Age-Dependent Hearing Impairment upon the Haploinsufficiency of Whrn

To investigate the impact of the haploinsufficiency of Whrn on hearing sensitivity throughout adulthood, mice with heterozygous deletion of long isoform of whirlin (Whrn+/–) were used. Using ABR recordings, we evaluated auditory transmission along the auditory pathway in Whrn+/- mice at different ages (from 1 to 6 months old). The ABRs using various sound stimuli (click, 4, 8, 16, and 32 kHz) were measured from WT and Whrn+/– mice. The intensity of sound stimuli (from 90 to 10 dB SPL for click stimuli and from 90 to 20 dB SPL for 4, 8, 16, and 32 kHz tone stimuli) at which discernable signal was no longer present was considered as a threshold (Fig. 1A). The threshold of individuals that had no detectable response at 90 decibels (dB SPL) were given a value of 100 dB SPL, and it was considered deaf. Although both WT and Whrn+/– mice have elevated thresholds across all frequencies along with ages from 1 to 6 months, Whrn+/– mice showed a stronger pattern of hearing impairment (Fig. 1B). In WT mice, the threshold at click stimuli was 34.83 ± 1.43 dB SPL (n = 59) at 1 month old and gradually increased to 60.00 ± 7.03 dB (n = 10) at 6 months old. In Whrn+/– mice, the threshold was 59.14 ± 4.47 dB SPL (n = 29) at 1 month old and reached to 76.82 ± 6.51 (n = 11) at 6 months old. Similarly, higher thresholds were observed across various tone of stimuli in Whrn+/– mice. The rate of age-related threshold increases was not significantly different between WT and Whrn+/– mice at low frequencies (click, 4 to 16 kHz tone). Notably, the progression of high-frequency hearing loss was greater in Whrn+/– mice at 32 kHz (pslope = 0.01, linear regression).

Fig. 1

Age-dependent hearing impairment Whrn heterozygous mice. A Representative traces of ABRs in response to click stimuli (90 to 30 dB SPL) from WT and Whrn+/– mice at 2 months old. B Summary of ABR threshold in response to click and tone stimuli (click, 4, 8, 16, and 32 kHz) in WT (black dots, n = 133 across 1–6 months age) and Whrn+/– mice (red dots, n = 135 across 1–6 months age). The linear regression was conducted at each group and were presented with 95% CI (black and red lines). C, D The proportion of deaf mice (not detecting any signal at 90 dB) among total recorded WT (C) and Whrn+/– mice (D) at 1–2 months, 3–4 months, 5–6 months old. Black bar indicates deaf mice portion, and those were presented with percentage

While analyzing the threshold change of ABR, the proportion of mice without responding across all frequencies of sound stimuli was gradually increased in Whrn+/– mice through ages. Therefore, considering the mice not responding to loud sound stimuli (90 dB SPL) as deaf, the proportion of deaf mice number was significantly increased in Whrn+/– mice comparing to WT mice (Fig. 1C, D). A small proportion of WT mice showed deafness at 1–2 months old (n = 46–84, click: 0.00%, 4 kHz: 7.32%, 8 kHz: 0.00%, 16 kHz: 2.17%, 32 kHz: 17.39%) across all stimuli. Although the threshold of WT mice increased at 5–6 months old, the proportion of mice with hearing impairment remained low (n = 26–27, click: 7.41%, 4 kHz: 19.23%, 8 kHz: 11.54%, 16 kHz: 19.23%, 32 kHz: 57.69%). However, the proportions of deafness in Whrn+/− mice were much higher than in WT even at 1–2 months old (n = 54–67, click: 13.43%, 4 kHz: 24.07%, 8 kHz: 18.18%, 16 kHz: 25.76%, 32 kHz: 68.18%) and distinctly increased with ages. At 5–6 months old, the hearing loss is more severe in Whrn+/− mice than WT showing almost complete loss of high-frequency hearing (n = 22, click: 36.36%, 4 kHz: 40.91%, 8 kHz: 40.91%, 16 kHz: 63.64%, 32 kHz: 95.46%, Fig. 1C, D). Taken together, heterozygous Whrn+/− mutant carriers showed hearing impairment with greater rate of deafness in early adulthood (at 1 month old) and significantly elevated thresholds of ABRs at all frequencies, indicating an early onset of age-related hearing loss.

The Effect of Whrn Haploinsufficiency on Hearing in Sex-Dependent Manner

Next, we examined whether sex influences the effect of whirlin haploinsufficiency on hearing sensitivity, because a large variation in ABR threshold within each group was observed (Fig. 1). There was a distinct sexual difference in age-dependent hearing loss between WT and Whrn+/− mice. Whrn+/− females exhibited hearing loss in young adulthood, whereas WT females showed age-related hearing loss in this period. In young adulthood females at 1–2 months old, Whrn+/− females already exhibited hearing loss with a higher threshold than age-matched WT females (two-way ANOVA, p < 0.0001; WT, n = 20–26; Whrn+/−, n = 34). This difference became smaller with age (two-way ANOVA, 3–4 months [WT, n = 5; Whrn+/−, n = 20], p = 0.0072; 5–6 months [WT, n = 13–14; Whrn+/−, n = 11], p = 0.1799), because WT females gradually lose their hearing sensitivity with aging (Fig. 2A).

Fig. 2

Sexual difference of Whrn haploinsufficiency effects. A, B The threshold of ABRs in responses to click and tone stimuli (4, 8, 16, and 32 kHz) in WT (black) and Whrn+/– mice at 1–2 months (top, WT = 14–19 males and 20–26 females, Whrn+/– = 24 males and 34 females), 3–4 months (middle, WT = 16 males and 5 females, Whrn+/– = 17 males and 20 females), and 5–6 months old (bottom, WT = 13 males and 13–14 females, Whrn+/– = 11 males and 11 females). Values from female (A) and male (B) mice are separately displayed. C The liner regression plotting of the ABR threshold against different ages (1–6 months old). Dots indicate individual values from females (left) and males (right) in WT (black, n = 48 males and 45 females) and Whrn+/– mice (red, n = 44 males and 57 females). Data were shown as mean ± s.e.m., and the regression model was plotted with 95% CI. To test statistical significancy between groups, two-way ANOVA and multiple comparison was used with Šidák correction. The significancy was presented as asterisk (n.s. = not significant, * < 0.05, ** < 0.01, *** < 0.001, **** < 0.0001)

Interestingly, the pattern of threshold changes in Whrn+/− males were different from females (Fig. 2B). At 1–2 months old, Whrn+/− males showed a significantly elevated threshold of ABRs for 32 kHz, indicating high-frequency hearing loss (multiple t-test, p = 0.0002). The ABR thresholds for other tones in Whrn+/− males were slightly higher than those from WT males at 1–2 months old (two-way ANOVA, p = 0.0156; WT, n = 14–19; Whrn+/−, n = 24), and the difference in ABR threshold between Whrn+/− and WT males became significantly increased with aging (two-way ANOVA, 3–4 months [WT, n = 16; Whrn+/−, n = 17], p = 0.0009; 5–6 months [WT, n = 13; Whrn+/−, n = 11], p < 0.0001). Whrn+/– males have progressive hearing loss indicating the early onset of age-related hearing loss comparing to WT males. To show the relationship between hearing loss and ages in males and females, we performed the linear regression of the change in ABR threshold (Click stimuli) against ages (1–6 months old). The linear regression showed a significant difference in the pattern of threshold change in females and males (Fig. 2C). The slope of linear model and Y-intercept were significantly different in females (Whrn+/– female: y = 0.853x + 69.41, WT female: y = 5.797x + 25.49, pslope = 0.034). It indicates that Whrn+/– females have severe hearing loss even as young adults, whereas WT females show age-related hearing loss. For males, although the slope was not different between WT and Whrn+/– males (pslope = 0.086, linear regression), the Y-intercept was significantly different (pintercept < 0.0001, linear regression). This indicates that Whrn+/– males have the early onset of age-related hearing loss compared with WT males showing a normal hearing within the age of 6 months.

Loss of Outer Hair Cell in Whrn +/− Mice

To determine the cellular mechanisms of hearing loss in Whrn+/– mice, we examined hair cell loss throughout apical to basal regions of the cochlea using immunostaining with antibodies against Myo7a (Fig. 3). The numbers of hair cells were counted at given length (10 μm). The density of OHCs were measured from 3 rows and they were averaged per mouse. The number of OHC in apical (1.282 ± 0.025, n = 5 WT mice vs 1.288 ± 0.044, n = 7 Whrn+/– males vs 1.300 ± 0.033, n = 5 Whrn+/– females, Kruskal–Wallis test, p = 0.9599) and middle region (1.299 ± 0.019, n = 4 WT vs 1.300 ± 0.031, n = 7 Whrn+/– male vs 1.321 ± 0.026, n = 5 Whrn+/– female, Kruskal–Wallis test, p = 0.8780) of cochlear were similar between WT and Whrn+/– mice regardless of sex. The number of OHCs in basal region showed a tendency of reduction (1.251 ± 0.066, n = 4 WT vs 1.122 ± 0.076, n = 6 Whrn+/– male vs 0.973 ± 0.148, n = 5 Whrn+/– female, Kruskal–Wallis test, p = 0.3869). Although no statistically significant difference was observed in the number of OHCs, there was a clear trend toward a reduced number of OHCs in the basal region of Whrn+/− mice. These findings suggest that the hearing loss observed in Whrn+/− mice is not directly attributable to the loss of OHCs, IHCs. However, the observed trend implies that OHCs in the basal region, which are responsible for responding to high-frequency sounds, may be particularly vulnerable to reduced levels of whirlin. Furthermore, the data suggest a potential sex-specific susceptibility, with female mice appearing more vulnerable to the effects of Whrn mutations.

Fig. 3

Loss of hair cells in Whrn+/– mice. A Expression Mayo7a in OHCs and IHCs in apical, middle, and basal membrane of the cochlea from Whrn+/– females (4–5 months old, n = 5), males (4–5 months old, n = 7), and male WT (4–5 months old, n = 4–5) as a control. White asterisks indicate the loss hair cells. B The number of OHCs of Whrn+/– females (green), Whrn+/– males (red), and WT (blue) at 4–5 months old. Data were shown as mean ± s.e.m. The statistical significance was tested using one-way ANOVA test followed by multiple comparison with Šidák correction, but there was no statistically significant difference between groups