V-ATPase is a multi-subunit enzyme complex also known as the vacuolar H⁺-ATPase. This proton pump is mainly responsible for the acidification of lysosomes and other membrane-bound compartments (Beyenbach and Wieczorek, 2006). V-ATPase comprises a peripheral V1 domain catalyzing ATP hydrolysis and a membrane integral V0 domain involved in proton translocation (Beyenbach and Wieczorek, 2006; Mindell, 2012). The V1 domain includes at least eight different subunits (A-H), while the V0 domain includes six different subunits (a,d,e,c,c′, and c″). The loss of any subunit will disrupt the assembly of V-ATPase and affect lysosome acidification, which could lead to various disorders (Ma et al., 2011).

ATP6V1B2, which encodes the V1B2 subunit in V-ATPase, is the causative gene for dominant deafness-onychodystrophy syndrome (DDOD syndrome, MIM: 124480) (Yuan et al., 2014); deafness, onychodystrophy, osteodystrophy, intellectual disability, and seizures syndrome (DOORS syndrome, MIM: 220500) (Beauregard-Lacroix et al., 2020); and Zimmermann–Laband syndrome (ZLS, MIM: 135500) (Kortum et al., 2015). These syndromes share the symptoms of congenital sensorineural deafness, onychodystrophy, and different extents of intellectual disability without or with epilepsy. ATP6V1B2 c.1516 C > T, p.Arg506^∗, was identified in all families with DDOD syndrome (Yuan et al., 2014; Menendez et al., 2017). To investigate the pathological role of mutant ATP6V1B2 in the neurosensory system, we generated a transgenic line of mice carrying c.1516 C > T (p.Arg506^∗) in Atp6v1b2, Atp6v1b2^{Arg506*/Arg506*} (homozygous mutant), and identified that the mutant mice displayed obvious cognitive defects for which impairment in the hippocampal CA1 region might be the pathological basis (Zhao et al., 2019). The interaction between the V1B2 and V1E subunits was found to be weaker in Atp6v1b2^{Arg506*/Arg506*}mice than in wild-type (WT) mice, indicating that the assembly of V-ATPase was affected by the mutation (Zhao et al., 2019). In another prior study, when pIRES2-EGFP-ATP6V1B2 WT and pIRES2-EGFP-ATP6V1B2 c.1516 C > T mutant plasmids were transfected into HEK293 cells, the lysosomal pH was increased, revealing the reduced acidification in the lysosome caused by the c.1516 C > T mutation (Yuan et al., 2014). Lysosomes, as scavengers of living cells, play a critical role in cellular metabolism. An abnormal lysosomal pH will cause dysfunctional macromolecule degradation and lead to lysosomal storage diseases. Loss of the V-ATPase subunits in the Drosophila fat body cells resulted in an abnormal pH in the lysosomal lumen, causing an accumulation of non-functional lysosomes and leading to a blockade of autophagic flux (Mauvezin et al., 2015). Additionally, abnormalities in autophagy caused by V-ATPase defects are associated with neurodegenerative diseases (Peric and Annaert, 2015; Cerri and Blandini, 2019).

Whether the Atp6v1b2 c.1516 C > T mutation results in autophagic dysfunction and leads to abnormal auditory function remains unanswered. However, the Atp6v1b2^{Arg506*/Arg506*} mice displayed normal auditory brainstem response (ABR) thresholds before 24 weeks of age (Zhao et al., 2019), while in the patients with ATP6V1B2-related syndromes, the hearing loss is congenital and severe. Genetic compensation can be induced by non-sense mutations that could occur between homologous genes (Ma et al., 2019). Two highly homologous genes encode V1B subunits: ATP6V1B1 and ATP6V1B2. The B2 subunit of V-ATPase was shown to functionally substitute for the B1 subunit (Paunescu et al., 2007). Therefore, we speculate that genetic compensation might be responsible for the phenotype of Atp6v1b2^{Arg506*/Arg506*} mice.

Herein, we performed long-term studies on hearing in Atp6v1b2^{Arg506*/Arg506*} mice, investigated the pathogenic mechanism of neurodegeneration in the auditory pathway, and confirmed the genetic compensation in Atp6v1b2^{Arg506*/Arg506*} mice to explain the difference between hearing impairments observed in patients and those produced in this mouse model.

The manifestation of normal auditory thresholds but reduced suprathreshold amplitude of the sound-evoked compound action potential of SGNs, shown as peak I of the ABR waveform, is defined as HHL (Kujawa and Liberman, 2009; Bharadwaj et al., 2014; Mehraei et al., 2016). The pathological mechanisms of HHL are complicated and are not yet fully understood. The electrophysiological results showed that Atp6v1b2^{Arg506*/Arg506*} mice younger than 20 weeks did not have an elevated ABR threshold but had lower P1 amplitudes and longer P1 latencies than WT mice, similar to patients with HHL (Mehraei et al., 2016). Current studies suggest that risk factors associated with HHL include a noisy environment (Liu H. et al., 2019), aging (Fischer et al., 2019), and ototoxic drugs (Liu et al., 2015). LOHL (late-onset hearing loss) is defined as hearing loss that is not present at birth but is identified at a later period. Both environmental and hereditary factors influence the development of LOHL. Clinically, there is no early diagnostic indicator for LOHL until hearing loss is detected. HHL and LOHL are regarded as two diseases with different pathogeneses (Song et al., 2020). We proposed that (1) the Atp6v1b2^{Arg506*/Arg506*} mouse model could be used as a natural model for HHL and LOHL; (2) patients with LOHL may have HHL early in life, and it is important for those who have been diagnosed with HHL to be cautious of risk factors such as noise, ototoxic drugs, etc. to avoid further hearing impairment.

Abnormalities in ribbon synaptic density and demyelination of the auditory nerve fibers have been identified as causes of HHL (Liberman and Kujawa, 2017; Wan and Corfas, 2017). In our study, we identified that HHL in Atp6v1b2^{Arg506*/Arg506*} mice was caused by demyelination of nerve fibers but not the ribbon synaptic density. Schwann cells are essential for the formation of nodes of Ranvier, which is a special structure along the myelinated fibers where voltage-gated sodium channels and potassium channels accumulate for regeneration of action potentials and fast synchronous transmission of electrical signals. This phenomenon explains why nerve fibers with myelin sheaths can conduct action potentials more quickly and efficiently than nerve fibers without myelin sheaths (Rasband and Peles, 2015).

Damage or loss of cells from the auditory pathway in the cochlea, such as IHCs and/or OHCs, supporting cells and SGNs, is a pathological feature of LOHL (Vreugde et al., 2002; Liu et al., 2016; Cheng et al., 2019; Gao et al., 2019; Han et al., 2020; Zhang S. et al., 2020; Zhang Y. et al., 2020; Chen et al., 2021). HCs are highly specialized cells attached to the basement membrane in the organ of Corti in the cochlea, and mainly function in transduce the sound mechanical vibration into the electrical signal (He et al., 2017, 2020, 2021; Tan et al., 2019; Jiang et al., 2020; Zhong et al., 2020; Zhou et al., 2020). These cells have hair-like protrusions (stereocilia) embedded in the tectorial membrane (Wang et al., 2017; Liu Y. et al., 2019; Qi et al., 2019, 2020; Cheng et al., 2021; Fu et al., 2021; Zhang et al., 2021). Incoming soundwaves distort the basement membrane, and the resulting mechanical distortion of the stereocilia is transduced into neural signals that are conveyed through the SGNs to the auditory regions of the brain (Ding et al., 2020; Guo et al., 2020; Jiang et al., 2020; Qian et al., 2020; Liu et al., 2021; Lv et al., 2021; Yuan et al., 2021). Improved culture system of SGNs facilitates the study of physiology and pathophysiology, and promotes identification of potential therapeutic targets for SGNs protection and regeneration (Guo et al., 2016, 2019, 2020; Waqas et al., 2017; Yan et al., 2018; Liu W. et al., 2019; Liu et al., 2021). In this study, Atp6v1b2^{Arg506*/Arg506*} mice developed demyelination of nerve fibers, followed by loss of nerve fibers and SGN cell bodies. This process appears to be neurodegenerative. However, the cochlear hair cells of Atp6v1b2^{Arg506*/Arg506*} mice were morphologically normal.

Hearing loss in patients with DDOD syndrome is primarily treated by cochlear implantation. However, although the implanted cochlea worked well, clinical follow-up of DDOD syndrome patients with cochlear implantation revealed that their language rehabilitation was unsatisfactory. Our findings regarding the pathological changes in SGNs and mild learning and memory problems in DDOD syndrome patients (Zhao et al., 2019) could explain this clinical puzzle.

V-ATPase has two main effects on the organelle: maintaining the acidic environment of the organelles and participating in and regulating the fusion between vesicle-type organelles, such as the fusion between lysosomes and autophagosomes (Colacurcio and Nixon, 2016). We have demonstrated that abnormalities in the assembly of the subunits of V-ATPase occurred in Atp6v1b2^{Arg506*/Arg506*} mice, which affects the transfer of hydrogen ions into the lysosome (Yuan et al., 2014; Zhao et al., 2019). In this study, we observed the presence of metabolic substrates inside lysosomes, which indicated there might be no fusion barrier between lysosomes and autophagosomes in Atp6v1b2^{Arg506*/Arg506*} mice. The degradation of lysosomes depends on the presence of multiple hydrolases, and the activity of these hydrolases requires an acidic environment (De Duve, 1957; Saftig and Haas, 2016). We found abnormal degradation of metabolic substrates in lysosomes of the SGNs in Atp6v1b2^{Arg506*/Arg506*} mice as a result of the abnormal acidic environment and the affected activity of the hydrolases in the lysosomes. In addition, redundant autophagosomes were observed outside the lysosomes in the SGNs of Atp6v1b2^{Arg50*/Arg506*} mice, indicating that high levels of undegraded substrates in the lysosomes affected the normal process of autophagic flow (Colacurcio and Nixon, 2016).

We observed an increase in caspase-3, the ultimate initiator of apoptosis, in the SGNs and Schwann cells of Atp6v1b2^{Arg506*/Arg506*} mice, which suggested that lysosomal dysfunction induced apoptosis. Mitochondrial dysfunction and premature termination of cell cycle occur when a series of lysosomal functional abnormalities present, such as increased intraluminal pH value, abnormal degradation, and excessive accumulation of exogenous substrate for lysosomes (Hughes and Gottschling, 2012; Molin and Demir, 2014; Ruckenstuhl et al., 2014). Cell damage triggers degradation of the Bcl-2 family of proteins, which in turn activates Bax. Activated Bax/Bak complexes bind to and reduce the permeability of the mitochondrial membrane, resulting in cytochrome C release to the cytosol. Cytochrome C induces apoptosome formation, which can directly activate caspase-3 (Kroemer et al., 2007; Galluzzi et al., 2012; Ilmarinen et al., 2014).

Next, we evaluated whether treatment with the Bax inhibitor BIP-V5 could ameliorate the hearing loss in the mutant mice. After 36-week systemic administration of BIP-V5, auditory function and pathological changes improved in two out of four Atp6v1b2^{Arg506*/Arg506*}-BIP-V5 mice, which further verifies that apoptosis induces the degeneration of SGNs and further impaired the hearing. As the reason for mortality was unknown in 50% (four out of eight) of the treated mice, the safety of BIP-V5 treatment needs further exploration. Bax plays an important role in programmed cell death or apoptosis and contributes to maintenance of normal psychological functions of various organs. Atp6v1b2 gene is highly expressed in the inner ear and central nervous system of mice (Yue et al., 2014), and we presented the Arg506^∗ mutation might induce Bax-mediated apoptosis. Therefore, administration of BIP-V5 to inhibit Bax is a reasonable way to maintain normal apoptosis in the inner ear and central nervous system. However, Atp6v1b2 gene is rarely expressed in some organs such as the adrenal gland, intestine, stomach, liver, etc. and the mutation has minimal effect on Bax. In this case, intervention of Bax is likely to disrupt the normal physiological functions of these organs and lead to abnormalities (Kanauchi et al., 2002; Anagnostopoulos et al., 2005; Zhou et al., 2017; Al Humayed et al., 2020). Therefore, although intraperitoneal administration can control the apoptosis of spiral ganglion cells to a certain extent, it also has serious side effects on other important organs, and even leads to death. Topical administration of the inner ear might avoid the high mortality caused by intraperitoneal administration in mice, but this also brings an additional problem: the mice require long-term drug intervention, and routine operation of multiple cochlear administrations in mice can damage the structure of the inner ear, thus affecting hearing. Therefore, in order to achieve the purpose of hearing recovery by targeted drug intervention in the inner ear of mice, it is crucial to explore a new surgical method in the future.

Patients with ATP6V1B2-related syndromes showed severe congenital sensorineural hearing loss, which was not consistent with the milder hearing phenotype of Atp6v1b2^{Arg506*/Arg506*}mice despite the presence of the same type of mutation (Menendez et al., 2017). Based on the finding that hair cells were not affected in Atp6v1b2^{Arg506*/Arg506*} mice, we hypothesized the presence of genetic compensation. Genetic compensation was already shown between isoforms of V-ATPases. For example, compensation exists between Atp6v1g1 and Atp6v1g2, both of which encode the V1G subunit. The V1G subunit plays an important role in the nervous system. Defects in the V1G subunit can lead to cognitive disorders. However, Atp6v1g2^–/– mice did not show any developmental defects or obvious behavioral abnormalities until adulthood, which is considered to be associated with the upregulation of Atp6v1g1 in the mouse brain (Kawamura et al., 2015).

The V1B subunits of the V-ATPase are encoded by Atp6v1b2 and Atp6v1b1. Atp6v1b2 mRNA is ubiquitously expressed in mouse tissues, including the inner ears. Atp6v1b1 is mainly expressed in the kidney, epididymis, eyes, and inner ears (Tian et al., 2017). The B2 subunit of V-ATPase was shown to compensate for the function of the B1 subunit of V-ATPase in the renal medullary intercalated cells of B1-deficient mice (Paunescu et al., 2007). In this study, we observed increased transcription of Atp6v1b1 in hair cells of Atp6v1b2^{Arg506*/Arg506*} mice and inferred that Atp6v1b1 compensated for the Atp6v1b2 dysfunction by increasing its own transcription level. Genetic compensation in hair cells explains the milder hearing impairment in Atp6v1b2^{Arg506*/Arg506*} mice. The mechanism of elevated transcription may be due to an enhancement in the promoter region of the gene by trimethylation (El-Brolosy et al., 2019; Ma et al., 2019). However, if a mutation occurs only at the transcriptome level and not at the DNA level, this compensatory mechanism will not be activated (Rossi et al., 2015; Ma et al., 2019). This finding was verified by our previous work: when we injected morpholinos designed to knock down Atp6v1b2 mRNA in the cochlea of mice, the hair cells were damaged, and the mice showed severe hearing loss (Yuan et al., 2014). In addition, the cognitive and memory impairment in Atp6v1b2^{Arg506*/Arg506*}mice was verified to be caused by apoptosis in the hippocampus (Supplementary Figure 8). We speculated that the function of Atp6v1b2 gene is more important or more specific both in the SGNs and the central nervous system than in the hair cells, and thus it might not be fully compensated by other genes, or that defect of Atp6v1b2 does not induce functional compensation in the regions of SGNs and hippocampus.

Based on the findings of this study and our previous studies, we elucidated the pathogenesis of Atp6v1b2 defects as follows: (1) the mutation causes dysfunctional assembly between V-ATPase V1 subunits, and H⁺ transfer to lysosomes is decreased, which leads to an increased pH value; (2) disequilibrium in the acidic environment of the lysosome further affects the function of acid-dependent hydrolytic enzymes, resulting in the accumulation of degradation substrates; (3) the Bcl-2 family detects abnormal lysosome autophagy and then induces the release of cytochrome C from the mitochondria; and (4) apoptosis occurs in SGNs, affecting hearing function (Figure 6).

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

The animal study was reviewed and approved by the institutional animal care and use committee of the Chinese PLA General Hospital.

SQ, WZ, and YY drafted the manuscript. SQ and WZ participated in the construction and analysis of Atp6v1b2 c.1516C > T knockin mice and performed western blotting, electron microscopy, RNAscope, and BIP-V5 treatment. DL and BG carried out the experiments of immunohistochemistry. XG and WW performed the auditory evaluation and statistical analysis. YY and PD conceived the study. PC, SY, and WH participated in its design and coordination. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.