Six-week-old BALB/c mice (Orient Charles River Technology, Seoul, Korea) with normal Preyer's reflex and a normal hearing threshold, were used. All animal experiments were performed with food and water available ad libitum and with the approval of the Animal Care Committee of University of Ulsan College of Medicine. The care and use of the animals reported in the present study were in accordance with the guidelines of the Laboratory Animal Unit of the Asan Institute for Life Sciences.

White noise (300-10,000 Hz) was generated by a personal computer using Cool Edit 1.52 software and amplifier (R-399, INTER M, Seoul, Korea) and delivered through a speaker (290-8L, Altec Lansing, OK, USA) in a sound-proof booth. Planned noise in a booth was calibrated by a sound level meter (B&K, Narum, Denmark) at the center and the edge of a booth.

When we collected the cochlea from mice and measured hearing threshold, mice were anesthetized by intramuscular injection of ketamine hydrochloride (30 mg/kg) and xylazine (2 mg/kg).

The hearing level of each mouse was determined by measuring ABR thresholds using auditory evoked potential workstation (Tucker-Davis Technologies, Alachua, FL, USA). Mice were anesthetized by intramuscular injection of ketamine hydrochloride (30 mg/kg) and xylazine (2 mg/kg), and both ears of each mouse were stimulated with ear probes sealed into the ear canal. The frequency-specific ABR in response to tone burst sound was recorded at 4, 8, 16, and 32 kHz. Sound stimuli were applied and the threshold was defined as the lowest intensity of stimulation that yielded a repeatable waveform based on an identifiable ABR wave V. Hearing threshold was measured the day after noise exposure.

Mice were continuously exposed to 122 dB peak equivalent sound pressure level white noise (300-10,000 Hz, continuous tone) for 3 hr per day for 3 consecutive days in the experimental group. Mice in the control group were kept in the noise booth for 3 hr per day for 3 consecutive days without noise.

Immediately after the measurement of hearing threshold, mice were deeply anesthetized by intramuscular injection of ketamine hydrochloride (50 mg/kg) and xylazine (2 mg/kg), and both cochleae were removed from each mouse. The stapes were removed and a small hole was made in each cochlear apex with a fine pick. Fixative (4% formalin and 1% glutaraldehyde in 0.1 M sodium phosphate buffer) was infused and the cochleae were immersed in fixative for 48 hr at 4℃. After washing with phosphate-buffered saline (PBS), the outer bony wall of each cochlea and the tectorial membrane were gently removed, and the organ of Corti was harvested with fine forceps beginning from the apex.

Fluorescein isothiocyanate (FITC)-phalloidin (1 µg/mL, Sigma, St. Louis, MO, USA) was used for F-actin staining. We mounted the specimen on slides after reaction with FITC-phalloidin for 1 hr at 4℃ in a darkroom and rinsing it with PBS.

For the nuclear staining, PI (4 µg/mL, Molecular Probes Inc.) was used. PI showed fluorescence when inserted in DNA in nucleus.

We observed a double-stained cell layer by confocal microscopy, which was able to examine fluorescently. The PI-labeled nuclei (red fluorescence) and FITC labeled F-actin (green fluorescence) were in one field of view simultaneously scanned in the cochlear section of interest and counted number of stained OHC.

When we observed cellular tissue, we made coverage come into sight in one field of view of a region and examined after dividing into 16 consecutive regions from the apex to the base of the cochlea. While the cells were fixed, every cell was stained by PI. Therefore, typical findings of PI staining corresponding to apoptosis were increased fluorescence intensity and prominent nucleus compared with surrounding cells. As apoptosis proceeded, chromatin and nucleus condensed. PI (+) was nuclear condensation featured by nuclear condensation with an increase in PI fluorescence staining. PI (-) was nuclear swelling characterized by enlargement of the nucleus, and a decrease in PI fluorescence intensity. PI (+) is evidence for the presence of apoptosis, and PI (-) denotes absence of apoptosis (11, 12). FITC (+) was intact F-actin; stereocilia was stained well with FITC and had no morphologic change. FITC (-) was breakdown of F-actin; stereocilia was not stained with FITC or had abnormal morphology. We conducted the quantitative analysis of F-actin-staining and PI-staining cells at each region. At this time, we counted the staining condition of OHC according to FITC (+) PI (-), FITC (+) PI (+), FITC (-) PI (+), FITC (-) PI (-) (Fig. 1). We defined a FITC (-) PI (-) cell in the course of staining as the missing cell.

Statistical analysis of the data was performed using SPSS 12.0 K (SPSS Inc., Chicago, IL, USA). We used the paired t test for analysis of changes in hearing threshold following noise exposure and the Student t test for comparison of hearing threshold between the experimental and control groups.

When we counted the number of cells in the fluorescent-stained cell layer under the magnification conditions, one field of view contained an average length of the 0.316-mm basement membrane, and we examined 16 consecutive lengths, for an average total length of 5.056 mm, or about 82% of the 6.16-mm long organ of Corti (13).

OHC cells of the control group showed no apoptosis with PI staining and showed normal FITC staining pattern (FITC (+) PI (-)) over the whole length of the cochlea. Normal OHC occupied 97.3±0.5% of the whole cochlea.