Estrogen inhibits apoptosis of pericytes in the cochlea stria vascularis through downregulating TMEM16A

Background: Currently, the speci�c mechanism of estrogen (E2) in protecting presbycusis is not clear. This study aimed to investigate whether E2 could affect the apoptosis of capillary pericytes (PCs) of cochlear stria vascularis (SV) in aged C57BL/6J mice by regulating transmembrane member 16A (TMEM16A), such that it plays a protective role in presbycusis. Methods: The model of C57BL/6J ovariectomized mice was established, and E2 was administered for 2 months. The hearing threshold was measured by auditory brainstem response (ABR). The changes in the cochlea were measured using hematoxylin-eosin (HE) and electron microscopy. qRT-PCR was used to examine the expression of TMEM16A and apoptosis-related protein mRNA. The PCs were cultured in vitro, and the cell senescence model was established by the continuous passage method. TMEM16A expression was assessed using immuno�uorescence. Flow cytometry was performed to explore the apoptosis rate. Results: The results of animal experiments showed that E2 intervention could reduce hearing loss and improve the atrophy of cochlear SV, loss of PC chromatin organelles, cytoplasmic swelling and nuclear porosity in aged mice. E2 also decreased the mRNA expression of TMEM16A, Caspase-3 and Bax in cochlear SV of aged mice and upregulated the expression of Bcl-2 mRNA. Cellular experiments showed that E2 could downregulate the expression of TMEM16A, and E2 or T16Ainh-A01(a speci�c blocker of TMEM16A) could reduce the apoptosis rate of PCs in aged mice. Conclusions: E2 may inhibit the apoptosis of PCs through downregulating the expression of TMEM16A, which plays a protective role in presbycusis. This study may provide a novel potential treatment and prevention method for presbycusis.


Introduction
With the aging of the population, organs age and change in function, and the number of patients with presbycusis is increasing (Ciorba et al. 2015).Presbycusis seriously affects the quality of life of the elderly, the effective use of hearing aids is bene cial to this population, improving their quality of life and health conditions (Carniel et al. 2017).To date, the speci c pathogenesis of presbycusis has not been clear.Age-related deterioration of the stria vascularis (SV) is one of the main causes of presbycusis (Gratton et al. 1996;Spicer and Schulte 2005).Pericytes (PCs) play an important role in maintaining SV homeostasis (Shi et al. 2008).Some studies have shown that the abnormal PCs in the cochlea can cause capillary leakage and lead to hearing loss (Shi 2009).Although the mechanism of these phenomena is still unclear, the data suggest that PCs may be a new direction for the prevention and treatment of presbycusis.
There are gender differences in the damage of presbycusis, with men developing presbycusis earlier than women (Helzner et al. 2005;Park et al. 2016), and estrogen (E2) replacement therapy can delay presbycusis (Hederstierna et al. 2007; Williamson et al. 2020).Some studies have shown that E2 is related to apoptosis.E2 not only can reduce apoptosis of vascular smooth muscle cells (Ling et al. 2004) but can also reduce the apoptosis of cochlear outer hair cells induced by gentamicin ototoxicity, which is achieved by inhibiting the JNK pathway (Nakamagoe et al. 2010).However, the speci c mechanism of E2 in protecting presbycusis is not clear.
Transmembrane member 16A (TMEM16A)-encoded Ca 2+ -activated Cl − channels play a variety of physiological roles in many organs and tissues (Yang et al. 2008).Accumulating evidence shows that TMEM16A is closely related to many cells in the cochlea (Jeon et al. 2011;Wang et al. 2015;Zhang et al. 2015), and previous research by our research team showed that TMEM16A exhibits age-related changes in cochlear SV PCs (Zhou et al. 2019).It has been con rmed that autophagy and apoptosis are increased in presbycusis mice (Menardo et al. 2012).However, whether the effect of TMEM16A on SV PC apoptosis in presbycusis has not yet been reported.
Currently, the speci c mechanism of E2 in presbycusis is not clear.Based on the protective effect of E2 on cochlear cell apoptosis, this study aimed to explore whether E2 can inhibit the apoptosis of PCs in cochlear SV through regulating the expression of TMEM16A.Our research results have laid an experimental foundation for elucidating the potential mechanism of E2 regulating auditory function and have provided new strategies for the prevention and treatment of presbycusis.Cell-based experiments were divided into a group of 5 th -generation (P5) PCs, a group of 13 th -generation (P13) PCs, a group of P13 PCs given DMSO, a group of P13 PCs given E2 (treated with 10 -8 mmol/L E2 for 24 h), a group of P13 PCs given T16Ainh-A01 (treated with 30 μmol/L T16Ainh-A01 for 24 h), and a group of P13 PCs given T16Ainh-A01 and E2 (treated with 30 μmol/L T16Ainh-A01 for 24 h and then treated with 10 -8 mmol/L E2 for 24 h).

Materials And Methods
Auditory brainstem response (ABR) detection C57BL/6J mice were anesthetized by intraperitoneal injection and were placed in a soundproof screening room.Three platinum needle electrodes were placed subdermally at the vertex (positive electrode), the left mastoid (negative electrode) and right hind leg (ground electrode).ABR waves were recorded in response to a series of click stimuli.The intensity of acoustic stimulation was 30-100 dB SPL, the scanning time was 10 sec, and the number of superimposed stimuli was 1,024.The stimulation protocol included stimulus intervals of 11.10 beats/sec.The intensity of each stimulus decreased every 20 dB SPL.When approaching the threshold, the decrease was every 5 dB SPL.The amplitude and threshold changes of ABR III waves in each group were observed.

Hematoxylin-eosin staining
The bilateral temporal bones of mice were immersed in 4% paraformaldehyde solution.The cochlea in all groups were decalci ed in 10% EDTA (25°C) for 7 days.Then, the cochlea was embedded in para n so that its orientation was parallel to the cochlear axis.A microtome was then used to section the tissues at a thickness of 3 μm.Finally, the sections were stained with HE, analyzed by light microscopy and then photographed (BX43, Olympus, JPN).

Transmission electron microscopy (TEM)
The cochlear segments of the mice were xed overnight in phosphate-buffered 3% glutaraldehyde, and the cochlear tissue was then cut into 1 mm × 1 mm ×1 mm pieces and post xed in 1% osmium.Then, the osmium was ushed from the tissue, and it was dehydrated, saturated and embedded in Epon 812 epoxy resin.An LKB-2188 ultrathin slicer was used to generate sections, and uranium dioxide acetate and lead citrate double electron staining was performed.Observation and photography were performed using a transmission electron microscope (JEM-1230, JEOL, JPN).

Primary culture of PCs
Primary cell culture was performed as described above, with some modi cations (Zhang et al. 2017).Using C57BL/6J mice aged 8-11 days (n = 3), cochlear tissues were removed under sterile conditions and were soaked in 4°C D-Hanks solution.The SV was gently separated under a dissecting microscope and transferred to 0.7 mL of cell culture medium.Next, the SV was minced into 0.15-0.20-mm 3 pieces, and the pieces were evenly arranged on the bottom surface of the culture dish using sterile ophthalmology.Tissues were incubated at 37°C in an atmosphere containing 5% CO 2 .After 24 h, 1 mL of cell culture solution was carefully added.On day 3, the unattached fragments were washed with phosphate-buffered saline, and new cell culture solution was added.After 2 -3 washes per week, the PCs were puri ed by digestion passage.
Nuclear staining was performed using DAPI (Solarbio Science and Technology Co.), and the cells were photographed by laser scanning confocal microscopy (Zeiss LSM 510 META, Carl Zeiss AG).
Immuno uorescence detection of TMEM16A expression was performed.The experimental steps were the same as those used to identify PC immuno uorescence but with the incubation of different antibodies.The slides were incubated with a rabbit anti-TMEM16A antibody (1:100; cat.no.ab64085; Abcam) at 4°C overnight.The next day, slides were incubated with a FITC-conjugated goat anti-rabbit antibody (1:100, ZF-0311; OriGene Technologies, Inc.) at 37°C for 1 h.

Cell aging model
The cells of the 5th, 9th and 13th generations were placed in six-well plates and were cultured under standard conditions for 24 h.A β-galactosidase staining kit (cat.no.C0602; Beyotime, Haimen, China) was then used according to the manufacturer's instructions, and 1 mL of β-galactosidase xed solution was added to each well.After being xed at room temperature for 15 min, 1 mL of working dye solution was added to each well.After the six-well plate was sealed with plastic wrap and incubated overnight in a 37℃ incubator, the results were photographed under an ordinary optical microscope.

Flow cytometry
Trypsin digestion was used to collect the cells.Cells were centrifuged at 1000 rpm for 6 min and then were washed twice with precooled PBS.Each tube contained 500 μL of binding buffer, in which the cells were resuspended.Then, 5 μl of Annexin V-FITC and 10 μl of PI staining solution were added.After mixing, the cells were incubated in the dark for 30 min.The resuspended liquid was transferred to a ow cytometry sample tube for machine detection.

Statistical analysis
Statistical analysis was performed using SPSS 19.0 software, and GraphPad Prism was used to generate graphs.All the results are expressed by ().Every data point was analyzed by one-way ANOVA.Results with P values of less than 0.05 were considered statistically signi cant.

ABR was detected in different groups of mice
The change in the ABR III wave threshold was observed in each group (Fig. 1).The results of ABR showed that there was hearing loss in 12-month-old mice ( ** P < 0.01), and the hearing loss in ovariectomized mice was more obvious than that in 12-month-old mice ( # P < 0.05).E2 intervention could alleviate this hearing loss ( & P < 0.05).

Fig. 1 ABR threshold results of each group
The results are presented as the mean ± SD, n = 10.** P < 0.01 vs. the 3 m group.# P < 0.05 vs. 12 m group.& P < 0.05 vs. the 12 m OVX group.
Morphological observation of cochlear stria vascularis and pericytes HE staining was used to observe the morphological changes of cochlear SV (Fig. 2A).The structural changes of PCs were observed under TEM (Fig. 2B).In the 3 m group, the shape of cochlear SV was normal, the PCs had a good shape and the cytoplasmic and nuclear components were abundant.In the 12 m group, there was a large degree of vacuolar degeneration in the cochlear SV, loss of chromatin organelles and cytoplasmic swelling in PCs.In the 12 m OVX group, the cochlear SV was obviously atrophic, the cytoplasm was decreased, the PC chromatin organelles were lost, and the nucleus was loose.When treated with E2, the vacuolar degeneration decreased, the atrophy improved in the cochlear SV, the chromatin was dense and the cytoplasm was not swollen in PCs.These results suggest that E2 can improve the degeneration of SV and PCs in aged mice.mRNA content of TMEM16A, Bax, Bcl-2 and Caspase-3 in the cochlear SV The mRNA expression of TMEM16A, Bax, Bcl-2 and Caspase-3 was analyzed in cochlear SV from the different groups (Fig. 3).With the increase in age, the expression of TMEM16A mRNA in cochlear SV was increased (n = 3, *** P < 0.001).The expression of TMEM16A mRNA in the 12 m OVX group was higher than that in the 12 m group (n = 3, # P < 0.05).E2 intervention can downregulate the expression of TMEM16A mRNA in cochlear SV (n = 3, &&& P < 0.001).Then, the expression of apoptosis-related protein mRNA was detected.The expression of Caspase-3 and Bax mRNA was increased in the cochlear SV of the 12 m group (n = 3, * P < 0.05), and Bcl-2 mRNA was signi cantly decreased (n = 3, *** P < 0.001).The expressions of Caspase-3 and Bax mRNA in the 12 m OVX group were higher than those in the 12 m group (n = 3, ## P < 0.01, ### P < 0.001), and Bcl-2 mRNA was signi cantly decreased (n = 3, # P < 0.05).
After the intervention of E2, the expression of Caspase-3 and Bax mRNA decreased signi cantly (n = 3, & P < 0.05, &&& P < 0.001), and Bcl-2 mRNA was signi cantly increased (n = 3, &&& P < 0.001).The above results showed that with increasing age, the expression of apoptosis in cochlear SV was increased, while E2 can inhibit the apoptosis of cochlear SV.To further explore whether E2 can inhibit the apoptosis of cochlear SV PCs by regulating TMEM16A, we carried out experiments at the cellular level.

Primary culture and identi cation of PCs determined the cell aging model
In the in vitro cell experiments, we performed primary culture of PCs and established an aging cell model.
The studies revealed the growth state of PCs under an ordinary optical microscope (Fig. 4A).On the second day, the cochlear stria vascular tissue fragments were apparent.On the 7th day, the PCs had begun to grow out from the cochlear stria vascular tissue fragments and showed typical long spindle or few polygonal cells.The cells proliferated rapidly throughout the culture dish by day 12, reaching 90% density and passaging.Primary cultured PCs were observed using a uorescence inverted microscope (×400, Fig. 4B).In the primary culture of PCs, the expression of Desmin and α-SM-actin was positive (green), and there was no observed expression of vWF (a speci c marker of endothelial cells).Thus, we generated pure primary cultures of PCs.PCs were stained with β-galactosidase after continuous passage (Fig. 4C).In the P13 group, the cultured cells were obviously aged, and the number of positive cells was signi cantly higher than that in the P5 group (n = 3, *** P < 0.001).Therefore, we selected the P13 PCs as the aging cell group.

Fig. 4. Primary culture of PCs and analysis of the cell aging group
A: Primary culture of PC growth processes under an inverted microscope (×200).B: Identi cation of primary cultured PCs.Fluorescence inverted microscope (×400).The expression of desmin and α-SMactin was positive (green).The EC-speci c marker vWF was negative.C: β-galactosidase staining of PCs.

Immuno uorescence detection of TMEM16A expression
We used immuno uorescence to observe the expression of TMEM16A in different groups of PCs (Fig. 5).Compared with the P5 group, the results showed that the expression of TMEM16A was increased in the P13 group (n = 6, ** P < 0.01).When the P13 PCs were treated with E2, the expression of TMEM16A was decreased (n = 6, ## P < 0.01).These results indicated that E2 could reduce the expression of TMEM16A in the aging PC group.

Apoptosis was detected by ow cytometry
To further determine whether E2 has an effect on PC apoptosis through TMEM16A, we detected the apoptosis rate by ow cytometry (Fig. 6).Compared with the P5 group, the apoptosis rate in the P13 group was signi cantly increased (n =3, *** P < 0.001).When the P13 PCs were treated with E2 or T16Ainh-A01, the apoptosis rate was decreased (n =3, ## P < 0.01, and ### P < 0.001), and after intervention with both T16Ainh-A01 and E2, the rate was signi cantly decreased (n =3, ### P < 0.001).According to the results of this experiment and previous immuno uorescence results, it was found that E2 could inhibit the apoptosis of PCs through downregulating the expression of TMEM16A.

Discussion
Presbycusis is the most common sensory disorder in the elderly, and approximately 1/3 of people over the age of 65 suffer from presbycusis (Homans et al. 2017).There is currently still a lack of effective prevention and treatment measures for presbycusis.In our study, we further explored the possible mechanism of E2 in protecting against presbycusis.The results showed that E2 can inhibit apoptosis of PCs in the cochlear SV through downregulating the expression of TMEM16A, thus improving the hearing loss of aged C57BL/6J mice.
Presbycusis is a type of sensorineural hearing loss caused by the degeneration of the inner ear structure and bilateral symmetry.The etiology of presbycusis is complex and has not been clearly studied (Frisina 2009;Ohlemiller et al. 2010).At present, it is believed that one of the fundamental causes of strial presbycusis is the dysfunction caused by the degeneration of SV (Spicer and Schulte 2005).Early studies have reported cochlear SV contraction in gerbils, and the contraction rate increases with age (Gratton and Schulte 1995).When the atrophy of the cochlear SV reaches a certain extent, the capillaries and PCs are reduced or lost (Ocho et al. 2000), the PC cell structure is abnormal, the cytoplasm and chromatin are lost, and the cell vacuoles are increased (Neng et al. 2015).Our results showed that in the 12 m group, the ABR threshold was increased, which was consistent with the process of presbycusis.The cochlear SV of 12month-old mice showed a large degree of vacuolar degeneration, loss of chromatin organelles and cytoplasmic swelling in PCs.It was veri ed that with increasing age, the morphological structure changes of cochlear SV PCs may lead to age-related hearing loss.Some studies have shown that estrogen (E2) has a protective effect on hearing (Hederstierna et al. 2007).The ABR threshold of the auditory brainstem response in postmenopausal women is signi cantly higher than that in young women and men (Jerger and Johnson 1988;Wharton and Church 1990), suggesting that this difference may be related to the level of E2.Animal experiments have shown that the hearing threshold in ovariectomized rats is signi cantly higher than that in normal rats (Coleman et al. 1994).In our experiment, we found that after mice were ovariectomized, the ABR threshold was signi cantly increased, the cochlear SV was obviously atrophic, and the PC degeneration was more obvious.In addition, after the intervention of E2, the ABR threshold was decreased and the cochlear SV and PC degeneration were improved.Thus, our data demonstrated that E2 may reduce hearing loss in presbycusis by improving cochlear SV atrophy and the morphological structure of PCs in aging mice.Some studies have reported that E2 has the effect of anti-apoptosis, and it has been proven that estrogen can protect the heart by inhibiting the apoptosis of human umbilical vein endothelial cells induced by endoplasmic reticulum stress (ERS) as an anti-apoptotic agent (Su et al. 2017).E2 can also reduce the apoptosis of cochlear outer hair cells against gentamicin ototoxicity through inhibition of the JNK pathway (Nakamagoe et al. 2010).As many studies on presbycusis have found that it can cause cochlear cell degeneration, autophagy and apoptosis were increased (Huang et al. 2017;Sanchez-Rodriguez et al. 2018;Yuan et al. 2018).Based on the protective effect of E2 on cochlear apoptosis, our animal experiments showed that the apoptosis of cochlear SV increased in ovariectomized mice, and E2 intervention decreased the apoptosis.In the cell experiment, it was further found that the apoptosis rate of PCs increased signi cantly in the aging group and decreased signi cantly after the intervention of E2.
TMEM16A is an important constituent protein of CaCCs (Forge and Wright 2002).In the study of hearing and TMEM16A, it was found that TMEM16A was expressed in cochlear spiral ganglion cells (Zhang et al. 2015), cochlear SV basal cells (Jeon et al. 2011) and cochlear inner supporting cells (Wang et al. 2015).Early studies in our laboratory showed that there were also age-related TMEM16A changes in the PCs of guinea pig cochlear SV (Zhou et al. 2019).In our study, the expression of TMEM16A mRNA in the cochlear SV of aging C57BL/6J mice was increased.Cell-level studies have shown that TMEM16A was also expressed in primary cultured PCs and that its expression increased in the aging cell group, which was in accordance with the animal experiment.The above results showed that the change in TMEM16A expression was related to presbycusis.The expression and function of TMEM16A vary with cell type.Some studies have shown that TMEM16A is closely related to the apoptosis of cancer (Wilkerson and Reis-Filho 2013), suggesting that TMEM16A may be a potential biomarker of some aggressive diseases (Deng et al. 2016;Song et al. 2018;Frobom et al. 2019).For normal cells of the body, such as podocytes (Lian et al. 2017) and smooth muscle cells (Zeng et al. 2018;Zeng et al. 2019), overexpression of TMEM16A can promote apoptosis of these cells and further lead to the occurrence of disease.However, it has not been reported whether TMEM16A can affect presbycusis by regulating cochlear cell apoptosis.Further experiments showed that the expression of TMEM16A and apoptosis increased in the aging PC group.After treatment with T16Ainh-A01, the apoptosis rate of PCs in the aging group decreased signi cantly.These results suggest that TMEM16A is related to the apoptosis of PCs of cochlear SV and that inhibition of TMEM16A expression can reduce the apoptosis of PCs.In addition, we also found that the expression of TMEM16A decreased signi cantly when the ovariectomized mice were treated with E2.At the cellular level, we found that the expression of TMEM16A in the PCs of the aging group decreased after E2 intervention.Combined with the previous effect of E2 on PC apoptosis, it is suggested that E2 may inhibit the apoptosis of PCs in the aging cochlear SV through downregulating the expression of TMEM16A.
However, this study reveals that E2 may only partially affect presbycusis by regulating the effect of TMEM16A on apoptosis.Understanding how E2 speci cally regulates the expression or function of TMEM16A and the associated molecular mechanism are the issues that need to be addressed in the future.Nevertheless, the present ndings provide a clearer understanding of the effects of E2 on the internal microcirculation and auditory function of the cochlea, and they provide a theoretical basis for the use of E2 as a potential drug in the prevention and treatment of presbycusis.

Conclusions
In summary, due to the increased expression of TMEM16A, which can lead to an increase in the apoptosis of PCs, estrogen can downregulate the expression of TMEM16A and can also reduce the apoptosis of PCs.These data suggest that estrogen may reduce apoptosis in PCs by downregulating the expression of TMEM16A, which provides a protective effect on presbycusis.vs. the 3 m group.#P < 0.05, ##P < 0.01, ###P < 0.001 vs. 12 m group.&P < 0.05, &&&P < 0.001 vs. 12 m OVX group.Figure 6 Experimental grouping and treatment Animals were provided by the Animal Experiment Center of Beijing Vital River Laboratory Animal Technology Co. (animal use license batch no.SCXK Beijing 2016-0006).The use of animals in the present study was approved by the Committee of Animal Experimental Ethics of The First A liated Hospital of Medical College, Shihezi University (permit no.A2018-160-01).A total of 40 C57BL/6J female mice were divided into the following four groups (10/group): 3-month-old (3 m) group, 12-month-old (12 m) group, 12-month-old ovariectomized (12 m OVX) group (OVX at the age of 9 months) and 12 m OVX group treated with E2 (OVX at the age of 9 months and after 1 month treated with E2 [0.10 mg/kg/d] until the age of 12 months(Wu et al. 2016)).

Fig
Fig. 2A HE staining of the cochlear SV in each group.

Fig
Fig. 2A1 to D1 (inverted microscope, ×100) and Fig. 3A2 to D2 (inverted microscope, ×400) show the HE staining images of the changes in the cochlear SV.

Fig
Fig. 2B Morphological structure of cochlear SV PCs under transmission electron microscopy

Fig. 6
Fig. 6 The apoptosis rate of PCs

Figures
Figures