Furosemide prevents noise-induced hearing loss and enhances the preventive effect of N-acetylcysteine

Abstract

Disruption of reactive oxygen species (ROS) homeostasis is a key mechanism underlying noise-induced sensory hair cell damage. Antioxidant treatments such as N-acetylcysteine (NAC) have been shown to attenuate noise-induced hearing loss (NIHL), supporting the role of ROS accumulation. However, no FDA-approved pharmaceutical therapy currently exists for the prevention or treatment of NIHL, likely due to the complexity of the damaging mechanisms and the presence of the blood-labyrinth barrier (BLB), which limits drug permeability and prevents therapeutic compounds from reaching effective concentrations via systemic administration.

Furosemide (FRS) has demonstrated potential to reduce NIHL and facilitate drug delivery into inner ear by transiently opening the BLB. In this study, we investigated the mechanisms by which FRS pretreatment prevents NIHL. A single dose of 200 mg/kg FRS administered immediately before noise exposure significantly reduced NIHL in FVB/NJ mice. One hour after FRS treatment, the endocochlear potential (EP) was temporarily reduced without altering cochlear sensitivity (ABR thresholds), outer hair cell function (DPOAE amplitudes), or synaptic transmission integrity between hair cells and auditory nerve fibers (suprathreshold ABR wave I amplitudes).

Furthermore, this dose of FRS selectively increased stria vascularis permeability to small molecules but not to large protein-bound tracers. Combined treatment with FRS and NAC enhanced NAC’s antioxidant effect and additively prevented noise-induced outer hair cell (OHC) loss and NIHL, with OHC loss almost entirely prevented. These findings provide important insight into future strategies for the prevention and treatment of NIHL.

1. Introduction

Noise-induced hearing loss (NIHL) is the most prevalent occupational disease in the United States and one of the leading causes of acquired hearing impairment worldwide, resulting from military service, industrial work, and recreational noise exposure (Cunningham & Tucci 2017, Masterson et al. 2013, Tak et al. 2009). Although the molecular pathophysiology following noise exposure is highly complex, the primary pathology involves the loss of sensory hair cells, with outer hair cells (OHCs) being more vulnerable to noise damage than inner hair cells (IHCs) (Wang et al. 2018, Cody & Russell 1985, Waqas et al. 2018). Mammalian cochlear sensory hair cells play a critical role in hearing function by converting mechanical sound vibrations into electrical signals, However, their loss leads to permanent hearing impairment, as these cells are unable to regenerate (Roberson & Rubel 1994, Choi et al. 2024).

One of the underlying mechanisms of noise-induced sensory hair cell damage is disruption of reactive oxygen species (ROS) homeostasis, which contributes to NIHL (Sha & Schacht 2017). This theory is well-supported, as several antioxidant treatments have been shown to attenuate NIHL (Yamashita et al. 2004, Wu et al. 2020). For example, N-acetylcysteine (NAC), a precursor of glutathione and as one of the body’s most important endogenous antioxidants, boosts glutathione levels and neutralizes ROS. Treatment with NAC has demonstrated to reduce noise-induced elevation in auditory thresholds and preserved hair cell integrity in chinchilla and rats (Choi et al. 2014, Rhee & Chang 2021). A systematic review and meta-analysis of randomized controlled clinical trials found that NAC may reduce auditory threshold elevation in the 0 to 6 kHz frequency range (Chang et al. 2022). However, no clinically approved pharmaceutical therapy current exists for the prevention or treatment of NIHL. This may be due to the complexity of the damaging mechanisms and the presence of the blood-labyrinth barrier (BLB) in the inner ear, which limits the permeability of many compounds and prevents them from reaching therapeutic concentration via systemic administration (Juhn et al. 1981, Rivera et al. 2012).

The BLB is a specialized barrier in the inner ear, functionally similar to the blood barrier, allowing essential nutrients and ions to pass while blocking harmful substances, toxins, and pathogens, thereby maintaining the ionic and molecular homeostasis necessary for normal auditory function (Cohen-Salmon et al. 2007). While the BLB poses a challenge for delivering therapeutic drug to the inner ear, strategies such as transient permeability modulation are being explored (Yi et al. 2025). One of such approach involves the use of furosemide (FRS), a well-known loop diuretic that inhibits the sodium-Potassium-Chloride Co-transporter (NKCC) (Zhao et al. 2025b, Orlov et al. 2015). Furosemide exerts its diuretic effect by inhibiting NKCC2 on the luminal side of the thick ascending limb of the loop of Henle in the kidney. Importantly, NKCC1 is expressed in the basolateral membrane of the stria vascularis (SV) in the cochlea and is sensitive to FRS (Shindo et al. 1992).

FRS has been reported to prevent NIHL, potentially through its ability to transiently reduce the endocochlear potential (EP) (Adelman et al. 2010, Kanno et al. 1993). EP is essential for normal cochlear function and can be temporarily reduced by loop diuretic drugs. On the other hand, FRS targets on BLB and modulates intrastrial fluid, facilitating the entry of other compounds into the endolymphatic compartment (Videhult Pierre et al. 2020). For example, co-administration of ethacrynic acid with gentamicin has been shown to increase gentamicin concentration in perilymph, resulting in more severe hair cell loss (Ding et al. 2016). This highlights the potential of loop diuretics temporarily modulate the blood-cochlea barrier, could be leveraged to enhance the delivery of otoprotective compounds to the inner ear.

In this study, we first characterized noise-induced threshold elevation in FVB/NJ mice. We found that male mice are sensitive to female mice, in line with previous observation (Milon et al. 2018). Next, we investigated the mechanisms by which FRS pretreatment attenuates NIHL, as well as its potential additive effect when combined with an antioxidant. Our findings indicate that an appropriate dose of FRS temporarily reduces the EP without elevating auditory thresholds, contributing to its preventive effect against NIHL. Furthermore, combined FRS with NAC treatment provides enhanced prevention of NIHL, likely due to FRS-induced EP reduction and increased NAC permeability into the SV, thereby amplifying NAC’s antioxidant efficacy.

2. Materials and methods

2.1. Animals

FVB/NJ breeder (stock #001800) at age of 4 weeks were purchased from the Jackson Laboratory. Both breeders and offspring were housed in a specific pathogen-free animal facility at the Medical University of South Carolina (MUSC). Animals were maintained under a standard 12:12 h light–dark cycle at an ambient temperature of approximately 22°C, with background noise levels around 50 dB SPL. Mice had ad libitum access to a regular mouse diet (Irradiated Lab Diet #5V75) and water. FVB/NJ offspring were used in this study. Noise exposure was conducted 8 weeks of age. Baseline ABRs for all mice were measured one week prior to noise exposure. Mice with baseline auditory thresholds at any frequency higher than 40 dB SPL were considered abnormal and excluded from the study. All experimental procedures were approved by the Institutional Animal Care & Use Committee at MUSC, and animal care was overseen by the Division of Laboratory Animal Resources.

Our results showed that male FVB/NJ mice are particularly sensitive to noise exposure. Specifically, exposure to 100 dB octave-band noise (OBN) for 2 h resulted in threshold elevations of approximately 80 dB SPL across six tested frequencies (from 8-32 kHz). This robust response provided a suitable model for elevating additive preventive strategies for NIHL. In contrast, females showed highly variable responses, ranging from no hearing loss to complete hearing loss. Therefore, only male mice were included in the NIHL prevention experiments.

2.2. Noise exposure

Both male and female FVB/NJ mice, aged 8 weeks, were exposed to octave-band noise (OBN) at 100 dB SPL with a frequency spectrum ranging from 8–16 kHz for 2 h. Unrestrained mice were placed individually in stainless steel wire cage (approximately 9 cm³) inside a sound chamber during morning sessions. Sound levels were measured using a sound level meter at multiple locations within the chamber to ensure uniformity of the sound field, and measurements were taken both before and after exposure to confirm stability.

Sound pressure level calibration was regularly performed with a Bruel and Kjaer condenser microphone, allowing precise calibration and monitoring of the sound exposure. The noise level varied by a maximum of 1–2 dB across measured sites within the exposure chamber.

The sound chamber was equipped with a loudspeaker (model 2450H; JBL) powered by Crown Audio amplifier (model XLS 202D), with audio signals delivered from a CD player (model CD-200; Tascam TEAC American). Sound files were created and equalized using Adobe Audition 3 (Adobe Systems, Inc.). The background sound intensity surrounding the chamber was measured at 65 dB SPL using a Quest Technologies sound level meter (model 1200).

Control mice were placed in identical cages within the same chamber for 2 h but were not exposed to noise; the loudspeaker remained off to maintain silence.

2.3. Protective drug treatment

Based on our recent publication (Wu et al. 2025), we tested different doses of furosemide (FRS; Baxter, #AIN00214) at 100 mg/kg and 200 mg/kg via a single intraperitoneal (IP) injection in FVB/NJ mice 8-week-old FVB/NJ mice. ABRs were measured at 1 h and 1 d post-FRS treatment to assess auditory function.

Given that treatment with 200 mg/kg FRS did not induce elevation of ABR thresholds and DPOAE amplitudes at 1 h post-treatment, we proceeded to test their efficacy in prevention of NIHL. A detailed schematic treatment of the treatment protocol was shown in Fig. 3A.

Fig. 3.

A single dose of FRS (200 mg/kg) administrated immediately prior to noise exposure significantly attenuates noise-induced hearing loss.

(A) Schematic illustrating the timeline of FRS (200 mg/kg) administration relative to noise exposure.

(B) FRS administrated immediately before noise significantly reduced noise-induced threshold elevations across 8–32 kHz at 14 days post-exposure. The preventive effect declined when FRS was administrated 2 h before noise exposure, and no prevention was observed when administrated immediately after exposure. Data are presented as mean ± SEM, with sample sizes (n) indicated in the figure. Statistical analysis: two-way ANOVA with Sidak’s multiple comparisons test.

(C) FRS administrated immediately before noise did not prevent noise-induced reduction in DPOAE amplitudes at 14 d post-exposure. Data are presented as mean ± SEM, with sample sizes (n) indicated in the figure. Statistical analysis: two-way ANOVA with Sidak’s multiple comparisons test.

(B’–C’) Detailed multiple-comparison data corresponding to panel B and C are shown in B’ and C’.

In accordance with our lab’s early publication and literature (Wu et al. 2020, Lu et al. 2014), a dose of 325 mg/kg N-Acetylcysteine (NAC; Sigma-Aldrich, #A7250) was used in this study. NAC was dissolved in 0.9% saline and adjusted the pH to 7.0 via 0.2M NaOH and administrated via IP injection. The timing of NAC administration relative to noise exposure and FRS treatment was illustrated in Fig. 4A.

Fig. 4.

Combined FRS (200 mg/kg) and NAC treatment additively prevents noise-induced outer hair cell loss and hearing loss.

(A) Schematic illustrating the timeline of NAC and FRS administration relative to noise exposure.

(B–B’) NAC alone significantly reduced ABR threshold elevation at 8, 20, and 24 kHz measured 14 d post-exposure. A single dose of FRS immediately before noise provided better prevention than NAC alone at 20 kHz. Combined treatment (NAC + FRS pre-noise) prevented threshold elevations across 8–32 kHz, demonstrating superior efficacy compared to either treatment alone. Data are presented as mean ± SEM, with sample sizes (n) indicated in the figure. Statistical analysis: two-way ANOVA with Tukey’s multiple comparisons test; detailed values for panel B are shown in B’.

(C–C’) Noise exposure suppressed DPOAE amplitudes across 8–36 kHz. Combined treatment significantly preserved amplitudes between 8–32 kHz, demonstrating greater efficacy than either treatment alone. Data are presented as mean ± SEM, with sample sizes (n) indicated in the figure. Statistical analysis: two-way ANOVA with Tukey’s multiple comparisons test; numerical values for panel C are shown in C’.

(D) Representative cochlear surface preparation images (basal turn, ~5-5.5 mm from the apex) labeled with myosin VIIa (red) and stained with phalloidin (green) from Ctrl, 100 dB, and NAC + FRS + 100 dB groups. Scale bar: 10 μm.

(E) Noise-induced OHC loss was observed from the cochlear base to 4.5 mm from the apex. Combined treatment fully prevented this loss, similar to control mice. Data are presented as mean ± SEM, with sample sizes (n) indicated in the figure. Statistical analysis: two-way ANOVA with Tukey’s multiple comparisons test. Asterisks (*) = Ctrl vs. 100 dB (***p < 0.001, ****p < 0.0001).

2.4. Measurements of auditory brainstem responses and distortion product otoacoustic emissions

Mice were anesthetized via intraperitoneal (IP) injection of a mixture of xylazine (10 mg/kg) and ketamine (100 mg/kg). Upon confirmation of the absence of pain response, animals were placed in a sound-proof chamber, and body temperature was maintained at 37°C using a heating pad.

Auditory brainstem responses (ABRs) were recorded from the left ear using tone bursts (3-ms duration) delivered via an ear bar inserted into the external auditory meatus. Subdermal needle electrodes were positioned at the vertex (active), below the ipsilateral pinna (reference), and below the contralateral pinna (ground). ABRs were measured at 8, 12, 16, 20, 24, and 32 using Tucker Davis Technology (TDT) System III hardware and SigGen/Biosig software (TDT, Alachua, FL, USA). Stimuli consisted of 15-ms tone bursts with 1-ms rise/fall time. Up to 1024 responses were averaged per stimulus level. Wave II, the most robust and reliable ABR components, was used to determine ABR thresholds by an expert blinded to experimental conditions.

Distortion product otoacoustic emissions (DPOAEs) were measured using a TDT RZ6 system and SigGen software. An ER-10B+ microphone connected to two transducers was sealed into the ear canal. Primary tones were presented at L1= 65 dB SPL and L2 = 55 dB SPL with an f2/f1 ratio of 1.2. Responses were recorded across a frequency range of 4–44 kHz. Data was analyzed for each frequency and averaged to assess OHC function.

2.5. Measurement of endocochlear potential (EP)

After anesthesia, the post-auricular region of the left ear was shaved and disinfected with 10% povidone-iodine and 70% alcohol. Under a surgical operating microscope, a retroauricular incision (about 1.0 cm) was made to expose the temporal bone (Hirose & Liberman 2003). After carefully removing overlying tissue and musculature, the tympanic bulla was opened using a fine drill mounted on a Buffalo No. 16 Engine to provide a clear view of the cochlea.

A small hole was created in the lateral wall of the basal turn of the cochlea by thinning the bone with the same fine drill, followed by penetration of the otic capsule using a 31 G needle. A glass capillary microelectrode (5–8 MΩ), filled with 150 mM KCl, was mounted on a Märzhäuser MM33 Micromanipulator. A silver/silver chloride ball electrode inserted into the neck muscles served as the ground. After a well-positioned opening was made in the otic capsule, mice received an IP injection of either 200mg/kg FRS or an equivalent volume of saline. The microelectrode was then advanced into the scala media until a positive potential was observed. Signals from the recording electrode were amplified using an AM Systems Model 1600 intracellular amplifier.

2.6. Acoustic startle response

The acoustic startle response (ASR) was measured using a startle reflex system (Maze Engineers), as previously described (Su et al. 2025). Mice were acclimated to the testing room overnight prior to the experiment. Each mouse was placed in the startle chamber equipped with a motion sensor to detect startle response.

The test session began with a 2-mim acclimated period in the chamber, during which a continuous background white noise of 80 dB SPL was maintained. Following acclimation, mice were exposed to 25 ms white noise bursts at 100 dB SPL, presented with an inter-trial interval of 20 ± 7 s. Each session included 20 stimulus presentations.

The startle responses were recorded as the peak amplitude within a 100-ms window following the onset of the acoustic stimulus. Data were analyzed using the MazeOriginWpf software to calculate the average startle amplitude at the given sound intensity.

2.7. Immunolabeling using cochlear whole-mount surface preparations for hair cell and synaptic ribbon counts, and ROS signal assessment

Immunolabelling for myosin VIIa (to assess sensory hair cell loss), CtBP2 (to assess IHC presynaptic ribbon), and 3-nitrotyrosin (3-NT, to assess ROS signaling) were performed as previously described (Hill et al. 2016, Yuan et al. 2015, Lai et al. 2023). Briefly, temporal bones were removed immediately after euthanasia and gently perfused through the round and oval windows with 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS, pH 7.4), followed by overnight fixation at 4°C.

After decalcification in 4% neutral sodium EDTA for 48 to 72 h at 4°C, cochleae were micro-dissected into apical, middle, and basal turns and adhered on a 10-mm round coverslip using Cell-Tak (Corning, #354240). Samples were permeabilized with 1% Triton X-100 (Sigma-Aldrich, #T9284) and blocked with 10% goat serum (GS) for 30 min at room temperature (RT). They were then incubated with primary antibodies, followed by appropriate secondary antibodies.

Hair cell counts: Samples were incubated with polyclonal rabbit anti-myosin VIIa (1:200; Proteus Bioscience, #25–6790) in darkness at 4 °C for 24 h, followed by Alexa-Fluor 594-conjugated secondary antibody (1:200) overnight at 4 °C.

IHC presynaptic ribbon counts: Samples were incubated with monoclonal mouse anti-CtBP2 IgG1 (1:200; BD Biosciences, #612044) in darkness at 37 °C overnight, followed by the Alexa Fluor-594 goat anti-mouse IgG1 (1:1,000) for 1 h at 37 °C in darkness.

For ROS signaling: Samples were incubated with monoclonal mouse anti-3-nitrotyrosine (1:200; Sigma, #N5538) in darkness at 4 °C for 24 h, followed by Alexa-Fluor-594-conjugated secondary antibody (1:200) overnight at 4 °C.

After washing, samples were incubated overnight in the dark 4 °C with Alexa-Flour-488-phalloidin (1:200; Invitrogen, #A12379). Between each step, samples were washed three times with PBS on a shaker. Finally, 8 μL of mounting medium (Electron Microscopy Sciences, #17985-10) was applied to the coverslips, which were then sealed with a second-round coverslip on microscope slides, then the two slips were sandwiched together and sealed with nail polish. Secondary antibody controls were processed in parallel without incubation with primary antibodies.

For hair cell counts, surface preparations were imaged using a 20× objective on a Zeiss confocal microscope. The cochlear epithelium was measured in millimeters. As IHC loss was not observed in FVB/NJ mice after noise exposure, only OHCs were counted along the entire length of the mouse cochlear spiral from apex to base. Percent OHC loss was calculated in 0.5-mm segments and plotted as a function of cochlear length to generate cytocochleogram.

For 3-NT signal analysis, confocal images were captured at 63× magnification from the apical, middle, and basal turns under identical setting for semi-quantitative comparison.

For CtBP2 presynaptic ribbon counts, Z-stack images were captured at 63× magnification from 0.12-mm segments (containing approximately 16 IHCs) in the apical, middle, and basal turns under identical settings to enable semi-quantitative analysis of the number of presynaptic ribbons per IHC (Hill et al. 2016).

2.8. Semi-quantification of the immunofluorescence signals and presynaptic ribbon counts from whole-mount surface preparations

To ensure consistency across samples, immunolabeling signals were semi-quantified from confocal images of cochlear whole-mount surface preparations. Immunolabeling for 3-NT and presynaptic ribbon counts were performed using cochleae from control and noise-exposed groups, which were fixed, stained, and processed in parallel using identical solutions and imaging conditions. Image analysis was using ImageJ software.

Semi-quantification of 3-NT signals:

Confocal images were acquired from the apical (~1 mm from the apex; ~8 kHz), middle (~2.4 mm; ~16 kHz), and lower basal (~4.5 mm, ~45 kHz) cochlear turns using a 63× objective under identical laser gain and acquisition setting. Surface preparations were counterstained with Alexa Fluor 488 phalloidin (green) to label hair cell structure and identify the comparable regions across samples. Regions of interest (ROIs) corresponding to individual OHCs were outlined using the circle tool based on phalloidin staining. Grayscale values were determined within each ROI to quantify fluorescence intensity. Immunolabeling intensity was assessed in 0.12-mm segments of the cochlear epithelium, each containing approximately 60 OHCs. Background intensity was subtracted and the average grayscale intensity per cell was calculated. For each experimental repetition, relative fluorescence intensity was normalizing the control group.

Semi-quantitative analysis of IHC presynaptic ribbons:

Z-stack images were acquired from the apical, middle, and basal (~3.9 mm, ~32 kHz) cochlear turns using a 63× objective under identical laser gain and acquisition setting. Images processing included background subtraction and a single despeckling step. A threshold was applied to isolate the ribbon signals, and image was converted to a binary format. Ribbon particles were quantified using the 3D Object Counter plugin in ImageJ. The total number of synaptic ribbons was divided by the number of IHCs within each image to calculate the average number of ribbons per IHC.

2.9. Assessment of stria vascular permeability, cardiovascular perfusion, and stria vascularis preparation

To evaluate SV permeability following FRS treatment, FVB/NJ mice received IP injection of FRS at a dose of 200 mg/kg. Permeability was assessed using a Alexa Fluor 555 cadaverine (ThermoFisher Scientific # A30677) or Evans Blue (EB) dye (Sigma-Aldrich # E2129), as previously described (Zhang et al. 2012).

Alexa Fluor 555 cadaverine assay:

Control mice (no FU treatment) and FU treated mice (1 h or 4 h post-treatment) received an intravenous injection of Alexa Fluor 555 cadaverine (5 mg/kg; MW: ~950 Da) via the tail vein. Five min after cadaverine injection, mice were anesthetized with ketamine (100 mg/kg) and xylazine (10 mg/kg) prior to cardiovascular perfusion. Each mouse was placed in a supine position, and a midline incision was made to expose the thoracic cavity. The right rib cage was opened to access the heart. A needle was inserted into the left ventricle, and a small incision was made in the right atrium to allow outflow. Perfusion was initiated with 5 mL of phosphate-buffered saline (PBS) for approximately 3 min, followed by 20 mL of 4% PFA at a flow rate of 2 mL/min for tissue fixation.

Following perfusion, cochleae were carefully removed from the temporal bone and gently perfused with 4% PFA through wound and oval windows under a microscope, then fixed overnight at 4°C. Without decalcification, the cochlear bone was removed under a microscope using forceps to expose the pigmented stria vascularis. The middle turn of the stria vascularis was carefully separated from the spiral ligament and flattened onto a coverslip pre-coated with Cell-Tak.

Samples were counterstained overnight at 4°C with Alexa Fluor 488 Isolectin GS-IB4 (1:2000; Fisher Scientific, #121411). After three washes, DAPI (1:1000) was applied at RT for 30 min. Finally, samples were mounted on round glass coverslips. Fluorescent images of the lateral wall were captured using a Zeiss 510 fluorescent microscope at 63× magnification.

EB dye assay:

Evans blue (2%, MW: 960.8 Da) was intravenously injected via the tail vein without anesthesia. One hour later, mice received an IP injection of FRS (200 mg/kg). One hour after FRS administration, mice were euthanized, and cochleae were harvested with or without cardiovascular perfusion. Cochleae were fixed by 4% PFA for overnight at 4°C. The following day, cochlear SV preparations were carefully dissected and immediately visualized under a laser-scanning confocal microscope.

2.10. Semi-quantification of the immunofluorescence signals from SV surface preparations

As described in the section “Semi-quantification of the immunofluorescence signals from whole-mount surface preparations”, original confocal images were acquired using a 63× objective. Images analysis was performed using semi-quantitative methods in ImageJ. Each image was converted to an 8-bit grayscale, and a consistent threshold was applied across all samples to minimize background noise and the grayscale intensity within the entire images was measured to quantify signal levels.

2.11. Statistical analysis

Data were analyzed using SYSTAT 8.0 and GraphPad 5.0 software for Windows. Biological sample sizes were determined based on the variability of measurements, the magnitude of the differences between groups, and experience from our previous studies, with stringent assessments of differences. We used the t-test to determine if there is a significant difference between the means of two groups. Differences for single-pair comparisons were analyzed using two-tailed unpaired Student’s t-tests. Data for relative ratios of single-pair comparisons were analyzed with one-sample t-tests. To determine whether there are any statistically significant differences between the means of three or more independent groups, we used one-way analysis of variance (ANOVA) with Turkey’s multiple comparisons. The percentages of OHC loss along the cochlear spiral, ABR thresholds, and DPOAE amplitudes along the frequencies were used two-way (ANOVA) with Turkey’s multiple comparisons. A p-value < 0.05 was considered statistically significant. Data are presented as individual mice and means ± SD or SEM, depending on sample size and variability within groups.

3. Results

3.1. Noise exposure induces auditory threshold elevation in FVB/NJ mice, with a consistently pronounced effect in males.

Exposure of 8-week-old FVB/NJ mice to 8–16 kHz OBN at 100 dB SPL for 2 h resulted in a significant elevation in auditory thresholds across all tested frequencies (8–32 kHz) in male mice, as measured 14 d post-exposure (Fig. 1A–B). Male mice exhibited thresholds of approximately 75–80 dB SPL across all frequencies. Baseline thresholds were around 30 dB SPL from 8–24 kHz and 40 dB at 32 kHz.

Fig.1.

Exposure to 100 dB SPL octave band noise induces threshold elevation in FVB/NJ male mice.

(A) Representative ABR waveforms at 16 kHz recorded from control (no exposure) and male mice 14 d post-noise exposure. ABR threshold was determined by the last identifiable wave II, highlighted in the red.

(B–C) ABR thresholds were significantly elevated across 8–32 kHz (B), and DPOAE amplitudes were diminished across 8–36 kHz (C) at 14 d post-exposure. Frequency corresponding to cochlear location are indicated below the figures. Data are presented as mean ± SEM, with sample size (n) noted in the figures. Statistical analysis: two-way ANOVA with Tukey’s multiple comparisons test. ****p < 0.0001.

(D) Representative images of the basal turn (~ 5 mm from the apex) from control and exposed mice 14 d post-exposure, labeled with myosin VIIa (red). Lost OHCs are marked with white asterisks for clarity. Scale bar: 10 μm.

(E) Noise-induced OHC loss followed a base-to apex gradient, with significant loss between 4.5 mm to 5.5 mm from the apex. Corresponding frequencies are indicated below. Data are presented as mean ± SEM, with sample sizes (n) provided in the figure. Statistical analysis: two-way ANOVA with Tukey’s multiple comparisons test. ***p < 0.001, ****p < 0.0001.

DPOAE measurement with amplitude-based presentations reflect the strength of the otoacoustic emission at each frequency and are commonly used to evaluate cochlear OHC function. Fourteen days after noise exposure, male mice showed significantly suppressed DPOAE amplitudes across 8–40 kHz range, with values reaching as low as −20 dB SPL (Fig. 1C). Furthermore, this noise exposure induced significant OHC loss following a based-to-apex gradient, with marked loss in the basal turn between 4.5 and 5.5 mm from the apex (Fig. 1D–E). However, no changes were observed in inner hair cell ribbon synapses 14 d post-exposure in male mice (Supplementary Fig. 1).

Female FVB/NJ mice showed highly variable responses to 100 dB SPL OBN exposure, ranging from no hearing loss to complete hearing loss (Supplementary Fig. 2).

These findings highlight the occurrence of noise-induced hearing loss, with male mice displaying a consistent and broader dynamic range for evaluating the efficacy of preventive treatments. Therefore, only male mice were used in the subsequent experiments.

3.2. A single dose of furosemide (200 mg/kg) transiently reduces endocochlear potential without affecting hearing sensitivity or outer hair cell function.

Based on our previous study in CBA/J mice (Wu et al. 2025), we measured ABRs and DPOAEs in male FVB/NJ mice following IP injection of FRS at 100 mg/kg and 200 mg/kg. Measurements were taken at two points (1 h and 1 d post-treatment) to determine the optimal dose for subsequent experiments. No threshold elevations were observed across all frequencies for either dose at either time point (Fig. 2A–D).

Fig. 2.

A single dose of FRS (200 mg/kg) significantly and transiently reduces EP without affecting hearing sensitivity.

(A–D) FRS at 100 mg/kg and 200 mg/kg did not alter ABR thresholds across all frequencies (A–B) or DPOAE amplitudes (C–D) at 1 h and 1 d post-treatment. Data are presented as mean ± SD, with sample sizes (n) indicated in the figure. Statistical analysis: two-way ANOVA showed no significant difference.

(E–G) FRS (200 mg/kg) did not affect suprathreshold (80 dB SPL) ABR wave I amplitudes at 12 kHz (E), 16 kHz (F), and 20 kHz (G) at 1 h and 1 d post-treatment. Data are presented as mean ± SD. Statistical analysis: one-way ANOVA showed no significant difference.

(H) Schematic of a cochlear cross-section illustrating the placement of the EP recording electrode in the basal turn of scala media.

(I) Experimental image showing EP electrode insertion into the basal turn of cochlea. Black dotted lines mark cochlear turn boundaries; the white dotted line indicates the electrode path.

(J) FRS (200 mg/kg, IP) significantly reduced EP 1 h after administration. EP was recorded directly from the basal turn of left cochleae in 8-week-old male FVB/NJ mice. Each data point represents one animal; data are plotted as mean ± SD, with sample sizes (n) indicated in the figure. Statistical analysis: Student’s t-test. ****p < 0.0001.

(K) Acoustic startle responses were unaffected 1 h after FRS (200 mg/kg) compared with saline-treated controls. Data are presented as individual points with mean ± SD, with sample sizes (n) indicated in the figure. Statistical analysis: Student’s t-test. ns = not significant.

Analysis of suprathreshold ABR wave I amplitude at 12, 16, and 20 kHz following 200 mg/kg FRS at 1 h and 1 d post-treatments revealed no differences compared to their baseline (Fig. 2E–G). The reduction of EP after FRS administration has been well documented in larger rodents such as chinchillas (Rybak et al. 1991), gerbils (Mills et al. 1993), and guinea pigs (Mori et al. 1990). The positive EP within the scala media (80-100 mV) is generated by the stria vascularis (Hirose & Liberman 2003). To assess this effect in mice, we measured EP by inserting a glass microelectrode through the lateral wall of the cochlea into the scala media of the basal turn (Fig. 2H–I). FRS at 200 mg/kg produced a significant EP reduction at 1 h post-treatment (t₇ = 8.874, p < 0.0001, Fig. 2J). The microelectrode recorded a positive potential of approximately +95 mV upon entering the scale media, consistent with previous reports (Li et al. 2020).

Additionally, the acoustic startle response, commonly used to assess sensory processing, motor function, and emotional states such as anxiety, showed no significant changes 1 h after 200 mg/kg FRS (t₆ = 1.841, p = 0.1153, Fig. 2K).

Based on these results, a single IP dose of 200 mg/kg FRS was selected for subsequent experiments to evaluate its potential in prevention of NIHL.

3.3. Treatment with FRS attenuates noise-induced hearing loss.

Next, we assessed whether a single dose of FRS could attenuate NIHL. Given the short half-life of FRS in mice with normal renal function, we evaluated its preventive efficacy at three-time intervals: FRS administered 2 h before noise exposure, immediately before noise exposure, and immediately after noise exposure, as illustrated in schematic diagram (Fig. 3A).

Noise exposure alone elevated auditory thresholds to approximately 80 dB SPL across all tested frequencies at 14 d post-exposure. In contrast, immediate FRS pretreatment significantly reduced thresholds to approximately 40 dB SPL at 8, 20, and 24 kHz, and to about 60 dB SPL at 12 and 16 kHz (Fig. 3B), but it did not result in a significant improvement in DPOAE amplitudes (Fig. 3C). The preventive effect of FRS diminished as the interval between administration and noise exposure increased. When administrated 2 h before noise exposure, only minimal but statistically significant attenuation (about 10 dB) was observed at 8, 16, and 20 kHz, whereas administrated immediately after noise exposure showed no significant attenuation (for detailed statistical values, see Table B’ in Fig. 3).

These findings indicate that a single dose of FRS administration prior to noise exposure significantly attenuates NIHL, with immediate administration providing the greatest preventive effect. This pattern may be associated with temporary suppression of EP, consistent with the pharmacokinetic half-life of FRS.

3.4. Combined treatment with FRS and NAC shows additive effects in prevention of noise-induced hair cell loss and hearing loss.

Bases on the above results, we selected immediate FRS administration for subsequent experiments in combination with NAC to assess additive prevention of NIHL, as illustrated in the schematic diagram (Fig. 4A).

Consistent with previous reports (Wu et al. 2022), our results showed that treatment with NAC alone attenuated noise-induced threshold elevations at 8, 20 and 24 kHz in FVB/NJ mice. However, combined treatment with both NAC and FRS significantly enhanced the preventive effects, reducing elevated thresholds across all tested frequencies and offering significantly greater prevention than either NAC or FRS alone (Fig. 4B; for detailed statistical values, see Table B’ in Fig. 4).

DOPEA measurements, which reflect OHC function, further supported this finding. The combined treatment significantly prevented noise-induced suppression of DPOAE amplitudes from 8–32 kHz. In contrast, neither NAC alone nor FRS alone attenuated DPOAE suppression across the 8–32 kHz range (Fig. 4C; for detailed statistical values, see Table C’ in Fig. 4).

To confirm the protective effect of the combined treatment against noise-induced hair cell loss, we quantified OHCs along the entire cochlear spiral using Myosin-VIIa immunolabeling and phalloidin-stained surface preparations (Fig. 4D). As previous documented by our lab and others (He et al. 2021), noise-induced OHC loss followed a base-to-apex gradient in mice, even with OBN exposure, beginning from the basal turn. Significant OHC loss was observed at 4.5 mm from the apex and complete OHC loss at 5.5 mm. Remarkably, the combined treatment almost completely prevented OHC loss, with no significant difference compared to unexposed control mice (Fig. 4E). These findings demonstrated an additive protective effect against noise-induced hearing loss with the combined treatment with of NAC and FRS.

3.5. Treatment with FRS increases the permeability of the stria vascularis to small molecules.

Since the combined treatment showed an additive effect in preventing NIHL, we further assessed whether the transient increases in SV permeability induced by FRS could enhance the efficacy of NAC. Cadaverine and Evans Blue (EB) were used as tracers to assess SV permeability to molecules of different sizes.

FRS treatment significantly increased Alexa Fluor^™ 555 cadaverine (950 Da) signal in the cochlear lateral wall at 1 h post-treatment compared to the saline control group, with levels returning to baseline by 4 h (Fig. 5A). Semi-quantitative analysis confirmed a significant increase in relative cadaverine intensity (t₄ = 4.882, p = 0.0081, Fig. 5B). EB dye, which has a molecular weight of 960.8 Da and binds to serum albumin (67,000 Da, resulting in an estimates total of 67,961 Da), did not show significant extravasation into the cochlea lateral wall in FRS-treated mice compared to saline controls (Fig. 5C). Because EB signal is substantially reduced after cardiovascular perfusion, we evaluated EB distribution both with and without perfusion. In both conditions, EB intensity in the cochlear lateral wall did not significantly differ between FRS- and saline-treated mice at 1 h post-treatment (Fig. 5D). Semi-quantitative analysis confirmed no significant change in relative EB intensity (t₃ = 0.256, p = 0.8142, Fig. 5E). These results indicate that FRS treatment transiently increases SV permeability to small molecules but not to large protein-bound tracers.

Fig. 5.

Administration of FRS (200 mg/kg) temporarily increases stria vascularis (SV) permeability to small molecules.

(A) Cadaverine signal (red) was significantly elevated in the SV and surrounding interstitium 1 h after FRS injection compared with saline controls, returning to baseline by 4 h. IB4 (green) marks the SV. Scale bar: 10 μm.

(B) Semi-quantitative analysis confirmed a significant increase in cadaverine intensity post-FRS treatment. Data are presented as individual points with mean ± SD, normalized to saline control (n = 5 for control and 1 h groups, n = 3 in for 4 h group). Statistical analysis: one sample and two samples unpaired Student’s t-test. **p < 0.01.

(C–D) Evans blue dye showed no significant extravasation in the cochlea or SV 1 h post-FRS compared with saline controls. Evans blue fluorescence: excitation 620 nm, emission 680 nm. Scale bar: 10 μm.

(E) Semi-quantitative analysis of relative Evans blue intensity confirmed no significant difference between FRS and saline groups. Data are presented as individual points with means ± SD, normalized to the saline control group (n = 4 per group). Statistical analysis: unpaired Student’s t-test.

3.6. Combined treatment with NAC and FRS additively reduces noise-induced enhancement of 3-NT levels in outer hair cells.

Oxidative stress is a well-established mechanism underlying NIHL (Sha & Schacht 2017). Consistent with our previous findings (Yuan et al. 2015), noise exposure significantly increased 3-NT immunolabeling in OHCs compared to unexposed saline-treated controls. This increase was significantly attenuated by treatment with NAC alone. Notably, combined treatment with NAC and FRS further reduced 3-NT immunolabeling levels compared to NAC alone, indicating an additive preventive effect. In contrast, FRS alone did not significantly alter 3-NT levels compared to the noise-exposed group (Fig. 6A). Semi-quantitative analysis of 3-NT immunolabeling in OHCs confirmed these findings (Fig. 6B; for detailed statistical values, see Fig. 6C). These results suggest that the transient increase in SV permeability induced by FRS enhances the protective effect of NAC against noise-induced ROS formation.

Fig. 6.

Combined NAC and FRS treatment reduces noise-induced 3-nitrotyrosine (3-NT) immunolabeling in outer hair cells.

(A) Representative images of 3-NT immunolabeling (red) in OHCs 1 h after noise exposure in mice treated with saline alone, FRS alone, NAC alone, or NAC + FRS, compared with unexposed controls. Top row: 3-NT channel; bottom row: merged channels. Phalloidin (green) stains hair cell structure. Scale bar: 10 μm.

(B) Semi-quantitative analysis of 3-NT intensity revealed significant differences among treatment groups. Combined NAC + FU treatment more effectively suppressed noise-induced 3-NT elevation than NAC alone, while FRS alone showed no effect. Data are presented as individual points with mean ± SD, normalized to the control (n = 5 per group). Statistical analysis: one-way ANOVA with multiple comparisons test. **p < 0.01, ****p < 0.0001.

(C) Detailed multiple comparison values for panel B are provided in the figure.

4. Discussion

The salient findings of this study are that pretreatment with a single, appropriately dosed administration of FRS immediately before noise exposure not only transiently reduces EP, thereby attenuating NIHL, but also increases the permeability of the SV to small molecules. This enhanced permeability facilitates the delivery of NAC into the inner ear, amplifying its antioxidant effects and providing additive prevention of NIHL. These results provide new insights into the potential therapeutic strategies for prevention of NIHL.

Furosemide temporarily reduces EP and modulates cochlear function: a chemical earplug

Although FRS administration induces temporary hearing loss in a dose-dependent manner (Li et al. 2011, Ding et al. 2012), our findings showed that a single dose of 200 mg/kg FRS administrated to 8 week-old FVB/NJ mice does not affect hearing sensitivity and OHC function assessed by measurement of ABRs and DPOAEs, and suprathreshold wave I amplitude analysis one-hour and one day post-administration. In contrast, administration of 400 mg/kg dose results in a substantial temporary threshold elevation that persists for at least 3 days following treatment. Moreover, anesthesia exacerbates FRS-induced hearing loss (data not shown). These results support the conclusion that a single 200 mg/kg dose of FRS does not cause temporary hearing loss in this model. However, susceptibility to FRS may vary depending on mouse strain and age. To date, we have tested single doses of FRS up to 400 mg/kg in both FVB/NJ and CBA/J mice at 8 weeks of age, and no permanent hearing loss has been observed.

Our results demonstrate that a single 200 mg/kg dose of FRS transiently reduces EP without affecting ABR thresholds and DPOAE amplitudes at 1 h post-administration. This transient EP reduction likely diminishes the driving force for K⁺ influx into hair cells, thereby temporarily suppressing mechanoelectrical transduction (Nin et al. 2008). Noise-induced increases in K⁺ influx can lead to K⁺ excitotoxicity, contributing to both hair cell degeneration and ribbon synapse loss, as demonstrated in recent reports (Zhao et al. 2025a, Zhao et al. 2021). These studies show that reducing K⁺ influx, such as through the administration of K⁺-channel blockers, can attenuate noise-induced hearing loss and synaptopathy, thereby supporting and reinforcing our interpretation that transient EP reduction may reduce the electrochemical driving force for K⁺ entry into hair cells and protect against excitotoxic damage. This phenomenon, termed “transient metabolic silencing,” may functionally serve as a “chemical earplug.” Our data support this hypothesis, showing that 200 mg/kg FRS pretreatment significantly attenuates NIHL, with the most pronounced preventive effects observed when FRS is administered immediately prior to noise exposure. This timing aligns with the known short half-life of FRS in rodents (Hammarlund & Paalzow 1982). In contrast, administration immediately after noise exposure did not confer prevention, further supporting the importance of timing and temporary EP suppression prior to acoustic trauma. Furthermore, our findings align with a recent report in JCI (Shearer et al. 2025), which demonstrated that administration of FRS (200 mg/kg) reduced hair cell loss in Tmprss3-mutate mice, reinforcing the mechanistic link between EP dynamics and cochlear vulnerability.

Overall, our experimental results support the hypothesis that transient EP suppression may serve as a preventive mechanism against acoustic trauma. These findings open intriguing possibilities for pharmacological strategies aimed at preventing NIHL.

Additive preventive effects of FRS and NAC implications: Implication for therapeutic Strategies

NAC is a well-established antioxidant known for replenishing intracellular glutathione and scavenging ROS/RNS, and has been widely used to prevent NIHL in animals (Fetoni et al. 2013, Bai et al. 2022). However, despite its biological plausible and robust preclinical efficacy, clinical trials have shown only minimal and statistically marginal protective effects (Kopke et al. 2015). In this context, our findings that a single dose of FRS combined with NAC pretreatment provides additive preventive effects on both hair cell survival and hearing preservation highlight the potential value of a multimodal therapeutic approach.

Although the combined treatment significantly reduced noise-induced OHC loss in the basal turn and attenuated hearing thresholds at high frequencies, it did not lead to a significant improvement in DPOAE amplitudes in the high frequency region. This apparent discrepancy likely reflects the well-recognized frequency-dependent sensitivity of DPOAE measurements. DPOAE amplitudes are strongest and most reliably quantified in the mid-frequency range (approximately 12–16 kHz), whereas at high frequencies the signals are inherently weaker, more variable, and less responsive to subtle changes in OHC integrity. Consequently, even though NAC+FRS treatment produced clear histological prevention of OHC loss and measurable recovery of hearing thresholds in the basal, high-frequency region, these improvements may not have translated into detectable changes in DPOAE amplitudes at the uppermost test frequencies due to the intrinsic limitations of the assay. Preservation or partial recovery of OHCs does not necessarily yield proportional gains in high-frequency DPOAEs, particularly when DPOAE output is already near the noise floor. This frequency-dependent constraint provides a likely explanation for why enhanced OHC survival and improved auditory thresholds in the basal turn did not correspond to a statistically significant functional benefit in high-frequency DPOAEs.

The observed synergy between FRS and NAC likely arises from complementary mechanisms: FRS reduces ionic stress by transiently silencing cochlear metabolic activity, whereas NAC enhances the neutralization of ROS/RNS, key mediators of noise-induced cochlear injury. This dual action is supported by the significant decrease in 3-NT levels in OHCs following combined treatment, indicating reduced nitrosative stress. Although NAC is a plausible contributor to this reduction, the observed decrease in 3-NT is correlative and may involve additional mechanisms beyond NAC-mediated antioxidant activity. Future studies employing NAC-specific tracking or genetic approaches will be necessary to determine the causal contribution of NAC to ROS/RNS suppression in this context.

Overall, our results reinforce a multifactorial model of NIHL in which both ionic dysregulation and oxidative damage paly central roles. Therapeutic strategies that target these distinct yet interconnected pathways may therefore offer superior preventive benefits compared to monotherapies. Importantly, the ability of FRS to modulate cochlear physiology without causing permanent damage, coupled with its potential to enhance drug permeability, positions it as a promising adjunct in combination therapies for NIHL. These results provide a rationale for further exploration of FRS-based combination strategies aimed at maximizing protection against noise-induced auditory injury.

FRS-enhanced transient permeability of the stria vascularis: implication for drug delivery

The BLB plays a critical role in protecting the cochlea from toxic substances and pathogens. However, this barrier also limits the entry of therapeutic agents into the inner ear (Nyberg et al. 2019). Our findings, consistent with previous reports (Ding et al. 2016, Videhult Pierre et al. 2020), demonstrate that a single, appropriately dosed administration of FRS transiently increases the SV permeability without causing permanent functional changes. This transient permeability enhancement may facilitate the delivery of otoprotective compounds, offering a novel strategy to overcome the challenge of cochlear drug delivery.

We observed that FRS enhances SV permeability to small molecules, suggesting an additional mechanism by which it may facilitate the efficacy of co-administrated agents. This effect was particularly evident when FRS was combined with NAC, improving its access to vulnerable cochlear regions and enhancing its preventive effects. However, the extent and duration of SV permeability changes varied across mouse trains. In CBA/J mice, permeability caused by 200 mg/kg FRS is recovered within 6 h (Wu et al. 2025), whereas in the FVB/NJ mice used in the present study, recovery occurred within 4 h.

The observed difference between EP reduction window and the BLB permeability recovery period following FRS treatment reflects distinct physiological mechanisms and recovery kinetics of EP and BLB integrity. EP is primarily maintained by active ion transport in the stria vascularis. FRS rapidly disrupts these processes rapidly, causing an acute drop in EP. Recovery generally occurs relatively quickly once the drug is cleared and transport resumes. For reference, the half-life of FRS in rodents is approximately 2.5 h and is expected to be shorter in mice (Hammarlund & Paalzow 1982). In contrast, BLB integrity depends on structural components such as tight junctions and endothelial cell function, which require more time to repair following disruption (Hirose & Liberman 2003). The 4-hour window for BLB recovery likely reflects the time needed for cellular and vascular restoration rather than ongoing EP dysfunction. Therefore, although monitoring durations differ, early EP recovery may help re-establish cochlear homeostasis and facilitate subsequent BLB repair.

Despite these promising findings, several limitations should be acknowledged. First, the study was conducted exclusively in FVB/NJ mice, which may limit the generalizability of the results to other strains or species. Second, although we observed transient increase in SV permeability, the molecular mechanisms underlying this effect remain unclear and warrant further investigation. Future studies should explore optimal dosing, timing, and long-term safety across species, as well as the translational potential of this combinatorial strategy in clinical settings.

FVB/NJ mice as a model for noise-induced hair cell loss

Additionally, our results demonstrate that male FVB/NJ mice are particularly sensitive to noise exposure. Specifically, exposure to 100 dB octave-band noise (OBN) for 2 h resulted in threshold elevations of approximately by 80 dB SPL across six tested frequencies (from 8–32 kHz). This robust and reproducible response provided a suitable model for elevating additive preventive strategies for NIHL. In contrast, females showed highly variable responses, ranging from no hearing loss to complete hearing loss. This sex-dependent vulnerability aligns with previous reports (McFadden et al. 1999) and may be influenced by hormonal modulation of cochlear physiology or differential expression of protective genes. Estrogen, for instance, has been shown to exert neuroprotective and antioxidant effects in the cochlea, potentially contributing to the relative resilience observed in females (Villavisanis et al. 2020).

However, the absent of ribbon degeneration 14 days after noise exposure in FVB/NJ male mice is indeed unexpected. This outcome contrasts with our previously published findings in CBA/J mice (Hill et al. 2016) and with the seminal report in CBA/CaJ mice (Kujawa & Liberman 2009), as well as extensive work in C57BL/6 mice (Wan et al. 2014). Across these strains, comparable noise exposure paradigms consistently induce cochlear synaptopathy, as evidenced by reductions in ribbon counts. Our findings also diverge from the only published study using FVB/NJ mice to data, which reported synaptic loss restricted to 24 kHz based on a small sample size examined 14 days after moderate noise exposure (Paquette et al. 2016). Despite this, our results were consistent across biological replicates, suggesting that FVB/NJ mice may exhibit an atypical pattern of post-noise cochlear pathology in which OHC vulnerability does not directly parallel IHC presynaptic vulnerability. This strain-specific dissociation highlights an important biological question requiring further mechanistic investigation.

One possible explanation is partial or complete recovery of ribbon synapses following transient injury. Macrophage-mediated synaptic remodeling has been implicated in such recovery processes. After noise exposure, macrophages infiltrate the cochlea, where they facilitate debris clearance, modulate local immune signaling, and promote neurite stabilization and reinnervation (Zhang et al. 2021). Macrophage activation has been associated with restoration of synaptic puncta and improved afferent terminal integrity, raising the possibility that the initially damaged synapses may return to near-baseline levels by 14 days post-exposure (Manickam et al. 2023, Kaur et al. 2019). Strain-specific immune profiles may further contribute to these differences. FVB mice exhibit a Th2-skewed, anti-inflammatory background, whereas C57BL/6 mice are Th1-biased and more pro-inflammatory, difference known to influence outcomes in multiple inflammation-related conditions (Kim et al. 2014). Such immune variation may shape the cochlear response to noise and the balance between synaptic injury and repair.

Additionally, noise-induced synapse alterations may occur without overt synapse loss. Changes in glutamatergic receptor composition, including AMPAR (GluA2), NMDAR (NR1), metabotropic GluR (mGluR7), and GABAergic receptors, could contribute to functional changes. A more detailed examination of ribbon ultrastructure, postsynaptic receptor organization, and longitudinal synaptic remodeling in FVB/NJ mice is therefore warranted.

In summary, our results demonstrated that a single dose of FRS (200 mg/kg) administrated immediately before noise exposure transiently reduces EP and attenuates NIHL. When combined with NAC, this approach additively prevents noise-induced hair cell loss and NIHL from complementary mechanisms: FRS reduces ionic stress by transiently silencing cochlear metabolism (acting as a “chemical ear plugs”), while NAC neutralizes ROS/RNS, the key mediators of noise-induced cochlear injury.

Abbreviations:

3-NT

3-nitrotyrosine

ABR

Auditory brainstem response

BLB

blood-labyrinth barrier

DPOAE

Distortion product optoacoustic emissions

EB

Evans Blue

EP

endocochlear potential

FDA

U.S. Food and Drug Administration

FRS

Furosemide

IHCs

inner hair cells

NAC

N-acetylcysteine

NIHL

Noise-induced hearing loss

OBN

octave-band noise

OHCs

outer hair cells

PBS

phosphate-buffered saline

PFA

paraformaldehyde

ROS

reactive oxygen species

SV

stria vascularis