GM sulfate (G1272) was purchased from Sigma-Aldrich; Merck KGaA (Darmstadt, Germany); C57BL/6J mice were acquired from the Model Animal Research Center of Nanjing University (Nanjing, China); 3-methyladenine (3-MA) was purchased from Sigma-Aldrich; Merck Millipore (M9281) and Dulbecco's modified Eagle's medium (DMEM) was obtained from Gibco; Thermo Fisher Scientific, Waltham, MA, USA. The following primary antibodies were used: Rabbit anti-Atg12 (2011S) and rabbit anti-Bcl-2 (50E3) from Cell Signaling Technology, Inc. (Danvers, MA, USA); rabbit anti-LC3 (L7543; Sigma-Aldrich; Merck Millipore) and mouse anti-Bcl-2 (SC-7382; Santa Cruz Biotechnology, Inc., Dallas, TX, USA). The following secondary antibodies were used: Cyanine (Cy3)-conjugated donkey anti-rabbit (A31572; Thermo Fisher Scientific, Inc.), horseradish peroxidase (HRP)-conjugated goat anti-rabbit (111–035-003; Jackson ImmunoResearch Laboratories, Inc., Westgrove, PA, USA) and IgG HRP-conjugated goat anti-mouse (SC-2005; Santa Cruz Biotechnology, Inc.). Fluorescein isothyanate (FITC)-phalloidin (P5282) and DAPI (D9542) were purchased from Sigma-Aldrich; Merck Millipore, and a Pierce Classic Immunoprecipitation kit was obtained from Thermo Fisher Scientific, Inc. (26146).

A total of 160 C57BL/6J mice (age, 4 days; male:female, 1:1; weight, 3–5 g) were purchased from the Model Animal Research Center of Nanjing University (Nanjing, China). Mice were housed under a 12-h light-dark cycle at 22±1°C and with a humidity of 56±5%, with 5 mice per cage. Mice received an ad libitum 0.3% sodium diet. The care and use of animals were in strict accordance with the ‘Guiding Directive for Humane treatment of Laboratory Animals’ issued by the Chinese National Ministry of Science and Technology. The Institutional Animal Care and Use Committee of Nanjing Medical University and Nanjing Drum Tower Hospital provided ethical approval for all procedures. All efforts were made to minimize the number of animals used and to prevent their suffering. Dissection and culturing was performed as previously described (22). Following cleaning with 75% ethanol, pups were decapitated and the cochleae were carefully dissected. Soft tissues and the cartilage were removed, and the cochlear epithelium was isolated. The organotypic cultures were maintained in serum-free media consisting of DMEM supplemented with glucose and 30 U/ml penicillin G (Sigma-Aldrich; Merck KGaA) at 37°C in a humidified atmosphere containing 5% CO₂. After 24 h, the culture medium was removed and replaced with fresh media containing 100 µm/l GM or 10 mmol/l 3-MA and incubated at 37°C for 24 h. Cochlear explants were randomly divided into three groups: Control (culture medium), GM-treated (culture medium + GM) and GM- and 3-MA-treated (culture medium + GM + 3-MA).

A total of 48 h after dissection, the organ of Corti were immediately processed for immunofluorescent staining. Cochlear explants were fixed with 4% paraformaldehyde for 30 min. Following fixation, they were rinsed with 0.1 mM phosphate-buffered saline (PBS) three times and blocked with 5% normal horse serum (Sigma-Aldrich; Merck KGaA) in PBS (pH 7.4) with 0.1% Triton X-100 for 1 h at room temperature. The tissues were incubated with an anti-LC3 primary antibody overnight at 4°C. The next day the tissues were washed three times with 0.1% Tween in PBS, followed by incubation with the secondary Cy3-conjugated donkey anti-rabbit antibody for 2 h at room temperature. Following this, FITC-phalloidin and DAPI were added to label hair cell stereocilia and the nucleus, respectively. Following immunostaining, samples were mounted in Southern Biotech Fluoromount-G™ Slide Mounting medium (Thermo Fisher Scientific, Inc.). Cochleae were dissected into apical, middle and basal turns, and were imaged using a Nikon confocal fluorescence microscope (TE2000-U; Nikon Corporation, Tokyo, Japan).

For HC quantification in the GM-treated and GM + 3-MA-treated samples, the entire cochlea was imaged using a quantified analytic system for cochlea hair cells as previously described (23,24). Missing inner hair cells (IHCs) and outer hair cells (OHCs) were counted over 0.24-mm intervals under a light microscope (magnification, ×400). Cochleograms were generated demonstrating the percentage hair cell loss as a function of percentage distance from the apex. Negative controls were included to determine the amount of background staining. The Cochlea Anatomy Labshell Program (supplied as a gift from Professor Dalian Ding, Buffalo University, Buffalo, NY, USA) was used to compute the percentage of missing hair cells in 10% intervals along the length of the cochlea, starting from the apex. For each cochlea, the mean density of HCs was determined (apical, 10–20% distance from the cochlear apex; middle, 40–50% distance from the cochlear apex; basal, 70–80% distance from the cochlear apex).

Cochlea were collected after 48 h and fixed in 2.5% glutaraldehyde in 0.1 M PBS at 4°C overnight. The samples were subsequently post-fixed in 1% osmium tetroxide and further processed by standard procedures, including dehydration, infiltration and polymerization in araldite. Ultrathin sections were post-stained and examined using a H-7650 transmission electron microscope (Hitachi, Ltd., Tokyo, Japan) (25). For SEM, the fixation procedure was performed as described above, following which samples were dehydrated in ethanol, critical point dried with a vacuum and coated with gold-palladium. The samples were observed by SEM for morphological description and for potential acoustic trauma assessment. Samples were imaged under a Hitachi S-3000 N Scanning Electron Microscope (Hitachi, Ltd.).

Western blot analysis was performed to assess the protein expression levels of LC3 and Bcl-2 from cochlear explants following exposure to the various experimental conditions as described above. The cochleas were lysed with ice-cold radioimmunoprecipitation assay buffer (PP109; Protein Biotechnology, Beijing, China), containing Phosphatase Inhibitor Cocktail (Roche, Basel, Switzerland), to obtain the tissue proteins. Protein concentrations were measured using a bicinchoninic acid Protein Quantification kit (PP202, Protein Biotechnology) according to the manufacturer's protocol, using GAPDH (KC-5G4, 1:2,000; KangChen Biotech Inc., Shanghai, China) as the reference protein. Equivalent amounts of tissue lysate protein (20 µg) were separated by gel electrophoresis on a 4–20% gradient SDS-PAGE gel and transferred onto polyvinylidene fluoride membranes. Blots were blocked with 5% non-fat milk and then incubated with anti-LC3 (1:1,000) and anti-Bcl-2 (1:1,000) rabbit polyclonal antibodies at 4°C overnight. Membranes were then washed with 0.1% Tween20 in PBS and probed with an HRP-conjugated goat anti-rabbit (1:5,000) secondary antibody for 2 h at room temperature. The bound antibody complexes were detected using an ECL Plus kit (Pierce; Thermo Fisher Scientific, Inc.) and a Storm 840 PhosphorImager system (Molecular Devices, LLC, Sunnyvale, CA, USA).

For endogenous co-immunoprecipitation, a rabbit monoclonal anti-Atg12 antibody was used. Precleared tissue lysates were incubated with anti-Atg12 at 4°C overnight, followed by incubation with protein G agarose beads for 1 h at room temperature. Beads were washed with lysis buffer and boiled in sample buffer. Western blot detection of Bcl-2 was performed with mouse anti-Bcl-2 (1:1,000) at 4°C overnight. Membranes were then washed with 0.1% Tween-20 in PBS and probed with an HRP-conjugated goat anti-mouse (1:5,000) secondary antibody for 2 h at room temperature. The bound antibody complexes were detected using an ECL Plus kit using a Storm 840 PhosphorImager system.

The data were presented as the mean ± standard deviation. All experiments were repeated at least three times. Statistical analyses were conducted using Microsoft Excel 2007 (Microsoft Corporation, Redmund, WA, USA) and GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA, USA). P<0.05 was considered to indicate a statistically significant difference.

GM-induced autophagy in the cochlea was evaluated by TEM, and the morphology of hair cells in GM-treated cochlea was investigated. As presented by representative micrographs, compared with the control (Fig. 1A), GM-treated cells (Fig. 1B) had autophagosomes with characteristic double or multiple membranes after 24 h. In addition, numerous autophagic vacuoles or vesicles were observed.

In order to assess cochleotoxicity and the protective underlying mechanism of 3-MA, cells were double-stained with FITC-phalloidin (green) to label hair cells, and DAPI (blue) as a nuclear stain (Fig. 2A). Previously, cochlear cultures from mice were incubated in medium containing 10 mmol/l 3-MA and 0.1 mM GM for 24 h. Loss of hair cells indirectly reflects cochleotoxicity; percentage hair cell loss was increased in the GM-treated group compared with the control group. However, the addition of 3-MA attenuated this effect, particularly in the middle region of the cochlea (Fig. 2B). Compared with untreated controls, GM severely distorted the anatomy of the organ of Corti in the GM-treated group, where the hair cells were deformed and disordered, and gaps were clearly observed. Following 3-MA treatment, the number and alignment of cells in the organ of Corti was improved compared with the GM group. FITC-phalloidin-labeled hair cells were counted in each group after 24 h. No significant differences were observed in the number of hair cells in the apex between groups. Conversely, in the GM-treated group, OHC and IHC counts were significantly decreased in the middle region of the cochlea. However, in GM + 3-MA-treated group, 3-MA significantly protected against OHC and IHC loss in the middle region of the cochlea. In summary, these results suggested that 3-MA treatment significantly protects against GM-induced hair cell loss in the middle region of the cochlea.

SEM was performed to examine the surface structure of the cochlea. Examination of the organ of Corti of the control group (Fig. 3A) indicated three rows of OHCs and one row of IHCs. Conversely, cochleae from the GM-induced ototoxic group (Fig. 3B) exhibited loss of the well-known architecture of hair cells of the organ of Corti. The stereocilia of the OHCs and IHCs were disrupted, disarrayed, fused, had focal loss or were completely absent. Treatment with 3-MA preserved the majority of hair cells; however, the stereocilia became more dispersed (Fig. 3C).

To examine whether GM injury affects protein expression levels of LC3, immunostaining and western blotting were performed on each group. LC3 was detected in the control and GM-treated groups; however, expression levels were markedly upregulated in the GM-treated group (Fig. 4A). In western blots, LC3 protein presents with double bands; one is unconjugated and the other is conjugated with phosphotidylethanolamine to form the LC3-II complex. This is recruited to the autophagosome membrane; increased expression levels reflects the level of autophagy in cells. LC3 protein levels were overexpressed in the GM-treated group compared with the control group (Fig. 4B).

To assess the interaction between Atg12 and Bcl-2 in cochleae, western blotting was performed to observe protein expression levels of Bcl-2 in the GM-, GM + 3-MA-treated and control groups. Bcl-2 was equally expressed in all three groups (Fig. 5A). Co-immunoprecipitation was performed using a rabbit anti-Atg12 antibody to observe the protein complexes involved in ototoxicity and further analyze the interaction between Atg12 and Bcl-2 (Fig. 5B). Bcl-2 and Atg12 were co-expressed in the GM-treated and control groups, indicating that they interact in the cochlea. However, when autophagy was inhibited by 3-MA, Bcl-2 protein expression levels were reduced, demonstrating that 3-MA interrupts the Atg12 and Bcl-2 interaction in GM-induced cochleotoxicity. As a result, the apoptotic function of Atg12 is inhibited, reflecting an increase in the number of hair cells. In summary, these results suggested that following GM-induced cochleotoxicity, Atg12 and Bcl-2 interact to promote cochlear apoptosis, and 3-MA rescues this effect by inhibiting the Atg12-Bcl-2 complex.