Disruption of Gap Junction-Mediated Intercellular Communication in the Spiral Ligament Causes Hearing and Outer Hair Cell Loss in the Cochlea of Mice

Abstract

It is well-known that outer hair cell (OHC) loss occurs in the cochlea of animal models of permanent hearing loss induced by intense noise exposure. Our earlier studies demonstrated the production of hydroxynonenal and peroxynitrite, as well as the disruption of gap junction-mediated intercellular communication (GJIC), in the cochlear spiral ligament prior to noise-induced sudden hearing loss. The goal of the present study was to evaluate the mechanism underlying cochlear OHC loss after sudden hearing loss induced by intense noise exposure. In organ of Corti explant cultures from mice, no significant OHC loss was observed after in vitro exposure to 4-hydroxynonenal (a product of lipid peroxidation), H₂O₂, SIN-1 (peroxynitrite generator), and carbenoxolone (a gap junction inhibitor). Interestingly, in vivo intracochlear carbenoxolone injection through the posterior semicircular canal caused marked OHC and hearing loss, as well as the disruption of gap junction-mediated intercellular communication in the cochlear spiral ligament. However, no significant OHC loss was observed in vivo in animals treated with 4-hydroxynonenal and SIN-1. Taken together, our data suggest that disruption of GJIC in the cochlear lateral wall structures is an important cause of cochlear OHC loss in models of hearing loss, including those induced by noise.

INTRODUCTION

The cochlea is a spiraled hollow/conical chamber of bone. The spiral canal of the cochlea is a section of the bony labyrinth of the inner ear. The cochlea has three chambers including the scala vestibule (vestibular duct), scala tympani (tympanic duct), and scala media (cochlear duct). The scala vestibule and scala tympani contain perilymph, whereas the scala media contains endolymph. The organ of Corti is the sensory epithelium and a cellular layer on the basilar membrane, in which sensory hair cells are powered by the potential difference between the perilymph and the endolymph. The hair cells include outer hair cells (OHCs) and inner hair cells. The cochlear lateral wall, located lateral to the cochlear sensory epithelium in the scala media, includes the spiral ligaments and stria vascularis and contributes to the maintenance of homeostasis in the cochlear endolymph (Fig. 1).

Fig. 1. The Cochlear Structure, in Vivo Intracochlear Injection, and in Vitro Explant Culture

In vivo intracochlear injection in the current study; the drug infused from the posterior semicircular canal is directly distributed within the perilymph in the scala vestibule and scala tympani. In vitro explant cultures of the organ of Corti, in which the organ of Corti and basilar membrane were dissected and cultured on cell culture inserts in culture medium.

Gap junctions are clusters of transmembrane channels that contribute to the intercellular exchange of ions and small molecules including secondary messengers and signaling molecules. A considerable number of reports has demonstrated that gap junction-mediated intercellular communication (GJIC) plays an essential role in a variety of cellular functions such as growth, migration, differentiation, and survival.^1–4) Gap junctions are formed by a family of connexin proteins, of which more than 20 different proteins have been identified in a variety of cell types. The cochlear gap junction is formed mainly by 2 types of connexins, namely connexin26 and connexin30, in non-sensory cells such as those from the lateral wall and spiral limbus, which are formed by various types of fibrocytes.^5–7) The cochlear non-sensory cells are connected by gap junctions that are involved in ionic exchange and biochemical coupling. Functional changes in GJIC are known to be involved in various pathological conditions including the attenuation or exacerbation of damage in various cells including neural cells and cancer cells.^8–11)

Noise-induced hearing loss is a well-known irreversible occupational disorder that is associated with severely diminished QOL; it is caused by the sudden exposure to intense noise generated by various accidents and the repeated and prolonged exposure to noise generated in the environment or workspace.¹²⁾ Furthermore, recent clinical studies have demonstrated that hearing loss increases the risk to dementia and mental decline.¹³⁾ Thus, in an aging society, it is important to elucidate the mechanisms underlying hearing loss to develop novel therapeutics. We previously established animal models of various types of noise-induced hearing loss, such as sudden hearing loss¹⁴⁾ and progressive hearing loss.¹⁵⁾ These earlier studies demonstrated an increase in reactive oxygen species (ROS)-induced 4-hydroxinonenal (4-HNE; a lipid peroxidation product)-adducts and reactive nitrogen species (RNS)-mediated tyrosine nitration products including peroxynitrite anion and nitrogen dioxide in the cochlear lateral wall structures and organ of Corti in animal models of sudden hearing loss induced by exposure to intense noise.¹⁴⁾ In addition, intense noise exposure disrupts GJIC through the internalization and downregulation of connexins in the spiral ligament fibrocytes of the cochlear lateral wall structure.¹⁶⁾ Many reports have also demonstrated cochlear OHC loss concomitant with permanent hearing loss induced by intense noise.^14,17) In addition, the OHC is well-known to have poor regenerative ability after damage.¹⁸⁾ Therefore, it is important to elucidate the mechanisms underlying hair cell loss. However, to date, the mechanism underlying OHC loss in the organ of Corti after intense noise exposure has not been elucidated. Thus, the goal of the present study was to evaluate this phenomenon after sudden hearing loss. In particular, we evaluated whether ROS/RNS and the disruption of GJIC are involved in OHC loss.

MATERIALS AND METHODS

Materials

4-HNE was purchased from Cayman Chemical (Ann Arbor, MI. U.S.A.). Carbenoxolone (CBX), which is a modestly potent, reasonably effective, water-soluble gap junction inhibitor,¹⁹⁾ was purchased from LKT Laboratories, Inc. (St. Paul, MN, U.S.A.). Neomycin trisulfate salt hydrate was purchased from Sigma-Aldrich Co. (St. Louis, MO, U.S.A.), Alexa Fluor 488-conjugated anti-rabbit immunoglobulin G (IgG) (H + L) antibody, Alexa Fluor 568™ Phalloidin (Alexa Fluor 568-conjugated Phalloidin), N-2 supplement (×100), and B-27 supplement (×50) were purchased from Thermo Fisher Scientific Inc. (Waltham, MA, U.S.A.). An anti-myosin VIIa rabbit polyclonal antibody was purchased from Proteus BioSciences Inc. (Ramona, CA, U.S.A.). Dako REAL Antibody Diluent was purchased from Dako Denmark A/S (Glostrup, Denmark). Calcein-AM and Blocking One Histo were purchased from Nacalai Tesque, Inc. (Kyoto, Japan). VECTASHIELD was purchased from Vector Laboratories (Burlingame, CA, U.S.A.). All other chemicals used were of the highest purity commercially available.

Animal Care and Drug Administration

The protocol used was approved by the Committee for Ethical Use of Experimental Animals at Setsunan University. All efforts were made to minimize animal suffering, to reduce the number of animals used, and to use alternatives to in vivo techniques. Mice were housed in plastic cages in a room with a 12-h/12-h light–dark cycle and a humidity of 55% at 23°C, and were given free access to food and water.

Each drug was administered to the inner ear via injection into the posterior semicircular canal. A retroauricular incision was made in the left ear to expose the posterior semicircular canal under anesthesia with chloral hydrate (500 mg/kg, intraperitoneally (i.p.)). Under microscopy, a small hole was made in the bony wall of the posterior semicircular canal to insert a fused silica glass needle (EiCOM, Kyoto, Japan) into the perilymphatic duct of the posterior semicircular canal (Fig. 1). Five microliters of either vehicle, 3 mM CBX, 250 µM 4-HNE, or 250 µM SIN-1 was injected at a flow rate of 1 µL/min using a micro-syringe pump. The doses of drugs injected were denoted as “mol/ear.” To determine the mechanical damage of intracochlear injection, we injected vehicle as a control under the same experimental conditions.

Noise-Induced Sudden Hearing Loss in Animals

Male ddY mice at 5 weeks of age were used to prepare an animal model of noise-induced sudden hearing loss, according to earlier reports.¹⁴⁾ To remove animals with natural auditory impairment, we measured their auditory brainstem response (ABR) before use and selected those animals with normal acoustic sense in the present study. Each animal was anesthetized with chloral hydrate (500 mg/kg, i.p.), placed in a plastic cage, and exposed to 110-dB sound pressure level (SPL) of octave band noise (centered at 8 kHz) for 1 h within the sound chamber. The sound chamber was fitted with a speaker (300HT; FOSTEX, Tokyo, Japan) driven by a noise generator (SF-06; RION, Tokyo, Japan) and power amplifier (DAD-M100proHT; FLYING MOLE, Shizuoka, Japan). To ensure stimulus uniformity, we calibrated the SPL using a sound-level meter (NL-26; RION, Tokyo, Japan), which was positioned at the level of the animal’s head. As a control, naïve animals were placed in the same cage without the noise.

ABR Measurement

For ABR measurements, steel-needle electrodes were placed at the vertex, ventrolateral to the left and right ears, while animals were under anesthesia via isoflurane inhalation. Electroencephalogram recordings were performed with an extracellular amplifier Digital Bioamp system BAL-1 (Tucker-Davis Technologies, FL, U.S.A.), and waveform storing and stimulus control were performed using Scope software of the Power Lab system Power Lab 2/20 (AD Instruments, Castle Hill, Australia). Sound stimuli were produced by a coupler-type speaker ES1spc (Bioresearch Center, Nagoya, Japan) inserted into the external auditory canal of the mouse. Tone burst stimuli [0.1 ms rise/fall time (cosine gate) and 1-ms flat segment] were generated using a Real Time Processor RP2.1 (Tucker-Davis Technologies). SPLs were quantified using a Programmable Attenuator PA5 (Tucker Davis Technologies) and calibrated with a sound-level meter, Type 6224 (ACO Co., Ltd., Tokyo, Japan). ABR waveforms were recorded for 12.8 ms at a sampling rate of 20000 Hz using 100–10000 Hz band bypass filter settings. Waveforms from more than 500 stimuli were averaged. For recordings, animals were anesthetized via isoflurane inhalation. The threshold ABR was determined before and on days 1 and 7 after treatment, at frequencies of 4, 12, and 20 kHz, using a 5-dB SPL minimum size step-down from the maximum amplitude. The hearing threshold was defined as the lowest stimulus intensity that produced a reliable wave I ABR. Because test tones were set to SPLs of less than 100 dB at all frequencies, the thresholds were recorded as 100 dB for the calculation of the threshold shift value when there was no response due to profound hearing impairment.

Fluorescence Recovery after Photobleaching (FRAP) Assay

Mice were administered either vehicle or drug to dissect the cochlear spiral ligament at various times after treatment. The dissected cochlear spiral ligament was plated on poly-L-lysine-coated dishes containing Dulbecco’s modified Eagle’s medium (DMEM) supplemented 10% fetal bovine serum at 37°C in a 5% CO₂/95% air, humidified incubator for 4 h. For FRAP, spiral ligament tissue cultures were preloaded for 20 min with calcein-AM, which is a permeable material for gap junctions. After calcein-AM was preloaded, tissues were bleached by pulsed laser (405 nm) irradiation for 15 s using a confocal laser operated microscope (FLUOVIEW-FV1000, Olympus, Tokyo, Japan). Thereafter, FRAP was monitored every 2 s for 180 s. For analysis, we defined a region of interest inside the bleached area for each time point. To determine the ratio of fluorescence recovery, we measured the fluorescence intensity at 22 and 180 s and calculated a percentage of difference between the fluorescence intensity at 22 and 180 s in the tissues dissected from animals treated with vehicle and CBX.

Explant Cultures of the Organ of Corti

The organ of Corti was dissected from the cochlea of 4-d-old (P4) ddY mice under microscopy, placed individually on transparent membranes with 0.3-µm pores, and maintained in a 24-well culture plate containing 500 µL/well culture medium, which consisted of DMEM, 0.3% (w/v) glucose, and 0.03% (w/v) penicillin G. The explants were placed in a humidified incubator for 24 h at 37°C and 5% CO₂ and replaced with the same medium containing individual drugs for histological assessments after 24-h treatments.

For 7-d organ of Corti explant cultures, we used culture medium consisting of DMEM, 0.3% (w/v) glucose, 0.03% (w/v) penicillin G, B-27 supplement (×50), and N-2 supplement (×100). The explants were placed in a humidified incubator at 37°C and 5% CO₂ for 24 h, and the same medium containing individual drugs was replaced every other day for histological assessments.

Histological Assessment

For the in vivo determination of OHC, cochleae were removed quickly after decapitation. The round window, oval window, and apex of the cochlea were opened, perfused with 4% paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.4), and subsequently kept at 4°C overnight in the same solution. The fixed cochlea was washed three times with phosphate-buffered saline (pH 7.4), and then incubated for at least 7 d in 4% ethylenediaminetetraacetic acid solution for demineralization. To visualize hair cells in the organ of Corti, we incubated the epithelium of this tissue on the basal and middle turns of the cochlea with a solution containing 0.03% triton X-100 in tris-buffered saline (TBST) and Alexa Fluor 568-conjugated Phalloidin (1 : 200 dilution) for 30 min at room temperature in the dark. After washing three times with 0.03% TBST, the stained specimens were mounted onto slides with VECTASHIELD, an antifade mounting medium, and then observed under a confocal fluorescence microscope using the FV1000D system (Olympus); the numbers of missing hair cells in the mid part of the cochleae were then counted. The ratio of missing-to-whole hair cells was expressed as a percentage.

For immunostaining, organ of Corti explant cultures were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer for 20 min and washed with 0.2% triton X-100 in phosphate-buffered saline (pH 7.4, PBST). After incubation for 20 min at room temperature, samples were blocked with Blocking One Histo and then incubated with primary antibody against myosin VIIa (1 : 200) at 4°C overnight. They were then washed with PBST and incubated with Alexa Fluor 488-conjugated anti-rabbit IgG secondary antibody (1 : 200) for 2 h at room temperature. After washing with PBST, samples were incubated with Alexa Fluor 568-conjugated Phalloidin (1 : 200) for 20 min in the dark. Specimens were finally mounted on slides with antifade mounting medium (VECTASHIELD) for observation under a confocal fluorescence microscope using the FV1000D system (Olympus). A series of 8–12 laser confocal images was taken for each section of the organ of Corti at depth intervals of 0.75 µm. The images are presented using projection sets of 8–12 images over the Z-axis.

Data Analysis

All data are expressed as the mean ± standard error of the mean (S.E.M.), and statistical significance was determined by performing a two-tailed Student t-test and a one-way ANOVA with a Bonferroni–Dunnett post hoc test.

RESULTS

Cochlear OHC Loss Is Induced by Exposure to Intense Noise

Our previous studies showed that a single exposure to intense noise (110 dB, 1 h) under the present experimental conditions could produce severe hearing impairment at the frequencies of 4, 12, and 20 kHz immediately to at least day 7 after exposure.^14,20,21) To assess OHC loss after the exposure to 110 dB for 1 h, we stained the organ of Corti, which was dissected immediately (time = 0) or up until day 14 after exposure, with Alexa Fluor 568-conjugated Phalloidin (Fig. 2). Noise exposure caused significant OHC loss in the middle turn of the cochlea on day 2 onward and progressively increased the ratio of OHC loss, peaking on days 7 and 14 post-exposure. These data suggest that noise exposure under these experimental conditions can cause delayed and permanent OHC loss in the middle turn of the cochlea.

Fig. 2. Time Course of Outer Hair Cells (OHC) Loss Following in Vivo Noise Exposure

Animals were exposed to noise at 8 kHz octave band noise 110-dB sound pressure level for 1 h. The cochlea was fixed on the indicated days post-noise exposure for dissection of the organ of Corti, which was stained with Alexa Fluor 568-conjugated Phalloidin to examine OHCs. Left panels denote typical microscopic images of the middle turn in the organ of Corti. Scale bar = 50 µm. The right graph denotes the percentage of OHC loss per total OHCs in the middle turn of the organ of Corti. Values are the means ± S.E.M. from five independent experiments. ** p < 0.01, significantly different from the value obtained for naïve animals (N).

Effect of 4-HNE, H₂O₂, and SIN-1 on Cochlear OHCs in Vivo and in Vitro

An earlier report demonstrated that noise exposure under the present experimental conditions produced 4-HNE and nitrotyrosine in the cochlea immediately after exposure and that noise-induced hearing and cochlear OHC loss could be prevented by an antioxidant and nitric oxide synthase inhibitor.¹⁴⁾ To determine whether direct treatment with 4-HNE or peroxynitrite anion causes OHC loss in the cochlea, we initially established an explant culture system using the organ of Corti dissected from P4 mice and exposed the explant cultures to 4-HNE and the peroxynitrite generator SIN-1.

Since neomycin is well-known to damage cochlear OHCs under in vivo and in vitro experimental conditions,^22,23) we used this compound as a positive control to establish explant cultures and evaluate OHC loss (Fig. 3). In explant cultures exposed to vehicle for 24 h, OHCs positive for myosin VIIa (hair cell marker) and phalloidin (F-actin marker) were regularly observed. As expected, treating explant cultures with neomycin for 24 h markedly decreased myosin VIIa-positive cells, suggesting that OHCs, under the present culture conditions, are sensitive to this compound (Fig. 3b).

Fig. 3. Explant Cultures of the Organ of Corti

(a) Protocol of explant culture: the organ of Corti was dissected from the cochlea of P4 mice. It was then placed on the cell culture insert, cultured for 24 h, and then exposed to either vehicle or each drug in culture medium. After incubation for various times, hair cells in the explant cultures were visualized by immunostaining for myosin VIIa and phalloidin staining. (b) The explant cultures were incubated with either vehicle or neomycin (1 mM) for 24 h and then subjected to immunostaining for myosin VIIa (green) and phalloidin staining (red). These experiments were carried out at least three times, with similar results were obtained under the same experimental conditions. Scale bar = 100 µm. (Color figure can be accessed in the online version.)

Exposing explant cultures to either 4-HNE, H₂O₂, or SIN-1 for 24 h at concentrations of 2–50 µM did not cause significant loss of cochlear OHCs (Table 1). In addition to the 24-h exposure, long-term exposure (7 d) to 4-HNE and SIN-1 at an even higher concentration (250 µM) did not affect OHC survival [OHC loss (% of control): vehicle, 5.4 ± 0.6; 4-HNE, 7.3 ± 0.8; SIN-1, 6.5 ± 1.2]. These data suggest that 4-HNE and peroxynitrite anion have no direct toxic effect on OHCs, at least under the experimental conditions tested. Next, we tested whether in vivo intracochlear injection of 4-HNE and SIN-1 could cause OHC loss in the cochlea. Intracochlear injection of neither 4-HNE (1.25 nmol/ear) nor SIN-1 (1.25 nmol/ear) caused OHC loss in the middle turn of the organ of Corti on at least day 7 post-treatment [OHC loss (% of total): vehicle, 2.2 ± 0.5; 4-HNE, 5.6 ± 2.9; SIN-1, 3.9 ± 2.2 (n = 3–7, not significant)].

Table 1. Effect of in Vitro 4-HNE, H₂O₂, and SIN-1 Treatment on the Survival of Outer Hair Cells (OHCs) under Explant Culture Conditions

Concentration (µM)	OHC loss (% of total)
	4-HNE	H2O2	SIN-1
0	5.7 ± 1.1	5.8 ± 1.6	4.8 ± 2.0
2	8.1 ± 1.7N.S.	4.2 ± 1.8N.S.	4.5 ± 1.9N.S.
10	4.6 ± 1.0N.S.	6.4 ± 3.2N.S.	6.1 ± 2.0N.S.
50	9.1 ± 3.5N.S.	5.1 ± 1.5N.S.	7.2 ± 2.9N.S.

Explant cultures were incubated with either vehicle, 4-HNE, H₂O₂, or SIN-1 at the indicated concentrations for 24 h; they were then fixed and stained with Alexa Fluor 568-conjugated Phalloidin. Values are presented as the means ± S.E.M. from 3–5 independent experiments. N.S.: not significant vs. vehicle alone (concentration = 0) for each exposure time.

Effect of CBX on Cochlear OHCs in Vivo and in Vitro

Our earlier studies showed that noise exposure under the present experimental conditions disrupts ion-trafficking systems such as GJIC and Na⁺ pumps in the spiral ligament prior to permanent hearing loss.^16,21) To elucidate whether gap junctions are involved in OHC loss in the cochlea, we administered the gap junction blocker CBX via in vivo intracochlear injection, and also in vitro through direct exposure using organ of Corti explant cultures.

Figure 4 shows the disruption of gap junctions in the spiral ligament after in vivo intracochlear injection of CBX based on FRAP assays. In vehicle-treated animals, pulsed laser irradiated areas (white circles) were bleached, in terms of the fluorescence intensity of calcein-AM, at 22 s after irradiation. At 180 s after irradiation, the gap junction-permeable calcein-AM returned to levels approximately 80% those before irradiation, indicating that GJIC in the spiral ligament dissected from the cochlea of vehicle-treated animals was virtually intact. However, as expected, the spiral ligament dissected from the cochlea of CBX-treated animals exhibited the marginal return of calcein-AM at 180 s after irradiation. These results suggest that in vivo intracochlear injection of CBX disrupts GJIC in the spiral ligament of the cochlea.

Fig. 4. Carbenoxolone (CBX)-Mediated Disruption of Gap Junction-Mediated Intracellular Communication (GJIC) in the Cochlear Spiral Ligament

Animals were administered either vehicle or CBX (15 nmol/ear) into the semicircular canal. At 24 h post-treatment, the spiral ligament tissues of the cochlear basal/middle turn were prepared and then subjected to FRAP assays to assess GJIC. Upper panels denote the typical microscopic images of the spiral ligament tissue before or after irradiation with the laser. White circles indicate the area irradiated with the laser. The lower graph shows fluorescence intensities in the irradiated area at contiguous times after irradiation, as a percentage of the value prior to irradiation (time = 0). Values are means ± S.E.M. from 6–8 independent experiments. * p < 0.05, significantly different between each value at 180 s post-radiation, obtained from animals treated with vehicle and CBX.

To determine whether CBX directly causes OHC loss in the cochlea, we treated organ of Corti explant cultures with CBX for 7 d. CBX at 300 µM did not cause OHC loss in the explant cultures (Fig. 5a). Next, we examined the effect of in vivo CBX treatment on cochlear OHCs. Surprisingly, intracochlear injection of CBX caused marked OHC loss in the cochlea 7 d post-treatment (Fig. 5b).

Fig. 5. Effect of Carbenoxolone (CBX) on Cochlear Outer Hair Cells (OHCs) in Vivo and in Explant Cultures

(a) Explant culture: explant cultures of the organ of Corti were prepared from P4 mice. The explant cultures were incubated in medium containing either vehicle or CBX (300 µM) for 7 d and stained with Alexa Fluor 568-conjugated Phalloidin to assess OHCs. (b) In vivo experiments: animals were administered either vehicle or CBX (15 nmol/ear) into the semicircular canal. On day 7 post-treatment, the cochlea was fixed for dissection of the organ of Corti, which was stained with Alexa Fluor 568-conjugated Phalloidin to assess OHCs. Graphs denote the percentage of OHC loss per total OHCs in the middle turn of the organ of Corti. Values are the means ± S.E.M. from four independent experiments. ** p < 0.01, significantly different between each value obtained for animals treated with vehicle and CBX.

Time–Course of OHC Loss and Disruption of GJIC Following in Vivo CBX Treatment

Next, we assessed OHC loss at various time points after in vivo CBX treatment. Whereas marked loss of OHCs was observed on day 7 post-treatment, as indicated in Fig. 5b, no significant OHC loss was observed from at least day 1 to 5 post-treatment (Fig. 6a). These results suggest that CBX can cause delayed OHC damage in the cochlea.

Fig. 6. Time Course of Outer Hair Cells (OHC) Loss and Disruption of Gap Junction-Mediated Intracellular Communication (GJIC) Following in Vivo Treatment with Carbenoxolone (CBX)

Animals were administered either vehicle or CBX (15 nmol/ear) into the semicircular canal. (a) On the indicated days post-treatment, the cochlea was fixed for dissection of the organ of Corti; basal/middle turns were stained with Alexa Fluor 568-conjugated Phalloidin to enumerate OHCs. The graph denotes the percentage OHC loss per total OHCs in the middle turn. Values are the means ± S.E.M. from four independent experiments. ** p < 0.01, significantly different between each value obtained for animals treated with vehicle and CBX. (b) On days 1 and 7 post-treatment, the spiral ligament tissues of the cochlear basal/middle turn were prepared and then subjected to FRAP assays to assess GJIC. The graph denotes fluorescence intensities of the irradiated area at 180 s post-irradiation as a percentage of the value prior to irradiation. Values are the means ± S.E.M. from 6–8 independent experiments. * p < 0.05, ** p < 0.01, significantly different between each value obtained for animals treated with vehicle and CBX.

Based on FRAP assays, in vivo treatment with CBX at 15 nmol/ear dramatically decreased the fluorescence recovery even on day 7, in addition to day 1 post-treatment (Fig. 6b). In addition, CBX at a dose of 5 nmol/ear was effective in decreasing the fluorescence recovery on day 1 [ratio of fluorescence recovery (%): vehicle, 82.6 ± 5.7; CBX, 62.1 ± 6.2 (p < 0.05)]. No significant change in the fluorescence recovery was seen after treatment with CBX at 5 nmol/ear on day 7 [ratio of fluorescence recovery (%): vehicle, 80.0 ± 5.6; CBX, 77.1 ± 6.1]. Importantly, the GJIC disruption by CBX at a lower dose (5 nmol/ear) was less than that with a dose of 15 nmol/ear on day 1 post-treatment [ratio of fluorescence recovery (%): 5 nmol, 62.1 ± 6.2; 15 nmol, 22.6 ± 12.7 (p < 0.01)]. These data suggest that CBX can mediate the disruption of GJIC, permanently in the case of high doses and in a dose-dependent manner, in the cochlear spiral ligament under the experimental conditions tested.

Effect of CBX on Hearing Ability

Finally, we examined the effect of intracochlear CBX injection on hearing ability at frequencies of 4, 12, and 20 kHz, as measured by ABR (Fig. 7). CBX treatment, even at a dose of 5 nmol/ear, significantly elevated the ABR threshold at all frequencies from at least day 1 to 7 post-treatment. Moreover, an increased dose resulted in an even higher ABR threshold. These results suggest that CBX can cause permanent (up to day 7) hearing loss in a dose-dependent manner at 5–25 nmol/ear.

Fig. 7. Effect of Carbenoxolone (CBX) on Hearing Ability in Vivo

Animals were administered either vehicle or CBX, at the indicated doses, into the semicircular canal. The auditory brainstem response (ABR) was measured at frequencies of 4, 12, and 20 kHz before and on days 1 and 7 after treatment. Graphs denote the average of ABR threshold shift at each frequency. Values are the means ± S.E.M. from 3–5 independent experiments. * p < 0.05, ** p < 0.01, significantly different from control value obtained for animals treated with vehicle.

DISCUSSION

The primary aim of the present study was to elucidate the mechanisms underlying OHC loss in the cochlea after exposure to intense noise. For this, we used an in vivo experimental model and explant cultures from the organ of Corti dissected from the cochlea of P4 mice for in vitro experiments. We focused on the generation of ROS/RNS and the disruption of GJIC as intracochlear events prior to noise-induced sudden hearing loss, and for the first time, provided evidence for the involvement of intracochlear GJIC disruption in OHC loss; this was based on the current data that in vivo disruption of GJIC elicited the dramatic loss of cochlear OHCs.

In the present study, we first focused on 4-HNE and RNS generated in the cochlea prior to sudden hearing loss induced by noise exposure. Our previous reports demonstrated that in vivo systemic treatment with a ROS-neutralizing agent or nitric oxide synthase inhibitor protected cochlear OHCs from damage under these conditions.¹⁴⁾ Further evidence for the involvement of excess ROS/RNS generated within the cochlea during OHC loss induced by acoustic overstimulation has come from numerous reports.^24–26) However, the current study showed that neither 4-HNE, H₂O₂ nor SIN-1 can effectively damage OHCs under our in vitro explant culture conditions. In addition to in vitro experiments, direct exposure of the cochlea to 4-HNE and SIN-1 in vivo, at least at the doses used, did not significantly damage cochlear OHCs. These findings led us to propose that ROS/RNS are not essential for OHC loss, in contrast to previous findings showing that a ROS neutralizing agent and nitric oxide synthase inhibitor can prevent noise-induced OHC loss. We speculated on these discrepancies as follows: OHC loss might occur in the presence of synergy between ROS/RNS and other events in the cochlea of animal models of hearing loss; this is because many events would be triggered by noise exposure prior to hearing loss. In this case, ROS/RNS might be required for OHC loss during noise-induced hearing loss. Further, events that are progressively induced by ROS/RNS might be involved in OHC loss. Although the current study does not fully resolve this dilemma, we did provide evidence that ROS/RNS are ineffective at directly damaging cochlear OHCs.

Second, we focused on cochlear GJIC, which is disrupted prior to hearing loss and cochlear OHC loss induced by noise exposure.¹⁶⁾ To examine the effect of GJIC disruption on hearing ability in the present in vivo study, we performed intracochlear administration of CBX, the amount of which was sufficient for the disruption of GJIC in the spiral ligament structures. In vivo treatment with CBX might cause at least two types of side-effects under the current experimental conditions. One is the mechanical damage by fluids flowing into the cochlea, whereas the other is the pharmacological side-effects of CBX itself. In the present study, the former was avoided by injecting vehicle as a control under the same experimental conditions. For the latter, the finding that CBX inhibits 11β-hydroxysteroid dehydrogenase, which reversibly catalyzes the conversion of cortisol to the inactive steroid cortisone,²⁷⁾ makes sense as the focus of the side-effect of carbenoxolone in the present study. However, a previous report demonstrated that the protein and mRNA expression of 11β-hydroxysteroid dehydrogenase cannot be detected in any of the inner ear tissues of adult rats.²⁸⁾ In light of these findings, the possibility that hearing loss mediated by CBX might be due to inhibition of 11β-hydroxysteroid dehydrogenase in inner ear tissues can be ruled out. Thus, it is likely that CBX-induced hearing loss is due to GJIC in the cochlea.

Earlier studies showed that the use of a calpain inhibitor to prevent noise-induced disruption of GJIC in the cochlear spiral ligament ameliorates cochlear OHC and hearing loss.²¹⁾ These findings led us to propose that disrupting GJIC in the spiral ligament causes OHC loss in the cochlea. As expected, in vivo intracochlear CBX treatment induced marked OHC loss and disruption of GJIC in the spiral ligament. However, CBX did not damage OHCs in in vitro explant cultures, which did not include the spiral ligament where GJIC plays a potential role for signal transduction that is derived from noise exposure. These data suggest that CBX-induced OHC loss is due to the disruption of GJIC in the spiral ligament, but not due to direct effects on the organ of Corti. In addition to previous studies, the current data led us to propose that the disruption of GJIC in the spiral ligament is essential for noise-induced OHC loss. In addition, this is strongly supported by the finding that in vivo CBX treatment caused severe hearing loss in a dose-dependent manner, similar to GJIC disruption by CBX.

GJIC plays an important role in maintaining the extracellular and intracellular ionic composition and content in numerous tissue structures. Particularly, GJIC in the cochlear spiral ligament is known to play important roles in hearing functions. Endolymph contains 150 mM K⁺, 2 mM Na⁺, and 20 µM Ca²⁺ and maintains the endocochlear potential at +80 mV, which is crucial for the maintenance of hearing ability.²⁹⁾ The role of GJIC is to maintain K⁺ at high concentrations in the endolymph of the cochlear duct to generate endocochlear potential. Endocochlear potential enhances the sensitivity of hair cells by increasing the driving force for K⁺ influx and Ca²⁺ permeation, which amplifies the motility of hair bundles.^30–32) Once K⁺ enters the hair cells, it must cycle back to the endolymph through an ion-trafficking system comprising Na⁺, K⁺-ATPase, ion transporters, and GJIC in supporting cells and lateral wall structures including the spiral ligament and stria vascularis.³¹⁾ Thus, it seems likely that the CBX-induced disruption of GJIC results in hearing loss through the attenuation of K⁺-recycling toward the stria vascularis via the spiral ligament in the cochlear lateral wall structures and decreased endocochlear potential. Therefore, our present proposition, specifically that disruption of GJIC by CBX induces hearing loss, would be feasible.

There are some reports on the anti-tumor activity of gap junction disruptors such as glycyrrhetinic acid, ursolic acid, and oleanolic acid. It is known that glycyrrhetinic acid inhibits gap junctions,³³⁾ whereas oleanolic acid and ursolic acid decrease the expression of connexin 32, which is a gap junction component.³⁴⁾ Pharmacological inhibition of GJIC by glycyrrhetinic acid results in cell damage in human hepatocellular carcinoma.³⁴⁾ A further study showed that connexin-mediated protection of the ARPE-19 human retinal pigment epithelial cell line in the presence of oxidative stress is abolished by glycyrrhetinic acid.³⁵⁾ Conversely, CBX is known to protect against cell damage in various cells.^36,37) However, the current findings that CBX indirectly causes OHC death does not account for the direct effect of GJIC disruptors in tissues and cells other than OHCs. Further studies are thus needed to elucidate the exact mechanism underlying GJIC disruption-induced OHC loss.

In conclusion, the present study clearly showed that disruption of cochlear GJIC induces OHC cell death. Thus, it is strongly anticipated that continued research will result in the development of drugs targeting the disruption of GJIC to treat sensorineural hearing loss including ototoxic-induced hearing loss and sudden hearing loss.

Acknowledgments

This work was supported in part by a Grant-in-Aid for Scientific Research to K.O. (#16K08288) and T.Y. (#17K15462) from the Japan Society for the Promotion of Science.

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Kumar NM, Gilula NB. The gap junction communication channel. Cell, 84, 381–388 (1996).
• 2) Sáez JC, Berthoud VM, Branes MC, Martinez AD, Beyer EC. Plasma membrane channels formed by connexins: their regulation and functions. Physiol. Rev., 83, 1359–1400 (2003).
• 3) Yao J, Zhu Y, Morioka T, Oite T, Kitamura M. Pathophysiological roles of gap junction in glomerular mesangial cells. J. Membr. Biol., 217, 123–130 (2007).
• 4) Yao J, Oite T, Kitamura M. Gap junctional intercellular communication in the juxtaglomerular apparatus. Am. J. Physiol., 296, F939–F946 (2009).
• 5) Ahmad S, Chen S, Sun J, Lin X. Connexins 26 and 30 are co-assembled to form gap junctions in the cochlea of mice. Biochem. Biophys. Res. Commun., 307, 362–368 (2003).
• 6) Forge A, Becker D, Casalotti S, Edwards J, Marziano N, Nevill G. Gap junctions in the inner ear: comparison of distribution patterns in different vertebrates and assessment of connexin composition in mammals. J. Comp. Neurol., 467, 207–231 (2003).
• 7) Kikuchi T, Kimura RS, Paul DL, Adams JC. Gap junctions in the rat cochlea: immunohistochemical and ultrastructural analysis. Anat. Embryol., 191, 101–118 (1995).
• 8) Azzam EI, de Toledo SM, Little JB. Direct evidence for the participation of gap junction-mediated intercellular communication in the transmission of damage signals from alpha-particle irradiated to nonirradiated cells. Proc. Natl. Acad. Sci. U.S.A., 98, 473–478 (2001).
• 9) Krutovskikh VA, Piccoli C, Yamasaki H. Gap junction intercellular communication propagates cell death in cancerous cells. Oncogene, 21, 1989–1999 (2002).
• 10) Nakase T, Fushiki S, Naus CC. Astrocytic gap junctions composed of connexin 43 reduce apoptotic neuronal damage in cerebral ischemia. Stroke, 34, 1987–1993 (2003).
• 11) García-Dorado D, Rodriguez-Sinovas A, Ruiz-Meana M. Gap junction-mediated spread of cell injury and death during myocardial ischemia-reperfusion. Cardiovasc. Res., 61, 386–401 (2004).
• 12) Alamgir H, Turner CA, Wong NJ, Cooper SP, Betancourt JA, Henry J, Senchak AJ, Hammill TL, Packer MD. The impact of hearing impairment and noise-induced hearing injury on quality of life in the active-duty military population: challenges to the study of this issue. Mil. Med. Res., 3, 11 (2016).
• 13) Livingston G, Sommerlad A, Orgeta V, Costafreda SG, Huntley J, Ames D, Ballard C, Banerjee S, Burns A, Cohen-Mansfield J, Cooper C, Fox N, Gitlin LN, Howard R, Kales HC, Larson EB, Ritchie K, Rockwood K, Sampson EL, Samus Q, Schneider LS, Selbæk G, Teri L, Mukadam N. Dementia prevention, intervention, and care. Lancet, 390, 2673–2734 (2017).
• 14) Nagashima R, Yamaguchi T, Tanaka H, Ogita K. Mechanism underlying the protective effect of tempol and N^ω-nitro-L-arginine methyl ester on acoustic injury: possible involvement of c-Jun N-terminal kinase pathway and connexin26 in the cochlear spiral ligament. J. Pharmacol. Sci., 114, 50–62 (2010).
• 15) Yamaguchi T, Yoneyama M, Onaka Y, Imaizumi A, Ogita K. Preventive effect of curcumin and its highly bioavailable preparation on hearing loss induced by single or repeated exposure to noise: a comparative and mechanistic study. J. Pharmacol. Sci., 134, 225–233 (2017).
• 16) Yamaguchi T, Nagashima R, Yoneyama M, Shiba T, Ogita K. Disruption of ion-trafficking system in the cochlear spiral ligament prior to permanent hearing loss induced by exposure to intense noise: possible involvement of 4-hydroxy-2-nonenal as a mediator of oxidative stress. PLOS ONE, 9, e102133 (2014).
• 17) Furness DN. Molecular basis of hair cell loss. Cell Tissue Res., 361, 387–399 (2015).
• 18) Kurabi A, Keithley EM, Housley GD, Ryan AF, Wong AC. Cellular mechanisms of noise-induced hearing loss. Hear. Res., 349, 129–137 (2017).
• 19) Connors BW. Tales of a dirty drug: carbenoxolone, gap junctions, and seizures. Epilepsy Curr., 12, 66–68 (2012).
• 20) Nagashima R, Yamaguchi T, Kuramoto N, Ogita K. Acoustic overstimulation activates 5′-AMP-activated protein kinase through a temporary decrease in ATP level in the cochlear spiral ligament prior to permanent hearing loss in mice. Neurochem. Int., 59, 812–820 (2011).
• 21) Yamaguchi T, Yoneyama M, Ogita K. Calpain inhibitor alleviates permanent hearing loss induced by intense noise by preventing disruption of gap junction-mediated intercellular communication in the cochlear spiral ligament. Eur. J. Pharmacol., 803, 187–194 (2017).
• 22) Liu L, Chen Y, Qi J, Zhang Y, He Y, Ni W, Li W, Zhang S, Sun S, Taketo MM, Wang L, Chai R, Li H. Wnt activation protects against neomycin-induced hair cell damage in the mouse cochlea. Cell Death Dis., 7, e2136 (2016).
• 23) Selimoglu E. Aminoglycoside-induced ototoxicity. Curr. Pharm. Des., 13, 119–126 (2007).
• 24) Böttger EC, Schacht J. The mitochondrion: a perpetrator of acquired hearing loss. Hear. Res., 303, 12–19 (2013).
• 25) Kamogashira T, Fujimoto C, Yamasoba T. Reactive oxygen species, apoptosis, and mitochondrial dysfunction in hearing loss. Biomed Res. Int., 2015, 617207 (2015).
• 26) Han WJ, Shi XR, Nuttall A. Distribution and change of peroxynitrite in the guinea pig cochlea following noise exposure. Biomed. Rep., 9, 135–141 (2018).
• 27) Duax WL, Ghosh D, Pletnev V. Steroid dehydrogenase structures, mechanism of action, and disease. Vitam. Horm., 58, 121–148 (2000).
• 28) Terakado M, Kumagami H, Takahashi H. Distribution of glucocorticoid receptors and 11 beta-hydroxysteroid dehydrogenase isoforms in the rat inner ear. Hear. Res., 280, 148–156 (2011).
• 29) Von Békésy G. Resting potentials inside the cochlear partition. Nature, 169, 241–242 (1952).
• 30) Weber PC, Cunningham CD 3rd, Schulte BA. Potassium recycling pathways in the human cochlea. Laryngoscope, 111, 1156–1165 (2001).
• 31) Wangemann P. K⁺ cycling and the endocochlear potential. Hear. Res., 165, 1–9 (2002).
• 32) Chan DK, Hudspeth AJ. Ca²⁺ current-driven nonlinear amplification by the mammalian cochlea in vitro. Nat. Neurosci., 8, 149–155 (2005).
• 33) Davidson JS, Baumgarten IM. Glycyrrhetinic acid derivatives: a novel class of inhibitors of gap-junctional intercellular communication. Structure–activity relationships. J. Pharmacol. Exp. Ther., 246, 1104–1107 (1988).
• 34) Satomi Y, Nishino H, Shibata S. Glycyrrhetinic acid and related compounds induce G1 arrest and apoptosis in human hepatocellular carcinoma HepG2. Anticancer Res., 25 (6B), 4043–4047 (2005).
• 35) Hutnik CM, Pocrnich CE, Liu H, Laird DW, Shao Q. The protective effect of functional connexin43 channels on a human epithelial cell line exposed to oxidative stress. Invest. Ophthalmol. Vis. Sci., 49, 800–806 (2008).
• 36) Du ZJ, Cui GQ, Zhang J, Liu XM, Zhang ZH, Jia Q, Ng JC, Peng C, Bo CX, Shao H. Inhibition of gap junction intercellular communication is involved in silica nanoparticles-induced H9c2 cardiomyocytes apoptosis via the mitochondrial pathway. Int. J. Nanomedicine, 12, 2179–2188 (2017).
• 37) Na YJ, Choi KJ, Park SB, Sung HR, Jung WH, Kim HY, Rhee SD, Kim KY. Protective effects of carbenoxolone, an 11β-HSD1 inhibitor, against chemical induced dry eye syndrome. Apoptosis, 22, 1441–1453 (2017).