Absence of oncomodulin increases susceptibility to noise-induced outer hair cell death and alters mitochondrial morphology

Abstract

Cochlear outer hair cells (OHCs) play a fundamental role in the hearing sensitivity and frequency selectivity of mammalian hearing and are especially vulnerable to noise-induced damage. The OHCs depend on Ca²⁺ homeostasis, which is a balance between Ca²⁺ influx and extrusion, as well as Ca²⁺ buffering by proteins and organelles. Alterations in OHC Ca²⁺ homeostasis is not only an immediate response to noise, but also associated with impaired auditory function. However, there is little known about the contribution of Ca²⁺ buffering proteins and organelles to the vulnerability of OHCs to noise. In this study, we used a knockout (KO) mouse model where oncomodulin (Ocm), the major Ca²⁺ binding protein preferentially expressed in OHCs, is deleted. We show that Ocm KO mice were more susceptible to noise induced hearing loss compared to wildtype (WT) mice. Following noise exposure (106 dB SPL, 2 h), Ocm KO mice had higher threshold shifts and increased OHC loss and TUNEL staining, compared to age-matched WT mice. Mitochondrial morphology was significantly altered in Ocm KO OHCs compared to WT OHCs. Before noise exposure, Ocm KO OHCs showed decreased mitochondrial abundance, volume, and branching compared to WT OHCs, as measured by immunocytochemical staining of outer mitochondrial membrane protein, TOM20. Following noise exposure, mitochondrial proteins were barely visible in Ocm KO OHCs. Using a mammalian cell culture model of prolonged cytosolic Ca²⁺ overload, we show that OCM has protective effects against changes in mitochondrial morphology and apoptosis. These experiments suggest that disruption of Ca²⁺ buffering leads to an increase in noise vulnerability and mitochondrial-associated changes in OHCs.

Keywords: Ca2+; Ca2+ buffer; cell death; cochlea; mitochondria; noise-induced hearing loss; oncomodulin; outer hair cells.

Figure 1

Ocm KO mice have higher ABR threshold levels compared to WT mice post-noise exposure. (A) Timeline of noise-exposure experiments. Three-week-old Ocm WT and Ocm KO mice were exposed to broadband noise at 106 dB SPL for 2 h. DPOAE and ABR measurements were taken prior to, and 2 days post-noise exposure (dpn). (B,C) ABR threshold levels at 8, 16 and 32 kHz pre-and 2 days post-noise exposure in Ocm WT (B) and Ocm KO (C) mice. (D,E) ABR threshold levels remain unchanged in control WT and KO mice. (F) Comparison of ABR threshold levels at 8, 16, and 32 kHz in Ocm WT and Ocm KO mice 2 days post-noise exposure (Two-way ANOVA with Bonferroni’s post hoc). (G) Average ABR threshold shift across all tested frequencies in control and noise exposed Ocm WT and Ocm KO mice. n ≥ 3 per group. All plotted values represent mean ± SEM. Asterisks represent statistical significance with: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p < 0.0001, ns, not significant.

Figure 2

DPOAE threshold levels and I/O functions are higher in Ocm KO mice compared to WT mice post-noise exposure. (A–E) DPOAE threshold levels at 5, 8, 11, 16, 22, 32, and 45 kHz pre-and post-noise exposure in Ocm WT (A) and Ocm KO (B) mice (C) Comparison of DPOAE threshold levels in Ocm WT and Ocm KO 2 days post-noise exposure. (D,E) DPOAE threshold levels remain unchanged in control WT and KO mice. (F) Average DPOAE threshold shift at 5, 8, 11, 16, 22, 32, and 45 kHz 2 days post-noise exposure in Ocm WT and Ocm KO mice. (G) Average DPOAE threshold shift across all tested frequencies pre-and 2 days post-noise exposure in Ocm WT and Ocm KO mice. (H,I) DPOAE I/O at 16 kHz pre-and post-noise exposure in Ocm WT (H) and Ocm KO (I) mice n = ≥3 per group. All plotted values represent mean ± SEM. Asterisks represent statistical significance with: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p < 0.0001, ns, not significant.

Figure 3

Ocm KO mice are more prone to outer hair cell loss compared to Ocm WT mice. (A–D) Representative images of cochlear whole mounts in the apical (~8–12 kHz) and basal (~28–48 kHz) frequency regions. Orange = Myo7a, green = Phalloidin, blue = Hoechst. 40× maximum intensity projections (MIPs). White asterisks denote missing OHCs. Scale bar = 10 μm. Shown are images from Ocm WT control (A), Ocm WT noise (B), Ocm KO control (C), and Ocm KO noise exposed mice (D). (E) Quantification of OHC loss in Ocm WT and KO control and noise exposed mice. n ≥ 5 per group. All plotted values represent mean ± SEM. Asterisks represent statistical significance with: *p ≤ 0.05, ***p ≤ 0.001; ns, not significant.

Figure 4

TUNEL staining is increased in noise-exposed Ocm KO mice compared to WT mice. (A,B) Representative MIPs of confocal z-stacks from control and noise exposed Ocm WT and KO basal OHCs. Cochlea were harvested 1 day post noise exposure. Three rows of OHCs are shown. White = TUNEL, blue = Hoechst, orange = Myo7a, green = Phalloidin. Scale bar = 10 μm. (C) Bar graphs represent the number of TUNEL positive surfaces per OHC region. n ≥ 5 per group. All plotted values represent mean ± SEM. Asterisks represent statistical significance with: ****p < 0.0001; ns, not significant.

Figure 5

TOM20 and COXIV show different expression patterns in apical and basal OHCs of unexposed and noise exposed Ocm WT and KO mice. (A–C) state Representative MIPs of basal OHCs are shown. (A) TOM20 immunostaining in OHCs. red = TOM20, blue = Hoechst. Scale bar = 2 μm. (B) Quantification of TOM20 fluorescence intensity. n ≥ 3 per group. (C) COXIV immunostaining in OHCs. orange = COXIV, blue = Hoechst. Scale bar = 2 μm. (D) Quantification of COXIV fluorescence intensity. n = 2–4 per group. All plotted values represent mean ± SEM. Asterisks represent statistical significance with: *p ≤ 0.05, ****p < 0.0001; ns, not significant.

Figure 6

Mitochondrial morphology is altered in control and noise-exposed Ocm KO OHCs. (A,B) Apical and basal OHCs from mid-modiolar sections of unexposed and noise exposed Ocm WT and KO cochleae harvested 2 dpn. Representative MIPs are shown. Red = TOM20, blue = Hoechst. Scale bar = 5 μm. All images were taken using the same imaging parameters. However, brightness/contrast display settings were optimized for each representative image shown here for visualization purposes. Dotted outlines with asterisks denote missing OHCs. Arrows highlight areas of mitochondrial aggregation along the lateral membrane (short arrows) and in the perinuclear region (long arrows) in Ocm WT unexposed and noise exposed, respectively. (C) Bar graph showing the number of mitochondria/OHC volume (μm³), (D) mitochondrial volume (μm³)/OHC volume (μm³), and (E) the mean number of branches per mitochondria in apical and basal OHCs of control and noise exposed Ocm WT and KO OHCs. n ≥ 3 per group. All plotted values represent mean ± SEM. Asterisks represent statistical significance with: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p < 0.0001; ns, not significant.

Figure 7

HEK293T cells expressing OCM are less vulnerable to Ca²⁺-related apoptotic inducers. HEK293T cells were transiently transfected with GFP (control) or OCM-GFP and incubated with thapsigargin (Tg, 1 μM) for 48 h. (A) Representative Airyscan images of control (untreated) and Tg treated GFP and OCM-GFP expressing cells. Magenta = TOM20, blue = Hoechst. Scale bar = 5 μm. Mitochondrial morphology was measured by calculating (B) the mean (integrated density) TOM20 intensity, (C) number of mitochondrial branches, and (D) the average mitochondrial size (μm²). Data is shown as a floating bar graph (min to max) with a line at the mean. n = 3 technical replicates. A TUNEL assay was performed using staurosporine treated (STS, 24 h) or untreated HEK293T cells transiently expressing mCh (control) or OCM-mCh. (E) Shown are representative confocal images from these experiments. TUNEL = white, blue = Hoechst. Scale bar = 20 μm. (F) The bar graph shows mean number of TUNEL positive cells over total mCh positive cells. n = 3 experimental replicates. All plotted values represent mean ± SEM. Asterisks represent statistical significance with: *p ≤ 0.05; ns, not significant.

Figure 8

Ocm KO OHCs are predisposed to decreased mitochondrial abundance. The illustration represents our proposed model of altered mitochondrial dynamics in Ocm WT and KO OHCs exposed to noise. OCM is shown in light blue. Arrows indicate the movement of Ca²⁺ inside the OHCs. (Left) In unexposed Ocm WT OHCs, mitochondria display a characteristic organization pattern, clustering around the lateral plasma membrane and under the subsurface cisternae. (Middle) Following noise exposure in Ocm WT OHCs, we found that mitochondria aggregate along the perinuclear region. (Right) Ocm KO OHC mitochondria are predisposed to decreased mitochondrial volume, abundance and lack organization. Following noise exposure, mitochondria of Ocm KO OHCs become sparse and fragmented, leaving the cell vulnerable to OHC death.