsynaptic density and puncta adherentia complex with the somatic spine (black arrowheads). A second postsynaptic density and puncta adherentia complex contacts the main somatic surface (white arrowheads). A mitochondrion–adherens complex (MAC) near the somatic synapse displays an abnormal vesicular chain (asterisk). B The vesicular chain of the MAC in higher magnification (color fill).