Hearing Loss in Id1−/−; Id3+/− and Id1+/−; Id3−/− Mice Is Associated With a High Incidence of Middle Ear Infection (Otitis Media)

Inhibitors of differentiation/DNA binding (Id) proteins are crucial for inner ear development, but whether Id mutations affect middle ear function remains unknown. In this study, we obtained Id1−/−; Id3+/− mice and Id1+/−; Id3−/− mice and carefully examined their middle ear morphology and auditory function. Our study revealed a high incidence (>50%) of middle ear infection in the compound mutant mice. These mutant mice demonstrated hearing impairment starting around 30 days of age, as the mutant mice presented elevated auditory brainstem response (ABR) thresholds compared to those of the littermate controls. The distortion product of otoacoustic emission (DPOAE) was also used to evaluate the conductive function of the middle ear, and we found much lower DPOAE amplitudes in the mutant mice, suggesting sound transduction in the mutant middle ear is compromised. This is the first study of the middle ears of Id compound mutant mice, and high incidence of middle ear infection determined by otoscopy and histological analysis of middle ear suggests that Id1/Id3 compound mutant mice are a novel model for human otitis media (OM).


INTRODUCTION
Inhibitors of differentiation/DNA binding (Id) proteins are major inhibitors of helix-loop-helix (HLH) transcription factors. The Id genes encode four related transcription factors: Id1, Id2, Id3, and Id4. Id transcription factors contain a helix-loop-helix region similar to that of the basic helix-loop-helix (bHLH) transcription factors, can form heterodimers with some of them and lose their binding to DNA. However, due to the lack of basic DNA-binding regions, these heterodimers cannot bind to DNA. Consequently, Id transcription factors negatively regulate the DNA binding capacity of bHLH proteins (Benezra et al., 1990;Wong et al., 2004). In other words, Id transcription factors usually hinder cell differentiation, especially the differentiation of dendritic cells, which plays a rather important role in immunity to infectious agents (Spits et al., 2000;Kee, 2009).
Inhibitors of differentiation/DNA binding proteins regulate different kinds of cellular processes, including cellular growth, differentiation, migration, senescence, and tumorigenesis (Sikder et al., 2003;Wong et al., 2004;Wu et al., 2018). Id1, Id2, and Id3 are expressed in the otic vesicle of mouse embryos (Jen et al., 1997). The importance of these three Id genes as differentiation regulators was also illustrated by their key role in the regulation of expression of Math1 and hair cell differentiation in the developing cochlea (Jones et al., 2006). Ids expression profiles in the normal mature middle ear has not been reported. However, a previous study by Lin et al. (2002) showed that Id1 and Id3 are unregulated in the middle ears of rats following pneumococcal infection, which indicates that Id1 and Id3 may be involved in middle ear diseases such as otitis media (OM). The main objective of this study was to clarify the correlation between the two Id genes with OM.
The most common middle ear problems include cholesteatoma, tympanic membrane (TM) perforation, middle ear infection, and otosclerosis. Among these diseases, middle ear infection, or OM, is the most common, as most children younger than 3 years will have at least one episode of OM (McCaig et al., 2002;O'Brien et al., 2009). Acute OM is the most common cause of meningitis, whereas neglected chronic OM can lead to permanent hearing loss (Pont and Mazon, 2017). Previous studies have shown that genetic lesions lead to a high incidence of OM in mutant animals such as Jeff mice, Sh3pxd2b mice, and Enpp1asj mice (Hardisty-Hughes et al., 2006;Yang et al., 2011;Tian et al., 2016). Animal models with OM susceptible to a defined genetic lesion will be important to reveal the pathogenesis and underlying genetic pathways linked to OM.
Individual null mutants for Id1, Id2, and Id3 have been generated and reported to have very few abnormalities, and none associated with hearing deficit (Yan et al., 1997;Lyden et al., 1999;Yokota et al., 1999). Id1 and Id3 double-knockout mice are commonly used as animal models to study the role of Id1 and Id3 in mammalian development (Zhao et al., 2011). Unfortunately, complete loss of these genes leads to aggregation of dilated and irregularly shaped blood vessels and brain hemorrhage by E12.5, and no embryos survived beyond E13.5 (Lyden et al., 1999;Fraidenraich et al., 2004). Because of the upregulated expression of Id1 and Id3 in middle ears after the acute episode of pneumococcal otitis media has been resolved, the two genes are presumed to be involved in the development of otitis media (Lin et al., 2002). In this study, the auditory brainstem response (ABR) and middle ear morphology in Id1 −/− ; Id3 +/− and Id1 +/− ; Id3 −/− mice were carefully examined. OM and associated elevated hearing thresholds were found in the majority of the mutant mice. Our data suggest that Id1/Id3 mice are valuable models for the study of OM pathogenesis and associated genetic factors.
Littermate mice with Id1 +/− ; Id3 +/− , Id1 +/+ ; Id3 +/− , Id1 +/− ; Id3 +/+ or Id1 +/+ ; Id3 +/+ genotypes were generated incidentally during breeding. We used littermate mice with Id1 +/+ ; Id3 +/+ genotype as wild-type controls. Mice of the other three genotypes were excluded because they did not show significant hearing loss (data not shown). No gender difference was found during the experiment. Both male and female mice from the same litter were included as the experimental and control groups. A total of 101, Id1 −/− ; Id3 +/− mice, 28 Id1 +/− ; Id3 −/− mice and 30 wild-type mice were used in this study. ABR test was performed at four time points: P30 (P: postnatal, same meaning as below), P60, P90, and P120. After ABR test, distortion product of otoacoustic emission (DPOAE) measurement was performed at two time points: P30 and P60. After ABR and DPOAE test at P30, otoscopic examination was carried out and then some of the mice were euthanized to conduct Hematoxylin-Eosin (H&E) staining. Only the data of the right ear were counted in the analysis of the experimental results. Specific details on the use of mice can be found in the Supplementary

Auditory Brainstem Response, Distortion Product Otoacoustic Emission Tests, and Otoscopic Examination
Auditory brainstem response threshold analyses were carried out using equipment from Intelligent Hearing Systems (Miami, FL, United States) as previously described (Zheng et al., 1999). Prior to examination, mice were anesthetized with 2, 2 and 2-tribromoethanol (Avertin; 0.5 mg/g body mass), and their body temperature was kept at 37-38°C by placing them on a heating pad in a soundproof chamber during testing. Specific auditory stimuli (broadband click and pure-tone burst stimuli of 8, 16, and 32 kHz) from ER2 and high-frequency transducers were delivered through plastic tubes to the ear canals. Evoked brainstem responses were amplified and averaged, and their wave patterns were displayed on a computer screen. Auditory Frontiers in Genetics | www.frontiersin.org 3 August 2021 | Volume 12 | Article 508750 thresholds were obtained for each stimulus by varying the SPL in 10 dB steps and, finally, a 5 dB step up and down to identify the lowest level at which an ABR pattern could be recognized. The DPOAE test was performed using SmartOAE 4.50 USBez software (Intelligent Hearing Systems) and an Etymotic 10B+ OAE probe (Etymotic Research, Elk Grove Village, IL) fitted with a high-frequency transducer (Intelligent Hearing Systems) that produced two pure tones, F1 and F2, respectively. The methods and process followed those described in a previous paper (Zhang et al., 2012).
After the ABR and DPOAE tests, otoscopy was performed on both ears to determine the condition of TM, including presence of middle ear fluid, inflammation, or infection.

Statistical Analysis
The data are presented as mean ± SEM. The difference between groups was determined using a one-way or two-way ANOVA or unpaired student t-test when applicable using GraphPad Prism software.

DISCUSSION
In this investigation, we found that middle ear function is affected in Id1 and Id3 mutant mice due to OM, which leads Frontiers in Genetics | www.frontiersin.org 5 August 2021 | Volume 12 | Article 508750 to conductive hearing loss in the affected mice. Because Id1/ Id3 double-mutant embryos die at approximately E12 (Lyden et al., 1999), and Id1 and Id3 single-knockout mice have normal hearing (Yan et al., 1997;Lyden et al., 1999;Yokota et al., 1999; data not shown), Id1/Id3 heterozygous (Id1 −/− ; Id3 +/− and Id1 +/− ; Id3 −/− ) mice were used to study the combined effect of missing Id1/Id3 alleles. Auditory brainstem response has been used extensively to assess mouse IE function and also offers a valid, simple physiologic test of mouse middle ear inflammation (MacArthur et al., 2006). Overall, most observed Id1 +/− ; Id3 −/− and Id1 −/− ; Id3 +/− mice had high ABR thresholds in at least one of the stimulus frequencies (click,8,16,and 32 kHz) in at least one ear. At P30, both Id1 +/− ; Id3 −/− and Id1 −/− ; Id3 +/− mice showed elevated ABR thresholds compared to age-matched littermate controls. However, at P60, ABR thresholds for Id1 −/− ; Id3 +/− mice at 16 and 32 kHz no longer showed significant differences compared to the controls. Moreover, at P60, the ABR thresholds of Id1 −/− ; Id3 +/− mice were lower than those of the Id1 +/− ; Id3 −/− mice for all four frequencies tested. Studies have shown that Id1/Id3 genes have redundant functions, and the loss of one gene function is compensated by another (Lowery et al., 2010). However, based on the ABR data, we can speculate that Id3 might play a slightly FIGURE 3 | Comparison of the distortion product of otoacoustic emission (DPOAE) amplitudes of right ears from age-matched littermate control mice, Id1 −/− ; Id3 +/− and Id1 +/− ; Id3 −/− mice at ages P30 and P60. The results indicated that at P30, the DPOAE amplitudes of Id1 −/− ; Id3 +/− (n = 12) and Id1 +/− ; Id3 −/− mice (n = 9) were significantly lower than those of the control mice (n = 13) at dominant frequencies. At P60, the DPOAE amplitudes at dominant frequencies in the Id1 −/− ; Id3 +/− mice (n = 35) were significantly lower than those of the controls (n = 8), and the DPOAE amplitudes at two frequencies in Id1 +/− ; Id3 −/− mice were significantly lower than those of the controls (n = 14). The error bar indicates SD from the mean or each group. *p < 0.05; **p < 0.01. greater role in middle ear function. A longitudinal tracking in our study indicated that Id1/Id3 mutant mice from P30 to P120 showed no progression of hearing loss; one explanation is that 4-month-old mice are still relatively young, and the deterioration of hearing loss in mutant mice might show up if we test the mutant mice for extended period, such as until 12 months. Distortion product of otoacoustic emission is commonly used to evaluate outer hair cell integrity in humans and research animals. Several groups have established this method as another way to evaluate the middle ear conductance and as an indirect measure of conductive hearing loss caused by middle ear dysfunction. This is primarily due to the presence of middle ear effusion, which prevents sound transduction to the inner ear for further processing. Besides, due to lesions in the middle ear, the DPOAE energy cannot be released through the middle ear and detected in the external ear canal. We showed that Id1 +/− ; Id3 −/− and Id1 −/− ; Id3 +/− mice have much lower DP values compared to littermate controls. If this is not due to outer hair cell dysfunction, then middle ear conductance is very likely to be affected in the mutant mice.
To identify the histopathology of the Id mutant mice that leads to conductive hearing loss, we first examined the gross morphology of the TM. Transparent and balanced TM is observed in the control mice. However, in the mutant mice, bubbles are the most common observation through the TM, which is an indication of middle ear effusion. TM retraction, another common observation, is caused by negative pressure in the middle ear cavity (MEC). In some cases, we observed white patches on the TM, which is usually due to soft tissue calcification, which eventually lead to increased stiffness of the TM and decreased membrane conductance. No TM perforation or missing ossicle bone, particularly the malleus, was observed in the mutant mice. These discoveries exclude the possibility of conductive hearing loss originating from abnormal outer ear canals and TMs.
Next, HE stained inner ear and middle ear cross sections were prepared for more detailed evaluation of the middle ear and inner ear morphology. The mutant mice's inner ears did not show any abnormalities. For example, there was no hair cell, spiral ganglion neuron loss or dislocation of Reissner's membrane and no stria vascularis defect. However, different degrees of effusion were present in the middle ear cavities of the mutant mice, which is in consistent with the observation through the TM. Inflammatory cells were also present in the MEC for an extended period, primarily polymorphonuclear cells. This is a major feature of chronic middle ear inflammation. Middle ear epithelia generally thickened in most of the mutant mice, another typical feature of middle ear inflammation. Combining ABR threshold data and histological examination, we found that OM and pathology correlated well with ABR threshold data from the Id1 +/− ; Id3 −/− and Id1 −/− ; /Id3 +/− and wild-type mice. These data together suggest that Id1/Id3 mutations cause middle ear inflammation, which leads to excess effusion, mucosal thickening and the presence of inflammatory cells in the MEC and the inflammation in middle ear causes conductive hearing loss in Id1/Id3 mutant mice.
Id1 and Id3 are two members of a family of four Id proteins (designated Id1 through Id4) that act as dominant negative inhibitors of bHLH transcription factors (Benezra et al., 1990;Christy et al., 1991;Sun et al., 1991;Riechmann et al., 1994). Id1 and Id3 gene in humans are located on chromosomes 20qll (Idl), lp36.1 (Id3), and they share a very similar genomic organization of exon-intron boundaries within their coding regions (Norton et al., 1998). They are co-expressed temporally and spatially during murine neurogenesis and angiogenesis (Lyden et al., 1999). Id genes are expressed in the otic vesicle's prosensory domains and are involved in hair cell development through an unknown mechanism (Jones et al., 2006;Kamaid et al., 2010). In the larger picture, Id proteins are crucial for the proper development and function of the immune system (Pan et al., 1999;Miyazaki et al., 2014).
The function of Id proteins associated with the immune system has been well documented in several studies (Miyazaki et al., 2014;Zook and Li, 2018;Han et al., 2019). Lack of Id3 leads to impaired B-cell proliferation and immune responses (Pan et al., 1999). A recent study showed that Id1 favors the differentiation of myeloid-derived suppressor cells (MDSCs), but not dendritic cells (Papaspyridonos et al., 2015). MDSCs are a heterogeneous group of immune cells from the myeloid lineage that suppress immunity against infectious agents. These studies point the possible etiology of OM in Id1/Id3 mutant mice to impaired immune responses. It has been well recognized that a compromised immune system is one of the major contributing factors to OM. We speculate that Id1/Id3 mutations lead to an impaired immune system, which compromises the ability of the mutant mice to fight against middle ear inflammation, leading to continuous presence of effusion, epithelia hyperplasia and inflammatory cells.
Vascular endothelial growth factor (VEGF) is one of the most potent angiogenic factors under inflammatory conditions in the middle ear (Juhn et al., 2008). A previous study used clinical specimens of cholesteatoma in the middle ear to identify a transcription factor that regulate the growth of cholesteatoma. They found Id1 is an essential regulator of VEGF in the cholesteatomal matrix and perimatrix (Fukudome et al., 2013). Some studies have demonstrated that TGF-β, VEGF, or hypoxiainducible factor (HIF) signaling plays a critical role in the pathogenesis of chronic otitis media in animal models (Tateossian et al., 2009(Tateossian et al., , 2013Cheeseman et al., 2011;Husseman et al., 2012). As the important differentiation regulators, Id proteins regulate different kinds of cellular processes. More studies need to be done to clarify the underlying mechanism of otitis media in Id1/Id3 mutant mice.
In conclusion, we have shown that the Id1 +/− ; Id3 −/− and Id1 −/− ; /Id3 +/− mice provide excellent models for studying OM. This model's susceptibility to OM may be related to the mice's weakened immune response toward infectious agents. Moreover, the individual variability observed in the Id1 +/− ; Id3 −/− and Id1 −/− ; /Id3 +/− mouse population may provide a valuable control in future explorations of this model. As such, with this mouse model, we can further elucidate causal relationships between the multiple features of OM and provide an optimal approach to minimizing hearing loss in affected individuals.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
The experimental protocols were approved by the Case Western Reserve University Animal Care and Use Committee and were in agreement with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

AUTHOR CONTRIBUTIONS
YZ and QZ contributed to the study concept and design.