Ursolic acid protects against cisplatin‑induced ototoxicity by inhibiting oxidative stress and TRPV1‑mediated Ca2+‑signaling.

Cisplatin (CDDP) is widely used in clinical settings for the treatment of various cancers. However, ototoxicity is a major side effect of CDDP, and there is an associated risk of irreversible hearing loss. We previously demonstrated that CDDP could induce ototoxicity via activation of the transient receptor potential vanilloid receptor 1 (TRPV1) pathway and subsequent induction of oxidative stress. The present study investigated whether ursolic acid (UA) treatment could protect against CDDP‑induced ototoxicity. UA is a triterpenoid with strong antioxidant activity widely used in China for the treatment of liver diseases. This traditional Chinese medicine is mainly isolated from bearberry, a Chinese herb. The present results showed that CDDP increased auditory brainstem response threshold shifts in frequencies associated with observed damage to the outer hair cells. Moreover, CDDP increased the expression of TRPV1, calpain 2 and caspase‑3 in the cochlea, and the levels of Ca2+ and 4‑hydroxynonenal. UA co‑treatment significantly attenuated CDDP‑induced hearing loss and inhibited TRPV1 pathway activation. In addition, UA enhanced CDDP‑induced growth inhibition in the human ovarian cancer cell line SKOV3, suggesting that UA synergizes with CDDP in vitro. Collectively, the present data suggested that UA could effectively attenuate CDDP‑induced hearing loss by inhibiting the TRPV1/Ca²+/calpain‑oxidative stress pathway without impairing the antitumor effects of CDDP.

Introduction

Cisplatin (CDDP) is a divalent platinum compound with broad-spectrum anticancer activity that synergizes with a variety of antitumor drugs without cross-resistance, making it a first-line anticancer drug (1). CDDP is commonly used for the treatment of various cancers including bladder cancer, lung cancer, ovarian cancer and other soft tissue tumors. In children, CDDP is also effective for the treatment of various solid tumors (2,3). However, CDDP has serious side effects, including irreversible hearing loss, especially in juvenile patients treated with CDDP and radiotherapy for at least six months (4-6).

The mechanism of CDDP-induced ototoxicity remains to be elucidated. CDDP could induce apoptosis of cochlear cells, which led to hearing loss (7-9). Several studies also showed that CDDP increased NADPH oxidase 3 (NOX 3) expression in the cochlea, which increased the production of reactive oxygen species (ROS) (10,11). ROS directly attacked the cochlear cell membrane and produced the toxic lipid peroxidation product 4-hydroxynonenal (4-HNE), which might also account for cochlear cell apoptosis (12). A study showed that CDDP-induced ototoxicity in the cochlea was related to calcium (Ca2+) over-load (13). As an important second messenger, Ca2+ is involved in various functions, including auditory signal transduction (14). Moreover, maintenance of a low intracellular Ca2+ concentration in the normal resting state plays an important role in sustaining homeostasis (15). Abnormal influx of free Ca2+ in cells (referred to as Ca2+ overload) can cause structural damage and disruption of intracellular metabolism of hair cells (HCs), eventually leading to apoptosis of HCs (16). In the cochlea, Ca2+ overload was related to high expression of the transient receptor potential vanilloid receptor 1 (TRPV1) (17). TRP channels are a family of non-selective cation ion channels that control important cellular functions and signaling pathways (18). TRP channels are also important targets for a wide range of drugs, such as menthol, which adjust human TRP melastatin 8 (hTRPM8)-induced signaling, and could alleviate hyperglycemia in alloxantreated mice (a model of type 1 diabetes) or reverse muscle atrophy in dexamethasone-treated mice (a model of muscle wasting) (19). TRP channel activation can lead to changes in Ca2+ homeostasis, resulting in long lasting modulation of Ca2+ levels (20,21). TRPV1 is expressed in many organs, including the heart, kidney, brain, dorsal root ganglions and sensory neurons (22). TRPV1 plays a critical role in the maintenance of cellular homeostasis, particularly in cochlear cells (23). Studies have suggested that CDDP induced Ca2+ overload by increasing TRPV1 expression, and that Ca2+ subsequently activates calpain, a neutral cysteine protease, to cause cochlear apoptosis (24). However, the association between Ca2+ overload and TRPV1 remains to be confirmed.

Ursolic acid (UA) is a plant-derived triterpene compound with strong antioxidant effects (25-27). UA is the main ingredient of several traditional Chinese herbal medicines used in the treatment of liver diseases and in lowering blood lipid (28,29). In PC12 cells, UA could assert its neuroprotective effects by inhibiting ROS production induced by amyloid beta peptide (30,31). In addition, UA could capture oxygen free radicals in the P450 monoamine oxidase system and liver microsome, thus exerting a strong inhibitory effect on lipid peroxidation (32). However, to the best of our knowl-edge, there are currently no reports on the anti-apoptotic and anti-oxidative effects of UA on ototoxicity induced by CDDP. The present study evaluated the protective effects of UA on CDDP-induced ototoxicity and the underlying mechanisms were investigated. It was hypothesized that UA could protect against CDDP-induced ototoxicity by inhibiting oxidative stress and TRPV1-mediated Ca2+-signaling in the cochlear cells.

Materials and methods

Animal and experimental protocols

A total of 48 BALB/c mice (3 days) were purchased from the Animal Experimental Center, Jinzhou Medical University, China. A total of 48 BALB/c mice (20-22 g, 4 weeks, male and female) with normal Preyer's reflexes were purchased from the Animal Experimental Center, Dalian Medical University, China [license no. SCXK (Liao) 2008-0002]. All animal studies complied with the regulations for the Management of Laboratory Animals published by the Ministry of Science and Technology of the People's Republic of China, and all animal protocols were approved by the Animal Care and Use Committee of Jinzhou Medical University. All mice were fed a standard commercial diet and housed at an ambient temperature of 20-24°C with a relative humidity of 50±5% under a 12/12 h light-dark cycle in a specific pathogen-free facility during the experiment. All animals were allowed 1 week to acclimate before treatment and then were randomly divided into four groups (n=12 in each group) and treated for five days as follows: i) Normal control group, where the mice received physiological saline [0.5 ml/100 g, intra-peritoneally (i.p)]; ii) UA group, where the mice received UA (purity ≥98%; cat. no. N1823, APExBIO Technology LLC; 80 mg/kg/day, i.p); iii) CDDP group, where the mice received CDDP (Sigma-Aldrich; Merck KGaA; 4.5 mg/kg/day, i.p,) and iv) UA plus CDDP (UA+CDDP) group, where the mice received UA (80 mg/kg/day, i.p) followed by CDDP (4.5 mg/kg/day, i.p) 30 min later each day (Fig. 1A).

Figure 1

UA prevents CDDP-induced hearing loss and cochlea apoptosis in mice. (A) The flow chart of the experimental protocol. (B) ABR results at tone bursts of 8, 12, 24 kHz. n=12 mice for each group. **P<0.01. (C) UA prevented CDDP-induced OHC loss. OHCs were labeled by tetramethylrhodamine-phalloidin staining in the basilar membrane. White asterisks indicate cell loss. Scale bar, 20 µm. OHCs exhibited regional variation in their susceptibility to cisplatin, with gradual severity of hair cell loss from the apical turn to the basal turn. Mice receiving UA + CDDP demonstrated fewer OHC loss. (D) The numbers of intact hair cells on the apical, middle and basal turns. The graph shows quantitative differences of OHC loss at apical, middle and basal turns of the cochlear basilar membrane The UA + CDDP group exhibited significantly fewer hair cell loss in comparison compared with the CDDP group. n=3 ears for each group. **P<0.01 (E) Western blot analysis was performed to detect protein expression levels of cleaved caspase-3 and caspase-3 in cochlea, with β-actin as a loading control. n=6 ears for each group. Mean values were obtained from n=3 independent experiments. **P<0.01. ABR, auditory brainstem response; UA, ursolic acid; i.p, intraperitoneal; CDDP, cisplatin; OHC, outer hair cell; SPL, sound pressure levels.

During the treatment period, the weight of the experimental animals was measured every day (Fig. S1). Although CDDP is toxic, the survival rate of mice is 100% during the administration (33).

Measurement of auditory brainstem response (ABR)

For the analysis of the auditory threshold, ABR was recorded 1 day before and 5 days after related drug treatment with tone bursts of 8, 12 and 24 kHz using the Smart EP & OAE auditory evoked potential recording system (Intelligent Hearing Systems). The mice were anesthetized using pentobarbital sodium (90 mg/kg) and kept warm with a heating pad during ABR recording. A subdermal (active) needle electrode was inserted at the vertex, while ground and reference electrodes were inserted subdermally in the loose skin beneath the pinnae of opposite ears. The technique used to record ABRs has been previously described in detail (34). The ABR waveforms were averaged over a 10 msec time window using the Smart EP & OAE auditory evoked potential recording system software (Launch Pad 2.32 HIS program). The sound intensity was varied at 5 dB intervals near the hearing threshold. The differences in ABR thresholds shift for each frequency between the starting and the terminal points of the experimental time course were noted. The threshold was determined off-line by two independent, experimentally blinded observers based on the ABR records. In addition, the mice were euthanized, and double cochleae were removed for further analysis.

Outer hair cell (OHC) counting

Immediately after completion of the ABR test, the animals were killed by cervical dislocation under anesthesia with an intraperitoneal injection of pentobarbital sodium (90 mg/kg). An array of criteria was used to verify the death of animals, including absence of respiration and heartbeat and loss of corneal and palpebral reflexes.

The temporal bones were removed, and the cochlea was immersed in 4% paraformaldehyde and incubated overnight at 4°C. The cochlea was decalcified in 4% EDTA for 7 days at 4°C. Subsequently, the basilar membrane was dissected under a dissecting microscope, and the stria vascularis and tectorial membrane were removed. The basilar membrane was then stained with phalloidin-tetramethylrhodamine (TRITC; Sigma-Aldrich; Merck KGaA) for 20 min at 20-24°C and protected from light. The specimens were then rinsed three times with 0.01 M phosphate-buffered saline (PBS; pH 7.4) and then mounted on slides. Images were obtained using the TCS-SP5 II laser-scanning confocal microscope (Leica Biosystems GmbH). Outer hair cells were counted from the captured images using a fluorescent BX41 microscope (Olympus Corporation) at ×200 magnification in 20 consecutive fields from the apex to the basal turn along the entire length of the cochlear epithelium. The percentages of outer hair cell loss in each 0.5-mm length of epithelium were plotted as a function of the cochlear length as a cytocochleogram (34).

Immunohistochemical staining of paraffin sections of the cochlea

After the last auditory brainstem response test, five mice from each group were decapitated under abdominal anesthesia using pentobarbital sodium (90 mg/kg), and the temporal bones were removed immediately. The round and oval windows were opened, cochleae were perfused with 40 g/l paraformaldehyde (pH 7.4), and the specimens were immersed in the 4% paraformaldehyde for 24 h at 4°C. Following decalcification in 40 g/l EDTA solution (pH 7.4) for 5 days at 4°C, the cochleae were dehydrated with ethanol, cleared with xylene and embedded in paraffin wax. Serial sections (5-µm thickness) were prepared, dewaxed, rehydrated in descending alcohol series, rinsed with distilled water and treated with high pressure (125°C; 103 kPa; 3 min) to retrieve antigens. Endogenous peroxidase activity was blocked with 3 g/l H2O2 for 10 min at 20-24°C, followed by a PBS wash. After incubation with normal goat serum (cat. no. C0265; Beyotime Institute of Biotechnology) for 15 min at 20-24°C, tissue sections were stained overnight at 4°C with primary antibodies for TRPV1 (1:500; cat. no. ab6166; Abcam) and calpain 2 (1:1,000; cat. no. sc-373967; Santa Cruz Biotechnology, Inc.). After washing with PBS, sections were stained with horseradish peroxidase-conjugated-biotinylated goat anti-rabbit immunoglobulin G (IgG) secondary anti-body (1:1,000; cat. no. ab205718; Abcam) for 1 h at 20-24°C. Following washing with PBS, color was developed with 0.05% 3,3'-diaminobenzidine for 3 min at 20-24°C. Sections were counterstained with hematoxylin for 10 min at 20-24°C, dehydrated with gradient concentrations of ethanol, vitrified with xylene and mounted with neutral gum. PBS was used as a negative control throughout the study. The sections were observed by using a BX41 fluorescent microscope at ×400 magnification (Olympus Corporation). Relative optical intensity was determined using Image-Pro Plus 6.0 software (Media Cybernetics, Inc.).

Western blot analysis

Total cochlea tissue was homogenized in RIPA buffer (cat. no. P0013B; Beyotime Institute of Biotechnology) containing protease inhibitor cocktail tablets and phosphatase inhibitors cocktail (Roche Diagnostics GmbH). The tissue homogenate was sonicated for 30 sec and centrifuged at 14,000 × g at 4°C for 30 min to extract the supernatant. Protein concentrations were determined using a bicinchoninic acid kit. The protein samples (50 µg/lane) were separated using SDS-PAGE (7.5% for TRPV1 and calpain 2; 10% for 4-HNE and procaspase 3 and 12.5% for cleaved caspase 3). Following electrophoresis, the proteins were transferred onto nitrocellulose membranes (Thermo Fisher Scientific, Inc.). The membranes were then blocked with 10% bovine serum albumin for 1 h at 4°C and incubated overnight at 4°C with the following primary antibodies: Anti-TRPV1 (1:500; cat. no. ab6166; Abcam), anti-calpain 2 (1:1,000; cat. no. sc-373967; Santa Cruz Biotechnology, Inc.), anti-4-HNE (1:800; cat. no. ab48506; Abcam), anti-cleaved caspase-3 (1:1,000; cat. no. ab13847; Abcam) and anti-caspase-3 (1:1,000; cat. no. sc-7272; Santa Cruz Biotechnology, Inc.). After washing three times with TBS-0.05% Tween-20, membranes were incubated with IRDye 800CW goat anti-rabbit IgG secondary antibody (1:10,000; cat. no. 925-32211; LI-COR Biosciences.) for 1 h at 4°C. Following extensive washing, the immunoreactive bands were visualized by ECL. The Odyssey CLx Infrared imaging system (LI-COR Biosciences) was used to analyze the optical density value of immunoreactive bands. β-actin (1:1,000; cat. no. sc-47778; Santa Cruz Biotechnology, Inc.) was used as a loading control.

Organotypic cultures of postnatal organ of Corti

The organ culture procedures were modified from a previously published report (35). In brief, Cochlear explants were isolated from newborn 3-day-old BALB/c mice. BALB/c pups were euthanized following antisepsis using 75% ethanol. The two otic capsules were extracted from surrounding tissue and immersed ice-cold Hank's balanced salt solution (HBSS; cat. nos. CC014 and CC016; M&C Gene Biotechnology). The lateral wall tissues (stria vascularis and spiral ligament) and the auditory nerve bundle were micro-dissected for sterile dissection under a binocular microscope (Olympus SZX7; Olympus Corporation). The explants were carefully placed onto a prepared culture dish containing a 15-µl polymerized drop of rat tail collagen in 1 ml culture medium consisting of DMEM/F12 (Invitrogen; Thermo Fisher Scientific, Inc.), 1% bovine serum albumin (Invitrogen; Thermo Fisher Scientific, Inc.), and 10 U/ml penicillin G. Following 4 h of incubation (37°C, 5% CO2), an additional 1 ml medium was added to submerge the explants.

Measurement of Ca2+ levels in OHCs

Cell-permeable Ca2+ indicator Fluo-4 AM (cat. no. F14201; Invitrogen; Thermo Fisher Scientific, Inc.) were used to monitor levels of intracellular Ca2+, based on the literature (36). Following immersion in HBSS without Ca2+ and Mg2+ for 20 min at 20-24°C, the organ explants were immersed in dye (cat. no. F14201; Invitrogen; Thermo Fisher Scientific, Inc.) for 25 min following Fluo4-AM loading at 24°C. Subsequently, the explants were fixed in 4% paraformaldehyde for 30 min at 20-24°C. Phalloidin (1:200) was applied for 1 h at 20-24°C and protected from light. For measurement of Fluo-4 Ca2+ signals, organs were imaged under a BX41 microscope at ×400 magnification (Olympus Corporation) with epifluorescence, and the images were obtained using a TCS-SP5II laser-scanning confocal micro-scope (×400 magnification; Leica Biosystems GmbH) in 20 consecutive fields.

Immunofluorescence for assessment of 4-HNE

Preparation of cochlear paraffin sections (5-µm thickness), deparaffinization, rehydration, antigen retrieval and blocking nonspecific binding were performed as described in a previous study (34). Tissue sections were incubated with rabbit anti-4-HNE antibody (1:200; cat. no. ab46545; Abcam), followed by Cy3-labeled goat anti-rabbit IgG secondary antibody (1:500; cat. no. A0516; Beyotime Institute of Biotechnology) incubation for 1 h at 24°C. Immunofluorescence images were obtained via TCS-SP5II laser-scanning confocal microscopy (×400 magnification; Leica Biosystems GmbH). Relative optical intensity was determined using Image-Pro Plus 6.0 software (Media Cybernetics, Inc.).

Measurement of cell growth inhibition by MTT assay

The human ovarian cancer cell line SKOV3 (Peking Union Medical College, Beijing) were seeded at a density of 3×103 cells/well in 96-well plates and treated with different concentrations (1, 2, 3, 4, 8, 10 and 16 µg/ml) of CDDP. Each concentration treatment was repeated in six separate wells. Cells treated with RPMI 1640 media (100 ul; Sigma-Aldrich; Merck KGaA) alone acted as a control. The cells were incubated at 37°C with 5% CO2 for 24 h, and MTT reagent (20 µl; 5 mg/ml in PBS; Sigma-Aldrich; Merck KGaA) was added to each well and incubated at 37°C with 5% CO2 for 4 h. The absorbance in each well was measured at a wavelength of 570 nm using an enzyme-linked immunoassay microplate reader. According to the tumor cell growth inhibition rate, 4 µg/ml of CDDP was determined as optimal inhibitory concentration. Subsequently, SKOV3 cells were seeded at a density of 3×103 cells/well in a 96-well plate and treated with CDDP (4 µg/ml). Different concentrations (20, 40, 80 and 160 mol/l) of UA were added later. Each concentration of UA treatment was repeated in six separate wells. The cells were incubated at 37°C with 5% CO2 for 24 h, and MTT reagent (20 µl; 5 mg/ml in PBS; Sigma-Aldrich; Merck KGaA) was added to each well and incubated at 37°C with 5% CO2 for 4 h. The absorbance in each well was measured at a wavelength of 570 nm using an enzyme-linked immunoassay microplate reader (BioTek Instruments, Inc.).

Statistical analysis

Data are presented as the mean ± SD and were analyzed using SPSS 17.0 software (SPSS, Inc.). Intergroup differences in mean values were compared by one-way ANOVA followed by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

Results

UA prevents CDDP-induced hearing loss and hair cell loss in vivo

To investigate whether UA could prevent CDDP-induced hearing loss, the ABR test was performed on mice. In the control group, ABR thresholds at 8, 12, and 24 kHz were observed to be stable (0.48±1.43, 0.50±1.37 and 0.67±1.44 dB, respectively). However, ABR threshold shifts in the CDDP group were significantly increased at 8, 12 and 24 kHz (25.78±2.06, 27.59±2.86 and 28.51±2.32 dB, respectively), compared with the control group (P<0.01). In the UA + CDDP group, ABR threshold shifts significantly reduced at 8, 12 and 24 kHz (P<0.01; 3.93±1.19, 4.63±1.16 and 5.38±1.36 dB, respectively), compared with the CDDP group. In the UA group, UA treatment alone did not affect the ABR threshold shifts compared with the control group. These results suggested that UA can mitigate CDDP-induced hearing loss in mice (Fig. 1B).

To further determine the in vivo protective effects of UA against CDDP, TRITC staining of OHCs was performed on the basilar membrane of mouse cochlea and the number of surviving OHCs was counted. In the control and UA groups, OHCs on the basilar membrane appeared normal in structure and the staining pattern was homogeneous. Moreover, the loss rate of OHCs at each turn of the basilar membrane was <10%. In the CDDP group, the stereocilia of OHCs were characterized by loss and disorder. OHCs exhibited regional variation in their susceptibility to cisplatin, with gradual severity of OHC loss from the apical to the basal turn, and loss rates of OHCs at apical, middle and basal turns reaching 16, 41 and 61%, respectively, which was significantly higher compared with the control group (P<0.01; Fig. 1C and D). However, compared with the CDDP group, animals receiving UA + CDDP demonstrated significantly fewer OHC loss, especially at basal and middle turns, with the loss rate of OHCs at apical, middle and basal turns reaching 13, 25 and 32%, respectively (P<0.01; Fig. 1C and D). These results suggested that UA effectively protects against CDDP-induced OHC structural damage. Numerous studies showed that CDDP mainly damaged OHCs, while the damage of inner hair cells was weak (8). Hence, the present study focused on the outer hair cell change in and inner hair cell counting was not investigated.

Western blot analysis showed that the CDDP group expressed significantly higher levels of cleaved caspase-3 protein (a marker for apoptosis) compared with the control group (P<0.01; Fig. 1E). By contrast, cleaved caspase-3 expression levels in the UA + CDDP group were significantly lower compared with the CDDP group (P<0.01; Fig. 1E). However, cleaved caspase-3 expression levels in the UA group were not significantly different compared with the control group (Fig. 1E). Again, these results are consistent with the notion that UA can prevent the auditory loss observed in mice treated with CDDP.

UA inhibits CDDP-induced activation of the TRPV1/Ca2+/calpain 2 signaling pathway in the cochleae

Based on immunohistochemistry results (Figs. 2A and 3A), TRPV1 and calpain 2 were uniformly distributed on the OHCs, spiral ganglion (SG), and stria vascularis (SV) of the cochlea in the control group. By contrast, TRPV1 and calpain 2 staining was unevenly distributed in the CDDP group, and the expression levels of these proteins significantly increased compared with the control group (P<0.01; Figs. 2B and 3B). In the UA + CDDP group, TRPV1 and calpain 2 staining was uniformly distributed, and expression significantly decreased compared with the CDDP group (P<0.01, respectively; Figs. 2B and 3B). UA treatment alone did not affect TRPV1 and calpain 2 expression in the mouse cochlea (Figs. 2A and B and 3A and B). These results were consistent with the results of western blot analysis (P<0.01; Figs. 2C and 3C).

Figure 2

Effect of UA on CDDP-induced expression of TRPV1 in mouse cochlea. (A) Representative immunohistochemical staining images (brown and yellow granules) of TRPV1 on the organ of Corti, SG and SV of the cochleae. Arrows indicate the location of TRPV1-positive expression. Scale bar, 20 µm. (B) Quantification of TRPV1 immunolabeling in the different groups. n=3 ears for each group. **P<0.01. (C) Western blot analysis of TRPV1 was used to detect protein expression of TRPV1 in cochlea, with β-actin as a loading control. n=6 ears for each group. Mean values were obtained from n=3 independent experiments. **P<0.01. UA, ursolic acid; CDDP, cisplatin; SG, spiral ganglion; SV, stria vascularis; TRPV1, transient receptor potential vanilloid receptor 1.

Figure 3

Effect of UA on CDDP-induced expression of calpain 2 in mouse cochlea. (A) Representative immunohistochemical staining images (brown and yellow granules) of calpain 2 on the organ of Corti, SG and SV of the cochleae. Arrows indicate the location of calpain 2-positive expression. Scale bar, 20 µm. (B) Quantification of calpain 2 immunolabeling in the different groups. n=3 ears for each group. **P<0.01. (C) Western blot analysis of calpain 2 was used to detect protein expression of calpain 2 in cochlea, with β-actin as a loading control. n=6 ears for each group. Mean values were obtained from n=3 independent experiments. **P<0.01. UA, ursolic acid; CDDP, cisplatin; SG, spiral ganglion; SV, stria vascularis.

In the control group, Ca2+ staining was uniformly distributed in OHCs of the basement membrane (Fig. 4A). In the CDDP group, Ca2+ staining was unevenly distributed (Fig. 4A), and its intensity significantly increased compared with the control group (P<0.01; Fig. 4B). In the UA + CDDP group, Ca2+ staining was more uniformly distributed (Fig. 4A), with significantly lower intensity compared with the CDDP group (P<0.01; Fig. 4B). The results suggested that CDDP could cause Ca2+ overload, and that UA could effectively inhibit Ca2+ influx in mice cochleae.

Figure 4

UA inhibits CDDP-induced calcium overload in OHCs. (A) Effect of UA on CDDP-induced measurement of Ca2+ levels (green) in OHCs from organotypic culture. OHCs were labeled by TRITC staining (red). Scale bar, 20 µm. (B) Quantitative levels of Fluo-4 Ca2+ signals from the cochlea. n=3 ears for each group. Mean values were obtained from n=3 independent experiments. **P<0.01. TRITC, tetramethylrhodamine; UA, ursolic acid; CDDP, cisplatin; OHC, outer hair cell.

UA alleviates oxidative stress in mice with CDDP-induced ototoxicity

To investigate whether UA could alleviate oxidative stress, the levels of 4-HNE were analyzed. Based on immunofluorescence results, CDDP significantly enhanced production of 4-HNE in the OHCs, SG, and SV of mouse cochlea compared with the control group (P<0.01; Fig. 5A and B). However, 4-HNE production was significantly inhibited in the UA + CDDP group compared with the CDDP group (P<0.01; Fig. 5A and B). Moreover, compared with the control group, UA treatment alone did not affect 4-HNE production in the mouse cochleae (Fig. 5B). These results were confirmed by western blot analysis of 4-HNE levels (P<0.01; Fig. 5C). The results suggested that UA could prevent an increase in oxidative stress-related markers in the cochlea of mouse treated with CDDP.

Figure 5

UA alleviates oxidative stress in the mouse cochlea with CDDP-induced ototoxicity. (A) Effect of UA on CDDP-induced expression of 4-HNE (red) in mouse cochlea. Images of paraffin sections of the organ of Corti, SG and SV were from the cochleae. Arrows indicate the location of 4-HNE-positive expression. Scale bar, 20 µm. (B) Comparison of the relative fluorescence intensity values for 4-HNE in three different sites of the mouse cochlea. n=3 ears for each group. **P<0.01. (C) Western blot analysis was used to detect protein expression of 4-HNE in the cochlea, with β-actin as a loading control. n=6 ears for each group. Mean values were obtained from n=3 independent experiments. **P<0.01. UA, ursolic acid; CDDP, cisplatin; OHC, outer hair cell; 4-HNE, 4-hydroxynonenal.

The effect of UA on the antitumor effect of CDDP

Since CDDP is used as an anticancer therapeutic, it was investigated whether UA co-treatment would impair its antitumor effects. MTT assay was performed to evaluate the viability of the human ovarian cancer cell line SKVO3 in response to CDDP, UA, and UA + CDDP. While CDDP inhibited the growth of SKVO3 cells in a dose-dependent manner (Fig. 6A), growth inhibition of SKVO3 cells was significantly enhanced following UA + CDDP (4 g/ml) treatment. Moreover, UA inhibited the growth of SKOV3 cells in a dose-dependent manner (P<0.01; Fig. 6B), suggesting that UA enhanced the antitumor effect of CDDP in vitro.

Figure 6

Effects of UA on the antitumor effect of CDDP. (A) Inhibitory effect of CDDP on the growth of SKVO3 cells. CDDP at a concentration of 4 µg/ml was determined as the optimal inhibitory concentration based on growth inhibition rate after 24 h. (B) The inhibitory effect of CDDP combined with different concentrations of UA on SKVO3 cells. n=6 for each group. Mean values were obtained from n=3 independent experiments. **P<0.01. UA, ursolic acid; CDDP, cisplatin.

Discussion

UA is a triterpenoid compound found in >60 varieties of plants, including bearberry, Chinese elder herb and hawthorn (37). The therapeutic effects of UA include sedative, anti-inflammatory, anti-ulcer and anti-diabetic effects (38,39). A previous study demonstrated that UA significantly alleviated hydrogen peroxide-induced apoptosis against HEI-OC1 cells in vitro (40). In the present study, UA was found to inhibit CDDP-induced cochlear OHC damage, which in turn significantly attenuated hearing loss at low and high frequencies in mice. Therefore, based on the present findings, it was proposed that UA could provide therapeutic benefits for CDDP-induced ototoxicity.

TRPV1 channels are highly permeable to intracellular calcium influx, and overexpression of TRPV1 induces Ca2+ overload and the activation of calpains, especially calpain 2 (41). Activation of calpain 2 might mediate caspase-3-dependent apoptosis of cochlear cells by down-regulating Bcl-xl and up-regulating Bax (42-45). The TRPV1 channel was previously reported to play a critical role in CDDP-induced ototoxicity (46). Moreover, several studies have shown that TRPV1 expression is significantly increased in OHC, SG and SV in the cochlea in response to CDDP treatment (47-49). The present study found that UA decreased the levels of CDDP-mediated TRPV1 protein expression in the cochlear OHC, SG, and SV. In addition, CDDP dysregulated Ca2+ homeostasis in the cochlea, which is accompanied by increased expression of calpain 2. UA was previously shown to alleviate coughing via TRPV1 channel desensitization (50). Whether a similar mechanism can account for inhibition of CDDP-mediated TRPV1 protein expression by UA remains to be elucidated. TRPV1 is also activated by heat (>42°C), capsaicin, low extracellular pH and oxidative stress products (51). Thus, the antioxidant activity of UA might inhibit activation of the TRPV1 channel by reducing oxidative stress products in the cochlea. The present results suggested that UA significantly blocked CDDP-induced Ca2+ overload in hair cells and also inhibited calpain 2 expression. These results suggested an association between the TRPV1 channel and the Ca2+/calpain signaling pathway, which might explain the decrease in CDDP-induced Ca2+ influx observed in the cochlea following UA treatment.

Oxidative stress is considered to be an important cause of CDDP-induced hearing loss (52). The main source of ROS is NADPH oxidase (NOX) (53). CDDP induced expression of NOX3 in the cochlea, resulting in the formation of ROS, and consequently lipid peroxidation of the cochlea (10). In the present study, the levels of 4-HNE, a highly toxic lipid peroxidation product of aldehydes (54), were observed to increase as a result. High 4-HNE levels can induce the release of cytochrome c from the mitochondria by activating the mitogen-activated protein kinase/JNK cell death signal cascade (55). In the cytoplasm, cytochrome c activates caspase-9 and caspase-3, eventually leading to apoptosis of cochlear cells (56,57). Studies have shown that UA exerts significant antioxidant effects, inhibiting excessive free radical production by upregulation of endothelial nitric oxide synthase expression and downregulation of inducible nitric oxide synthase and NOX expression (58-60). In addition, UA might play an antioxidant role by enhancing free radical scavenging (61). Consistent with these studies, the present study found that UA inhibited the production of the highly toxic lipid peroxidation product 4-HNE in the cochlea, indicating that UA could inhibit CDDP-induced lipid peroxidation.

Antioxidants such as tiopronin, resveratrol, and vitamin E can alleviate inner ear damage caused by CDDP (62-64). However, most of these antioxidants have been found to attenuate the antitumor effects of CDDP, making them a poor therapeutic option for reducing CDDP-induced ototoxicity (65). Previous research has shown that UA can exert antitumor effects by promoting cancer cell apoptosis, inhibiting tumor angiogenesis and reducing cytotoxicity, and that it might have therapeutic effects on several tumors, including gastric, liver and ovarian cancers (66-68). The present study found that UA enhanced CDDP-induced tumor cell apoptosis in vitro. It was hypothesized that UA may complement cisplatin in its antitumor effect. These results suggested the potential of UA and CDDP co-treatment for the treatment of solid tumors.

In conclusion, the present study found a role in the TRPV1/Ca2+/calpain signaling pathway, which may facilitate CDDP-induced hearing loss. UA can reduce the extent of calcium overload and oxidative stress, and subsequently decrease the induction of apoptotic pathways associated with OHC loss (Fig. 7). These results suggested that UA can function as a lead compound for the pharmacological control of drug-induced ototoxicity.

Figure 7

Possible mechanism of UA preventing CDDP-induced hearing loss in mice. CDDP, cisplatin; UA, ursolic acid; TRPV1, transient receptor potential vanilloid receptor 1; 4-HNE, 4-hydroxynonenal.