Noise-induced cochlear synaptopathy in C57BL/6 N mice as a function of trauma strength: ribbons are more vulnerable than postsynapses

Abstract

Noise-induced cochlear synaptopathy is characterized by irreversible loss of synapses between inner hair cells (IHCs) and spiral ganglion neurons (SGNs) despite normal hearing thresholds. We analyzed hearing performance and cochlear structure in C57BL/6 N mice exposed to 100, 106, or 112 dB SPL broadband noise (8-16 kHz) for 2 h. Auditory brainstem responses (ABRs) were assessed before, directly after, and up to 28 days post-trauma. Finally, the number, size, and pairing of IHC presynaptic (CtBP2-positive) ribbons and postsynaptic AMPA receptor scaffold (Homer1-positive) clusters were analyzed along the cochlea. Four weeks after the 100 dB SPL trauma, a permanent threshold shift (PTS) was observed at 45 kHz, which after the higher traumata extended toward middle to low frequencies. Loss in ABR wave I amplitudes scaled with trauma strength indicating loss of functional IHC synaptic connections. Latencies of wave I mostly increased with trauma strength. No trauma-related OHC loss was found. The number of synaptic pairs was reduced in the midbasal and basal cochlear region in all trauma conditions, with ribbon loss amounting up to 46% of control. Ribbons surviving the trauma were paired, whereas 4-6 unpaired postsynapses/IHC were found in the medial, midbasal, and basal regions irrespective of trauma strength, contrasting findings in CBA/CaJ mice. Our data confirm the susceptibility of ribbon synapses and ABR wave I amplitudes to a noise trauma of 100 dB SPL or larger. Notably, peripheral dendrites bearing IHC postsynapses were less vulnerable than presynaptic ribbons in C57BL/6 N mice.

Keywords: ABR; auditory nerve; cochlear synaptopathy; hair cell; hidden hearing loss; noise trauma; postsynapse; ribbon.

Figure 1

Experimental design of hearing measurements, trauma application, and cochlear assignment. (A) ABR thresholds were initially recorded in mice aged 7–8 weeks at day-2 for each animal. At day 0, the animal received a 2 h band noise trauma, 8–16 kHz, of either 100 dB, 106 dB, or 112 dB SPL. Days of further ABR measurements are indicated. ABRs of the no trauma control group were recorded on days-2 and 28 only. After a final ABR recording on day 28, animals were euthanized and 4 of them were processed for immunohistochemistry (see Methods). (B) Sketch with the assignment of auditory frequencies to regions of the cochlear spiral in the mouse. Adapted from Müller et al. (2005) and Engel et al. (2006).

Figure 2

Hearing performance of the control and of the trauma groups before, directly after, and 28 days after noise trauma. Click ABR and pure-tone audiograms of the trauma groups before (day-2), directly after (day 0), and 28 days after the noise trauma (band noise, 8–16 kHz for 2 h) of either 100, 106, or 112 dB SPL and of the control group at days-2 and 28. (A) Click ABR thresholds (mean ± SD.) for the untreated control (ctrl) and the three trauma groups at days-2, 0, and 28. (B–D) Pure-tone audiograms showing mean f-ABR thresholds (±SD., only one direction is shown) recorded at day-2 (B), after trauma (day 0, C), and at day 28 (D). Data of the control group from day-2 are indicated in panel (C), connected by dashed lines for comparison. The permanent threshold shift with respect to the untreated control 4 weeks after trauma was evaluated for each trauma strength and frequency (D). Kruskal–Wallis test, with Bonferroni correction for multiple comparisons; (A, B) *p < 0.05; **p < 0.01; ***p < 0.001; statistical analysis for (C,D), see Table 1 for clarity. Number of animals: day-2: control, 7; 100 dB, 9; 106 dB, 20; 112 dB, 10; day 0: 100 dB, 8; 106 dB, 20; 112 dB, 10; day 28: control, 7; 100 dB, 8; 106 dB, 20; 112 dB, 10.

Figure 3

Effect of increasing strength of the noise trauma on growth functions of ABR wave I amplitudes 4 weeks after trauma. (A–D) Growth functions of the average peak-to-peak amplitudes of ABR wave I (±SD., only one direction is shown) for the control group (no trauma) and the groups exposed to 100 dB SPL, 106 dB SPL, and 112 dB SPL at 11 kHz (A), 16 kHz (B), 22 kHz (C), and 32 kHz (D) at day 28. Note that because of the occurrence of permanent threshold increases, less amplitude values could be extracted for higher frequencies and higher trauma levels. The minimum number of data required for presenting a mean value was set to three in each condition.

Figure 4

Sound level-dependent latencies of ABR wave I increase with trauma strength for the mid-to-high auditory frequencies. (A–D) Average level-dependent latencies of ABR wave I measured at the negative peak I (see Methods) are displayed for the unexposed control and the three trauma groups at day 28 after trauma for the frequencies 11 kHz (A), 16 kHz (B), 22 kHz (C), and 32 kHz (D). Note that because of the permanent threshold shift, fewer values were obtained for higher frequencies and higher trauma levels. The minimum number of data required for presenting a mean value was set to three in each condition.

Figure 5

Average loss of OHCs in high-frequency cochlear regions is small and is not related to trauma strength. (A) Example MIPs of confocal stacks of the midbasal part of a whole-mount organ of Corti from a 12-week-old mouse that had been exposed to a 2 h noise trauma from 8 to 16 kHz of 100 dB SPL 4 weeks after trauma. The number of OHC nuclei (DAPI-stained, blue) was counted in regions of nominally 10 OHCs per row in length (i.e., nominally 30 OHCs for the three OHC rows) as indicated by the dashed lines. Missing nuclei are indicated by red circles. (B) Percentage of OHC loss in regions of 3 × 10 OHCs (mean + SD) for the control and the trauma groups as a function of cochlear location. Numbers below the bars indicate the number of regions of OHC counts providing the space for 30 OHCs obtained from 4 different mice. Scale bar: 50 μm.

Figure 6

Effects of noise traumata of 100, 106, and 112 dB SPL on IHC synapses 4 weeks after trauma as evaluated by immunolabeling for CtBP2 and Homer1. MIPs of confocal stacks of whole-mount organs of Corti showing the basolateral pole of eight IHCs each immunolabeled for presynaptic ribbons (CtBP2, green) and postsynaptic glutamate receptors using Homer1 (magenta). (A1) Specimen of an unexposed 12-week-old control mouse. (B1–D1) Example specimen of mice that had been exposed to a noise trauma at 8 weeks of age of 100 dB SPL (B1), 106 dB SPL (C1), and 112 dB SPL (D1) at day 28 after trauma. Unpaired ribbons are indicated by open arrowheads and unpaired postsynaptic spots by closed arrowheads. Nuclei are stained in blue with DAPI; IHC nuclei are additionally indicated by white stars. An outline of one IHC is indicated by a white dotted line in each panel. The thumbnails to the right (A2–A4,B2–B4,C2–C4,D2–D4) show enlargements of paired synapses or unpaired synaptic components selected from the respective main panels (A1–D1). Scale bars: 10 μm in main panels, 1 μm for thumbnails.

Figure 7

Noise trauma strongly reduces the number of presynaptic ribbons but less so the number of postsynaptic Homer1 clusters in the mid-to high-frequency range 4 weeks after trauma. (A,B) Number of presynaptic ribbons per IHC (A) and number of postsynaptic Homer1 clusters per IHC (B) as a function of trauma strength and cochlear location. Data are given as mean ± SD; numbers of regions comprising usually 8 IHCs: control (no trauma): apical 19, medial 37, midbasal 82, basal 12; 100 dB SPL: apical 8, medial 10, midbasal 15, basal 4; 106 dB SPL: apical 18, medial 33, midbasal 65, basal 4; 112 dB SPL: apical 46, medial 51, midbasal 48, basal 10, from 4 mice in total. One-way ANOVA for apical ribbons and medial postsynapses, Kruskal–Wallis test otherwise, with Bonferroni correction for multiple comparisons; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 8

Noise trauma increases the sizes of both presynaptic ribbons and Homer1 clusters 4 weeks after trauma. (A,B) Mean area ± SD of presynaptic ribbons (A) and mean area ± SD of postsynaptic Homer1 clusters (B) Figure 6 as a function of trauma strength and cochlear location. Data are given as mean ± SD; numbers of regions comprising usually 8 IHCs: control (no trauma): apical 19, medial 37, midbasal 82, basal 12; 100 dB SPL: apical 8, medial 10, midbasal 15, basal 4; 106 dB SPL: apical 18, medial 33, midbasal 65, basal 4; 112 dB SPL: apical 46, medial 51, midbasal 48, basal 10, from 4 mice in total. Kruskal–Wallis test with Bonferroni correction for multiple comparisons, * p < 0.05; ** p < 0.01; *** p < 0.001.

Figure 9

Postsynapses are less vulnerable to noise trauma than presynaptic ribbons (A,B) Box-and-whisker plots of the number of unpaired ribbons per IHC (A) and of orphan postsynapses per IHC (Homer1 clusters, B) as a function of cochlear location 4 weeks after trauma of the respective strength. Regions with usually 8 IHCs that went into the analysis of Figures 7, 8 were further analyzed for the number of unpaired pre-and postsynapses. Age-matched groups that did not receive any trauma, but were tested for their hearing function at days -2 and 28, served as controls. One-way ANOVA for unpaired ribbons of the midbasal and basal region; Kruskal–Wallis test otherwise, Bonferroni correction for multiple comparisons, * p < 0.05; ** p < 0.01; *** p < 0.001.