All experiments were done in compliance with National Institutes of Health and institutional animal care guidelines and were approved by the Institutional Animal Care and Use Committee of Baylor University. Ocm^–/– (Ocm KO) mice were generated from spontaneous germline transmission of the KO allele from the parental line, Actb^Cre;Ocm^flox/flox on the C57Bl/6J background (Tong et al., 2016). The original C57 Ocm KO mice were backcrossed onto the CBA/CaJ background for 10 generations. At each generation, mice were genotyped for the Cdh23-ahl locus and only animals with the CBA/CaJ Cdh23 allele were bred following procedures used by Liberman and colleagues (Liu et al., 2007). The Cdh23 WT or ahl gene region was amplified using primers that would generate a 360 bp product. Cdh23^WT/ahl For: 5′-GATCAAGACAAGACCAGACCTCTGTC-3′. Rev: 5′- GAGCTACCAGGAACAGCTTGGGCCTG-3′. The PCR product was sequenced to confirm lack of Cdh23^ahl (Supplementary Figures 1A,B). Confirmation of congenicity was done by whole genome scan (Jackson Laboratories, Bar Harbor, ME, United States). Our C57 Ocm KO mice were crossed with B6.129P2-Pvalbtm1Swal/J on the C57Bl/6J background by Jackson Laboratories (Bar Harbor, ME, United States). Homozygous CBA Ocm KO and C57 Ocm KO and WT littermates from each sex were used for all experiments.

Tissues were incubated with primary antibody overnight at 37°C. Primary antibodies used: goat anti-OCM, (1:500, Santa Cruz sc-7446, AB_2267583), rabbit anti-prestin [1:5,000, kindly provided by Robert Fettiplace (Furness et al., 2008)], goat anti-choline acetyltransferase [ChAT, 1:500, Millipore, AB144P, AB 2079751, (Misgeld et al., 2002)]. All primary antibodies were labeled with species appropriate Alexa Fluor (ThermoFisher) or Northern Lights (R&D Systems) secondary antibodies for 2 h at 37°C.

For measurement of auditory brainstem responses (ABRs) and distortion product otoacoustic emissions (DPOAEs), adult mice were anesthetized with xylazine (20 mg/kg, i.p.) and ketamine (100 mg/kg, i.p.). Acoustic stimuli were delivered using a custom acoustic assembly previously described by Maison et al. (2012). Briefly, two electrostatic earphones (EC-1, Tucker Davis Technologies) were used to generate primary tones and a Knowles miniature microphone (EK-3103) was used to record ear-canal sound pressure. Stimuli were generated digitally with 4s sampling. Ear-canal sound pressure and electrode voltage were amplified and digitally sampled at 20s for analysis of response amplitudes. Both outputs and inputs were processed with a digital I-O board (National Instruments PXI-4461). For measurement of ABRs, needle electrodes were inserted at vertex and pinna, with a ground electrode near the tail. ABR potentials were evoked with 5 ms tone pips (0.5 ms rise-fall with a cos2 onset, delivered at 35/s). The response was amplified (10,000), filtered (100 Hz–3 kHz), digitized, and averaged in a LabVIEW-driven data-acquisition system. Sound level was raised in 10 dB steps from 10 dB below threshold up to 80 dB sound pressure level (SPL). At each sound level, 1,024 responses were averaged (with stimulus polarity alternated), using an “artifact reject” whereby response waveforms were discarded when peak-to-peak amplitude exceeded 15V (e.g., electro-cardiogram or myogenic potentials). Threshold was defined as the lowest SPL level at which wave-I peak could be identified, usually corresponding to the level step just below that at which the peak-to-peak response amplitude rose significantly above the noise floor. For amplitude vs. level functions, the wave-I peak was identified by visual inspection at each sound level and the peak-to-peak amplitude computed. For measurement of DPOAEs at 2f1 – f2, the primary tones were set so that the frequency ratio, (f2/f1), was 1.2 and so that f2 level was 10 dB below f1 level. For each f2/f1primary pair, levels were swept in 10 dB steps from 20 dB SPL to 80 dB SPL (for f2). At each level, both waveform and spectral averaging were used to increase the signal-to-noise ratio of the recorded ear-canal sound pressure, and the amplitude of the DPOAE at 2f1 – f2 was extracted from the averaged spectra, along with the noise floor at nearby points in the spectrum. Iso-response curves were interpolated from plots of DPOAE amplitude vs. sound level. Threshold was defined as the f2 level required to produce a DPOAE at 0 dB SPL. We used a total of 25 female and 18 male WT and 13 female and 13 male KO CBA mice, and a total of 40 female and 38 male WT and 17 female and 25 male KO C57 mice. Right ears were used for all hearing tests. For both background strains, variances between female and male mice were negligible across the ages tested.

For histological analysis and immunocytochemistry, anesthetized mice (Euthasol, 150 mg/kg, i.p.) were perfused transcardially with 4% paraformaldehyde (PFA) in 0.1M phosphate buffer (PBS) followed by PBS wash. After removal, cochleae scalae were flushed with PFA then rotated in fixative overnight at 4°C. Cochleae were decalcified in 0.1M EDTA for 3–5 days at 4°C with rotation. Cochleae were prepared as cochlear whole mounts of approximately 6 pieces of sensory epithelium or as 100 μm mid-modiolar sections embedded in gelatin-agarose solution (Simmons et al., 2010; Maison et al., 2012). Cochlear pieces were suspended in 30% sucrose for 30–60 min at room temperature with gentle shaking, frozen at −80°C for 30 min, then thawed for 30min at 37°C. Pieces were washed 3 times in PBS, then blocked in 5% normal horse serum (NHS) for 2 h at room temperature. Images were acquired using the LSM800 microscope (Zeiss) using a high-resolution, oil-immersion objective (Plan-Apochromat 63x, 1.4 NA). Cohorts of samples were immunostained at the same time and imaged under the same optical conditions to allow direct comparisons.

Low-power images of each microdissected piece of surface preparations were obtained with a 10x air objective (N.A. 0.3) on a LSM800 confocal microscope. Cytocochleograms were constructed from these images by tracing the cochlear spiral and superimposing hash marks using a custom ImageJ Measure Line plugin from Eaton-Peabody.² The plugin superimposed frequency correlates on the microdissected spiral image by application of the cochlear frequency map for mice. Organ of Corti was imaged at the 8, 16, and 32 kHz regions with the LSM800 confocal microscope using a 63x oil immersion objective (N.A. 1.4). At each of the desired locations, 3 adjacent microscopic fields (9–12 IHCs per field) were imaged with a 4-channel z-stack spanning the height of the hair cells to capture the stereocilia, nuclei and synaptic junctions. Hair cells were only counted if they had a nucleus or hair bundle/cuticular plate present. ChAT-labeled efferent clusters of terminal swellings were counted in equally spaced distance regions using a LSM5 confocal microscope using a 40x oil immersion objective (N.A. 1.2). In each cochlear location, we acquired two adjacent z-stacks (each spanning 112 μm of the cochlear spiral), taking care to span the entire region containing ChAT-positive terminals along the z-axis. Because of the difficulty counting ChAT-labeled terminals in older animals, we resorted to counts only of clusters of labeled efferent terminals. Counts were performed by a blinded observer.

For OCM subcellular distribution, quantification was performed in ImageJ using a region of interest (ROI) from the nucleus compared with a similar ROI in the cytoplasm. After background removal, cytoplasmic OCM fluorescence intensity was divided by nuclear OCM fluorescence intensity. 2-5 cochlea from different animals were analyzed per condition.

All statistical analysis were performed in Prism (v9.1x GraphPad Software). All data were tested for homogeneity of variance. For multiple comparisons, Brown-Forsythe and Welch ANOVA test was applied when SDs are significantly different (p < 0.05), followed by Dunnett’s multiple comparisons test. Mann-Whitney U-test was applied when normal distribution cannot be assumed. Mean values are quoted in text and figures as means ± SEM (DPOAE and ABR measurement) and means ± SD (missing OHC/ChAT counting and OCM distribution). p < 0.05 was selected as the criterion for statistical significance.

We previously showed that targeted deletion of Ocm in C57Bl/6j mice have elevated hearing thresholds by 3 – 4 mo (Tong et al., 2016). We wanted to compare the contribution of OCM to auditory function in mice with different genetic backgrounds, specifically, C57 mice with short hearing lifespans to CBA mice with longer hearing spans. We transferred the Ocm KO allele from the original C57 mouse line to the CBA background (see section “Materials and Methods”). To test hearing function, ABR waveforms of C57 KO (Figures 1A–D) and CBA KO (Figures 1E–H) mice at early adult ages up to the ages where ABR potential responses were lost were compared with age-matched WT C57 and CBA mice. ABR potentials at 8 and 12/16 kHz frequencies are similar for WT and KO mice at 2 mo in both backgrounds (Figures 1A,B,E,F). Ocm KOs showed a lack of response under these same conditions as early as 4 mo in C57 mice (Figures 1C,D) and 7 mo in CBA mice (Figures 1G,H), whereas the WT animals demonstrated robust ABR potentials at 8 mo in both backgrounds. C57 Ocm KO ABR thresholds shifted 20 - 60 dB at middle and high frequencies between 1 and 3 mo of age (Tong et al., 2016). Tong et al. (2016) showed a progressive elevation of ABR thresholds in the C57 KO. In CBA mice, we measured ABR thresholds at 3 different ages using 8, 16, and 32 kHz using the appearance of the wave-I peak as the criteria for threshold (Figures 1I,J). WT and KO animals showed similar thresholds at 2 mo, but KO animals showed threshold shifts by 5 mo, particularly at higher frequencies, and were essentially deaf by 7 mo (Figure 1I). In the CBA KO, 2 - 7 mo DPOAE thresholds were significantly different from each other at 16 kHz (Welch’s ANOVA: p = 0.0003; Dunnett’s post-test: p = 0.0002 for 2 – 7 mo). In contrast, WT animals demonstrate similar ABR thresholds at low and middle frequencies from 2 to 7 mo, and only show substantial threshold shifts in the high frequencies at 5 - 7 mo and middle frequencies at a year (Figure 1J). At 16 kHz in the CBA WT, there was no significant difference in ABR thresholds up to 7 mo (Welch’s ANOVA: p = 0.104, followed by Dunnett’s test).

To confirm that germ-line transmission of the Ocm KO in C57Bl/6J mice did not change the original hearing loss phenotype described by Tong et al. (2016), we measured DPOAE thresholds in age-matched KO and WT C57 and CBA mice over the course of their hearing span (Figures 2A–F). The C57 WT control mice showed little threshold shifts between 1 and 5 mo. The C57 KO mice show a progressive elevation of DPOAE thresholds from 1 to 6 mo (Figure 2A). At 16 kHz, there were significant differences in DPOAE thresholds between 1 – 3 mo and 3 – 6 mo (Welch’s ANOVA: p < 0.0001; Dunnett’s test: p = 0.009 for 1 – 3 mo, p = 0.002 for 3 – 6 mo). In the C57 KO between 3 and 6 mo, we observed sizable threshold shifts in all but the lowest frequencies (8 - 32 kHz), while in the WT, sizable threshold shifts were observed between 5 and 16 mo (Figures 2A,B). The largest threshold shifts in the C57 KO were in the middle frequencies. Although C57 KO mice were at or near the limit of DPOAE threshold detection by 6 mo, the C57 WT mice had measurable DPOAEs in most frequencies at 16 mo. In the C57 WT, there were no significant differences in 16 kHz thresholds of 1, 4, and 5 mo mice (Dunnett’s test). There was a significant difference in thresholds between 1 and 16 mo at 16 kHz (p = 0.009, Dunnett’s test). Since OCM is predominantly expressed in OHC of the organ of Corti at these ages, these results confirm that OHC defects are primarily responsible for the differences in ABR potentials between the Ocm KO early onset hearing loss and the WT.

Similar to ABR threshold responses (Figure 1I), CBA KO animals showed middle- to high-frequency DPOAE threshold shifts from 2 to 5 mo (Figure 2C). The KOs also showed nearly maximum threshold shifts at all frequencies from 7 to 12 mo. At 16 kHz, DPOAE thresholds were not statistically different between 2 and 5 mo but were significantly different between 2 and 7 mo and 2 – 12 mo (Welch’s ANOVA: p < 0.0001; Dunnett’s post-test: p < 0.0001 for 2 – 7 mo, p < 0.0001 for 2 – 12 mo). In contrast, WT mice maintained healthy threshold levels at least up to 16 mo of age and produced measurable DPOAE threshold responses in most frequencies by 28 mo (Figure 2D).

At 16 kHz in CBA mice, only DPOAE thresholds comparisons between 2 and 28 mo were significantly different (p < 0.0005, Dunnett’s test). Interestingly, young adult KO mice (2 mo) show enhanced DPOAE input-output functions compared with WT animals suggesting enhanced electromotility at least for higher frequencies. We compared DPOAE input-output functions at 32 kHz across ages (Figures 2E,F). The KO animals required less signal (dB SPL) to produce a DPOAE output that crosses threshold (Figure 2E). This efficiency is lost by 5 mo, and by 12 mo the KO input-output responses do not cross threshold. WT animals lose their input-output efficiency between 5 and 12 mo, requiring higher dB SPL input to cross threshold than at younger ages (Figure 2F). Overall, the only significant difference in amplitudes in the CBA WT were between the 2 mo and 28 mo animals (p < 0.05).

Taken together, Ocm KO reduces the hearing lifetime (hearing span) of mice regardless of genetic background (Figure 3). Though CBA mice are a well-established mouse model that hears throughout most its life, Ocm deletion in OHCs reduces the hearing span to less than half of the time of WT counterparts with severe hearing dysfunction in the first 1/3 of life (Figure 3, 2nd bar). Deletion of Ocm similarly reduces the hearing span of C57 mice, but at an earlier age. When comparing hearing thresholds between the two genetic backgrounds, loss of OCM in the CBA background recapitulates a similar hearing phenotype of the C57 background (Figure 3, middle bars). These data demonstrate that OCM is an essential contributor to hearing health by delaying progressive hearing loss as a function of age.

Since Ocm KO mice appear to accelerate early progressive hearing loss, we wanted to compare aging OHC phenotypes in the WT and KO mice across genetic backgrounds. In CBA animals, WT animals manifest substantial OHC loss after 1 year (Figures 4A–C,I). Interestingly, OCM expression is maintained even when there is substantial OHC loss. However, the subcellular localization and intensity of OCM immunostaining varies with age and frequency region across genetic background. Surface preparations were collected, stained and imaged under the same optical conditions to visualize OCM redistribution and apparent expression by age. In CBA WT mice, we find OCM immunoreactivity is more intense in the nucleus than in the cytosol of 2-month-old animals in the 8, 16, and 32 kHz regions (Figure 4 and Supplementary Figures 1C–F, 2A,B). OCM expression becomes more cytoplasmic with age with a few cells with intensely labeled OCM in the cytosol and lateral membrane at 12 mo (Figures 4A–C, 12mo^∗). This phenomenon is more extreme in basal regions and at older ages (Supplementary Figures 2A,B). Additionally, the WT OHC damage and/or loss was more pronounced in apical regions beginning at 1 year and becoming more pronounced by 30 months (Figures 4, 5). Since we know that WT animals have poorer hearing thresholds from 12 to 30 mo, it is possible that the intense OCM expression profile of older animals (Figure 4 and Supplementary Figures 2A,B) is a feature of dysfunctional/aging OHCs. Older animals still maintain expression of the OHC specific motor protein prestin, responsible for OHC electromotive function, though this too may be dysfunctional (Figures 4D,G, white). WT animals maintain robust OHC numbers throughout the first year of life and are only missing approximately 50% of the OHCs along the entire cochlear spiral by 30 mo (Figure 4I, top). By contrast, KO animals have lost the majority of OHCs by 12 mo (Figure 4I, top). Similar to our previous report (Tong et al., 2016), WT OHC numbers were maintained through 5 mo of age in the C57 background (Figures 4F–I) unlike Ocm KO animals (Figures 4H,I, bottom). Prestin was also abundant in the remaining OHCs of aged WT mice (Figures 4F,G) and KO OHCs at 5 - 6 mo (Figures 4D,H). Similar to the CBA WTs, OCM is expressed more intensely in the remaining C57 OHCs with age (Figure 4F) and across all frequencies (data not shown).

Synaptic dysfunction, rewiring, and OHC loss are common phenotypes of the aging cochlea. Activation of cholinergic medial olivocochlear (MOC) efferent axons protects OHCs from excessive noise and Ca²⁺ by modulating OHCs directly and altering their Ca²⁺-sensitive motility (Simmons, 2002; Maison et al., 2003; Frolenkov, 2006; Boero et al., 2020). Loss of MOC efferents could enhance the dysfunctional state of aging OHCs and thus result in increased hearing thresholds (Fu et al., 2010). Therefore, we looked for the presence and absence of ChAT-labeled efferent terminal clusters in aging WT and KO animals. In both CBA and C57 mice, we co-labeled surface preparations (Figures 5A,B,E,F) and mid-modiolar cochlear sections (Figures 5C,D, CBA only) with DAPI to stain the nuclei and ChAT to label efferent fibers and terminal clusters. In CBA WT and KO mice, ChAT-labeled clusters were abundant at 2 – 4 mo. However, in older WT and KO animals, we observed many instances of OHC loss and absence of ChAT labeling on remaining OHCs. Apical OHCs lose ChAT clusters by 12 mo in WT animals (Figure 5A#). At 30 mo in the 8 kHz region, the few remaining OHCs do not have ChAT-labeled clusters (Figure 5# and Supplementary Figure 2E). The KOs show some hair cell loss and efferent loss as early as 2 mo in the 32 kHz region, but the most dramatic efferent loss occurs in both apical and basal regions by 12 mo (Figure 5B^∗#). Both KO OHC and efferent loss appear similar to recent data collected from young and old human temporal bones (Wu et al., 2019; Wu et al., 2020). The WT OHC damage and/or loss was more pronounced in apical regions while the KO OHC loss was pronounced in basal regions.

In C57 mice, the pattern of OHC and efferent loss was more pronounced. The WT apical OHC and efferent loss was evident as early as 5 mo and by 15 mo both apical and basal OHCs and efferent clusters were mostly absent (Figure 5E). In the KO at 2 mo, we observe apical efferent loss and basal OHC loss (Figure 5F). At 5 mo, the majority of basal OHCs were absent and there were also no efferents clusters. Apical regions had less OHC and efferent loss. WT mice showed similar apical and basal OHC and efferent loss while the KO showed more OHC and efferent loss in basal regions. Though the WT animals showed signs of efferent loss in the apex compared with the KO at 5 – 7 mo, there was no significant difference in their ChAT numbers at either region (Supplementary Figure 2C). The greatest significant difference at these ages was in the basal regions of the KO (Supplementary Figure 2D).

We analyzed the loss of OHCs and ChAT-labeled efferent connections in relation to specific DPOAE frequency regions in CBA KO animals (Figure 6). We co-labeled surface preparations with antibodies to prestin and ChAT (Figures 6A–F). Single animal (Figures 6A–H) and cohort (Figures 6I,J) data are shown. At 3mo, we found robust DPOAE responses at 16 and 22 kHz (Figure 6H, purple) with the majority of hair cells present except in the extreme base (Figure 6G, purple). By 7 mo, responses were near the measurement ceiling at these frequency regions (Figure 6H, green) and yet there are still hair cells present, particularly at 16 kHz regions (Figure 6G, green). Indeed, DPOAE responses were elevated near maximal sound levels across the entire cochlea at 7 mo, but the majority of hair cells in all three OHC rows are present in the 4-16 kHz regions and prestin was present in the remaining cells. ChAT was also present in the majority of the remaining cells, though not all (Figures 6C,D#). Efferent loss was progressive, with the greatest loss occurring by 7 mo in the middle and high frequency regions (Figure 6I). We also plotted the percent OHC present as a function of cochlear frequency (Figure 6J). Middle frequency regions of the cochlea show the least efferent loss. The greatest loss of efferent clusters occurs in the regions of greatest OHC loss. Apical regions of the cochlea also show a loss of efferent clusters with little OHC loss. Taken together, there is a complex relationship between OHC loss, efferent terminal loss and hearing loss across cochlear frequency regions. In agreement with previous literature, high frequency regions are the most susceptible to trauma.

Oncomodulin plays a major role in the maintenance of OHC intracellular Ca²⁺ levels. OHCs have very high levels of OCM with estimates reaching as high as 2 – 4 mM suggesting that OCM provides the bulk of Ca²⁺ buffering in OHCs (Hackney et al., 2005). Ca²⁺ may be involved in regulating the motor capability underlying OHC cochlear amplification. The high concentration of OCM in OHCs, similar only to αPV in skeletal muscle, may protect against deleterious consequences of Ca²⁺ loading after acoustic overstimulation. Thus, without OCM to buffer the Ca²⁺ levels in OHCs, cytoplasmic Ca²⁺ regulation could be severely disrupted and lead to OHC dysfunction and loss if nothing effectively compensates for its absence (Climer et al., 2019). To date, unlike previous genetic disruption studies of other mobile Ca²⁺ buffers that show little, if any, impact on hearing, OCM is the only known CaBP for which targeted deletion causes progressive hearing loss (Tong et al., 2016), suggesting the essential role of Ca²⁺ signaling in maintaining hearing health. Within the inner ear, targeted deletion of any of the other major EF-hand CaBPs (e.g., calbindin D28k, calretinin and αPV) show no hearing loss phenotype (Airaksinen et al., 2000; Schwaller et al., 2002). Using a targeted deletion of Ocm, Tong et al. (2016) demonstrated that the lack of OCM led to progressive cochlear dysfunction beginning after 1 mo and resulted in ABR threshold shifts and loss of DPOAEs by 4 – 5 mo of age. Functionally, the absence of OCM mimicked an accelerated aging or noise damage process. The presence of normal ABR and DPOAE thresholds in the Ocm KO at 1 mo suggested that OCM was not essential for the development of cochlear function, but critically protects OHCs from damage in the adult ear. Similarly in the present study, we used a C57 mouse with germ-line deletion of OCM. The majority of C57 KO mice had significantly higher ABR and DPOAE threshold responses at 5 – 7 mo across all frequencies, but the absence of responses could occur as early as 4 mo at some frequencies. Compared to the WT mice at 5 mo, these Ocm KO mice also exhibited increased OHC loss especially in basal regions and altered OHC prestin immunoreactivity. In fact, the Ocm KO mice had hearing thresholds, OHC loss and prestin immunoreactivity comparable to 15 mo old WT C57 mice.

The two most utilized mouse models for studies of ARHL are the C57Bl/6J and CBA/CaJ strains. The C57 strain has a rapid, high-frequency hearing loss and loses most of its high-frequency hearing during the first 12 mo. In contrast, the CBA strain loses its hearing slowly with age. At early adult ages from 1 to 2 mo, functional measures of hearing for these two mouse strains are virtually indistinguishable. In the present study, the CBA Ocm mutant mice present with an early onset ARHL as seen in the C57 WT and KO mice. Our CBA Ocm mutants, beginning around 5 – 7 mo, and aged WT mice show similar hallmarks of ARHL: (1) progressive elevation of DPOAE thresholds, (2) loss of OHC efferent terminals, and (3) OHC loss. In Ocm KOs on a CBA background, our data show a progressive elevation of DPOAE thresholds over 12 mo. In contrast, WT hearing thresholds and response magnitudes change little from 2 to 12 mo. Similar to our observations in the C57 strain, young (2 mo) CBA Ocm KO mice had DPOAE thresholds nearly identical to WT controls at lower frequencies and somewhat enhanced at higher frequencies. In older CBA Ocm KO mice, maximum DPOAE threshold shifts occurred at frequencies 16 kHz and higher, which is within the region of greatest hearing sensitivity in these animals. Ocm KO mice had significantly higher thresholds at all measured frequencies between 5 and 7 mo. At 12 mo, DPOAE thresholds were only slightly elevated in WT control mice and were not measurable in the Ocm KO. In the present study, WT CBA mice had measurable DPOAEs up to 28 mo and we did not find significant ARHL until after 28 mo. These results suggest that Ocm deletion in CBA mice leads to an accelerated ARHL phenotype that is faster than the WT C57, but more delayed than in C57 KO mice. Although the specific temporal pattern of hearing loss differs, the results were similar to our previous report and validate that OCM expression is critical for either the maintenance of cochlear function and/or protecting OHCs from damage as previously hypothesized (Tong et al., 2016). Further, having a similar early progression of hearing loss across genetic strains argues that OCM possibly mediates sensitivity to ARHL. As illustrated in Figure 3, OCM deletion reduces the period of measurable hearing thresholds by over 50% in CBA and C57 mice.

In Ocm KO mice with high DPOAE thresholds, the majority of OHCs are still present. Although dysregulated Ca²⁺ signaling is a known contributor to apoptotic cell death, OHC loss occurs well after the loss of DPOAEs in the Ocm KO mouse. Since DPOAE responses are a direct measure of OHC function, elevated DPOAE thresholds suggest the remaining OHCs in the Ocm KO must be either “silent” or unresponsive to sounds. In CBA mutants, supra threshold DPOAE responses show rapid attenuation between 2 and 5 mo. At 32 kHz, 2 mo sensitivity to sounds deteriorated nearly 40 dB by 5 mo even though the majority of OHCs were still present. The lack of DPOAEs might suggest a decrease in expression of the motor protein prestin, as loss of prestin is correlated with elevated DPs (Liberman et al., 2002). However, this is unlikely due to the robust prestin immunoreactivity in these unresponsive OHCs. The lack of DPOAEs but presence of prestin protein would suggest that Ca²⁺ dysregulation makes OHCs dysfunctional in other ways such as in their biophysical characteristics. Recent studies of aged OHCs are consistent with the idea that OHCs undergo a period of dysfunction prior to OHC loss. In a study of aged OHCs from early onset and late onset ARHL mouse models, Jeng et al. (2021) suggested that age-related OHC dysfunction is not due to apoptosis. They primarily focused their studies on OHCs from the 9–12 kHz region. At 12 – 13 mo irrespective of whether OHCs were from early and late ARHL mice, they found OHCs had similar biophysical properties related to their peak current-voltage relationships, potassium currents, membrane capacitance, and membrane voltage measurements. With the exception of membrane voltage, they showed that these biophysical properties decreased similarly with age, being highest in the youngest animals. In addition to OHC loss, they found decreases in OHC size as indicated by decreases in membrane capacitance, OHC ribbons, and the mRNA expression of Slc26a5 and Ocm (Jeng et al., 2021). Most, if not all, of these features changed similarly, independent of the ARHL onset suggesting that they are general features of aging in OHCs, and not necessarily related to the OHC dysfunction observed in ARHL. Thus, it is possible that OHC-dependent ARHL is associated with disruption of cellular processes of transduction or motility. In the present study, we observed both prestin and OCM immunoreactivity in aged WT C57 and CBA OHCs (Figure 4). However, both proteins label more intensely with age (Figure 4), most likely either due to changes in cellular volume or expression. The seemingly enhanced expression of OCM could simply represent an artifact of shrunken OHC volume or it might convey that there is increased Ca²⁺ activity. In this regard, altered Ca²⁺ buffering could change the dynamics of organelles (e.g., mitochondria and endoplasmic reticulum), which store and release Ca²⁺ as well as alter ATP utilization. Although it is possible the overall mRNA expression of Ocm and Slc26a5 (prestin) in the cochlea decreases with age as reported by Jeng et al. (2020b), the ramifications for protein levels remain unclear. A more relevant question is whether their functions are somehow compromised. Jeng et al. (2020b) measured non-linear capacitance in aged OHCs and found that it decreased, but when normalized, the decrease appears independent from an ARHL phenotype. It is possible that smaller cell volumes due to an age-related shrinkage causes a relative increase in prestin protein levels. A tangential question is raised: though prestin is still present in aged OHCs, how much function does it retain? Overexpression of dysfunctional prestin in the mutant knock-in is more deadly to OHCs than complete lack of prestin suggesting features associated with prestin such as electromotility and axial stiffness could be contributors to OHC silencing during ARHL (Dallos et al., 1997; He et al., 2003; Keller et al., 2014). Our finding that OCM localization appears to undergo changes associated with age in WT mice might indicate age-associated changes in Ca²⁺ activity. It seems one of our challenges will be separating general features of aging in OHCs from those that lead to ARHL.

Sound- or electrically induced activity of the cholinergic MOC pathway suppresses OHC responses leading to reduced cochlear output and protection from acoustic injury (Rajan and Johnstone, 1989; Zheng et al., 1997; Maison and Liberman, 2000). Previous studies have shown that the function of the MOC system declines with age prior to OHC degeneration in both humans and mice (Sun and Kim, 1999; Kim et al., 2002; Jacobson et al., 2003; Zhu et al., 2007; Boero et al., 2020). In adult mice, OHCs are primarily innervated by the cholinergic MOC neurons, which modulate amplification of the cochlear partition (Guinan, 2006). At MOC – OHC synapses, release of acetylcholine (ACh) causes Ca²⁺ entry through α9α10 nicotinic ACh receptors (α9α10 nAChRs). The Ca²⁺ influx activates a hyperpolarizing current through Ca²⁺-activated small conductance K⁺ (SK2) channels on the OHC and causes OHCs to elongate (Wersinger and Fuchs, 2011; Guinan, 2018; Fuchs and Lauer, 2019). The overall effect of OHC elongation is believed to be protective (Frolenkov, 2006; Ciganovic et al., 2018). In both our Ocm KO and in aged WT controls, the presence of cholinergic efferent terminals contacting OHCs decreased in both low and high frequency regions with age (Figure 5). Jeng et al. (2021) also investigated whether the efferent innervation was retained in the aged apical cochlea of early and late onset ARHL mice. Although Jeng et al. (2020b) found no significant change in either the percentage of OHCs with SK2 immunoreactivity or the percentage of SK2 puncta juxtaposed to cholinergic terminals, they did note that some apical OHCs showed only efferent terminals or SK2 puncta. It has been known for some time that a reduction in OAEs correlates with impaired cochlear responses (de la Cruz et al., 1998; Zhu et al., 2007). If efferent terminals or SK2 puncta are lost, OHCs would lose that protection and have increased susceptibility to noise damage. MOC-mediated resistance to ARHL has been induced by enhancing α9α10 nAChR complexes on OHCs (Boero et al., 2020). Thus, the efferent loss or decline during aging could exacerbate ARHL, particularly in regard to discriminating between different stimuli under sundry conditions (Kobrina et al., 2020). Future studies should investigate the time course of efferent synaptic loss on the pre- and post-synaptic OHC junction.