ROS-Induced Activation of DNA Damage Responses Drives Senescence-Like State in Postmitotic Cochlear Cells: Implication for Hearing Preservation
In our aging society, age-related hearing loss (ARHL) has become a major socioeconomic issue. Reactive oxygen species (ROS) may be one of the main causal factors of age-related cochlear cell degeneration. We examined whether ROS-induced DNA damage response drives cochlear cell senescence and contributes to ARHL from the cellular up to the system level. Our results revealed that sublethal concentrations of hydrogen peroxide (H2O2) exposure initiated a DNA damage response illustrated by increased γH2AX and 53BP1 expression and foci formation mainly in sensory hair cells, together with increased levels of p-Chk2 and p53. Interestingly, postmitotic cochlear cells exposed to H2O2 displayed key hallmarks of senescent cells, including dramatically increased levels of p21, p38, and p-p38 expression, concomitant with decreased p19 and BubR1 expression and positive senescence-associated β-galactosidase labeling. Importantly, the synthetic superoxide dismutase/catalase mimetic EUK-207 attenuated H2O2-induced DNA damage and senescence phenotypes in cochlear cells in vitro. Furthermore, systemic administration of EUK-207 reduced age-related loss of hearing and hair cell degeneration in senescence-accelerated mouse-prone 8 (SAMP8) mice. Altogether, these findings highlight that ROS-induced DNA damage responses drive cochlear cell senescence and contribute to accelerated ARHL. EUK-207 and likely other antioxidants with similar mechanisms of action could potentially postpone cochlear aging and prevent ARHL in humans.
KeywordsOxidative stress DNA damage responses Senescence Age-related hearing loss Hearing preservation
age-related hearing loss
DNA damage–binding protein 2
DNA damage response
inner hair cells
nuclear factor erythroid 2–related factor 2
outer hair cells
checkpoint kinase 1
checkpoint kinase 2
reactive oxygen species
senescence-accelerated prone 8 mice
senescence-accelerated resistant 1 mice
Age-related hearing loss (ARHL) or presbycusis is the third most prevalent chronic disease among older adults . It can vary in severity from mild to severe, the more severe forms affecting communication and contributing to social isolation, depression, and possibly dementia . There is general agreement that cumulative effects of aging on hearing are exacerbated by genetic predisposition and environmental factors such as noise or drug exposure. With increasing noise pollution and life expectancy in our modern society, the prevalence of presbycusis is expected to grow dramatically. The precise mechanisms behind the age-related degeneration of the cochlear structures remain unclear, however. This is in part due to the complexity of each causal factor and to the interaction of the different mechanistic pathways leading to ARHL.
Growing evidence links oxidative stress to ARHL [3, 4]. The presence of lipid peroxidation , oxidative mitochondrial DNA damage, and glutathione-conjugated proteins [5, 6] is each strongly associated with cochlear aging in mice. In addition, reduced expression of antioxidant enzymes such as catalase, mitochondrial manganese superoxide dismutase (MnSOD or SOD2), and cytosolic SOD1 has been observed in older mice [4–6]. Furthermore, mice lacking the antioxidant enzyme SOD1 display enhanced age-related cochlear hair cell loss, reduced thickness of the stria vascularis, and severe degeneration of spiral ganglion neurons (SGNs) . On the other hand, an increase in SOD2 expression gradient in ganglion cells has been reported along the basal to apical axis of rodent and primate cochlea , consistent with the differential vulnerability of the varying areas in the cochlea, from the most vulnerable high-frequency to least vulnerable low-frequency regions in most ARHL. Finally, low serum levels of the ROS scavenger melatonin are significantly associated with the occurrence of high-frequency hearing loss in the elderly . Taken together, this experimental and clinical evidence indicates the potential involvement of oxidative stress in ARHL.
It is generally accepted that oxidative stress may cause irreversible DNA damage . Over 100 oxidative modifications to DNA have been identified, including adducts and both single-strand and double-strand breaks . In the cochlea, we have recently reported the activation of the DNA damage response (DDR) in postmitotic inner ear cells following treatment with the ototoxic anticancer drug cisplatin . However, the link between increased ROS, DNA damage, and cochlear cell aging remains obscure. Understanding ROS-induced cochlear cell degeneration may lead to identifying key molecular targets to protect residual hearing in patients suffering ARHL.
The synthetic superoxide dismutase (SOD)/catalase mimetic EUK-207 is a member of the class of metal-containing salen–manganese (Mn) complexes. EUK-207 suppresses oxidative stress, including in the mitochondria [13, 14], through transforming oxygen free radicals into oxygen via a superoxide dismutase-like activity, and hydrogen peroxide into oxygen and water by a catalase-like activity . EUK-207 has been used successfully to suppress age-related [15, 16] and radiation-induced cognitive impairment in mice .
Here, we explore the hypothesis that ROS-induced DDR drive cochlear cell senescence and contribute to ARHL from the cellular to the system level. In our in vitro study, hydrogen peroxide (H2O2) exposure caused DNA damage and initiated a DNA damage response in cochlear cells, as demonstrated by activation of the Chk2–p53 pathway. In addition, the H2O2-exposed cochlear cells exhibited several additional senescence-associated properties, including high senescence-associated β-galactosidase activity mainly in sensory hair cells, together with massively increased levels of p21, p38, and p-p38. Our in vivo results from senescence-accelerated mouse-prone 8 (SAMP8) mice, the well-characterized model of premature cochlear aging , demonstrated a similar accumulation of high amounts of DNA damage and similar senescence-like features in the cochlear tissues during aging. Importantly, EUK-207 attenuated H2O2-induced DNA damage and senescence phenotype in cochlear cells in vitro and reduced age-related loss of both hearing and hair cells in SAMP8 mice. These results highlight the attractive potential of antioxidants as a tool with which ARHL could be slowed down.
Material and Methods
Neonate Swiss mice were purchased from Janvier Laboratories. The senescence-accelerated mouse strain is derived from AKR/J mice by continuous brother–sister mating selected for a phenotype toward either accelerated senescence (SAMP) or resistance senescence (SAMR) . Among these strains, SAMP8 mice have strain-specific characteristics of rapid aging . Here, SAMP8 and senescence-accelerated mouse-resistant 1 (SAMR1) of both sexes were purchased from Harlan SARL Laboratory. Mice were housed in pathogen-free animal care facilities accredited by the French Ministry of Agriculture and Forestry (C-34-172-36; December 19, 2014).
H2O2 was purchased from Acros Organics (#202460250). The salen–manganese (Mn) SOD-catalase mimetic, EUK-207, was a gift from Dr. Bernard Malfroy-Camine, of MindSet Rx, Arlington, MA, USA.
For in vitro experiments, H2O2 was freshly prepared in culture medium to final concentrations ranging from 0 to 1.25 mM. The EUK-207 was prepared at 10 mM in pure water and freshly diluted in culture medium to a final concentration of 10 μM. This final concentration was chosen based on our preliminary evaluations of the dose–response effects of EUK-207, at concentrations ranging from 0 to 30 μM, on H2O2-induced γH2AX expression on day 3 in H2O2-exposed cochlear explants. For in vivo experiments, EUK-207 was freshly prepared at 1.5 mM in 5% mannitol and was administered through a subcutaneously implanted Alzet micro-osmotic pump (DURECT Corporation, 0.11 μl per hour, 28 days #1004) at a dose of 0.2 mg/kg/day.
In Vitro Assessments
Organ of Corti and Whole Cochlea Cultures
Mouse whole cochleae and organ of Corti explants were collected from postnatal day 3 mice and prepared according to the procedures described previously [12, 20]. The whole cochleae were kept in suspension, and the organ of Corti explants in adherent conditions in a 6-well culture plate containing 2 ml/well of culture medium. Culture medium consisted of Dulbecco’s modified Eagle’s medium/nutrient mixture F-12 (DMEM/F-12, #21331020) containing 2 mM L-glutamine (#25030024), N-2 complement (#17502048) at 1× and insulin transferrin selenium (#41400045) at 1× purchased from Gibco Life technologies, and 8.25 mM D-glucose (#G6152) and 30 U/ml penicillin G (#P3032) from Sigma-Aldrich.
H2O2-Induced Oxidative Stress
The oxidizing agent used to induce oxidative stress in vitro (H2O2) was chosen based on its wide use in a variety of cell types [21, 22]. Here, cochlear samples were firstly exposed to culture medium alone for 24 h in a humidified incubator (37 °C, 5% CO2). To probe the dose effect of H2O2, the culture medium was then replaced with fresh medium containing H2O2 at various concentrations (0, 0.25, 0.4, 0.5, 0.75, 1, and 1.25 mM) for 5 h before being replaced again with the culture medium alone and maintained for 5 days. The culture medium was renewed once on day 2 or 3 following its replacement. To determine time-dependent cochlear cell degeneration, organ of Corti explants were exposed to medium containing H2O2 at its half maximal effective concentration (EC50) and also at the concentration just below the EC50 (0.5 mM) for 5 h and then maintained in the culture medium alone for 0, 1, 3, and 5 days. All control samples were maintained in culture medium alone and were run concurrently alongside the experimental cultures.
Dose-Dependent Cytotoxicity and Apoptosis
To evaluate the toxicity of H2O2, cultured organ of Corti explants were immunolabeled with a rabbit polyclonal antibody against myosin 7A (1/300, Proteus Biosciences Inc. #25-6790) and a mouse monoclonal antibody against neurofilament (NF 200, 1/600, Sigma-Aldrich #N0142) to label hair cells and spiral ganglion neurons, respectively. Hair cell counting (six cochleae per condition and per time point) was performed using standard techniques . Due to the resistance of apical turn hair cells to H2O2 cytotoxicity, counting concerned only a 1.5-mm length of the cochlear duct at the basal turn (4 to 5.5 mm from the apex, see ).
The TUNEL kit (DeadEnd™ fluorometric TUNEL System, Promega #G3250) was used to identify apoptotic DNA fragmentation. The cochlear samples were counterstained with a rabbit polyclonal antibody against myosin 7A and a mouse monoclonal antibody against neurofilament. All secondary antibodies were used at a dilution of 1/1000. This included donkey anti-mouse and anti-rabbit IgG conjugated to Alexa 488 or Alexa 568 (Molecular Probes #A-21202, #A-21206, #A-10037, #A-10042). Fluorescent tags were visualized using a confocal microscope (LSM 5 Live Duo, Zeiss). In control specimens without primary antibodies, neither Alexa 488 nor 568 fluorescent tags were observed. All experiments were performed in triplicate.
Measurement of Oxidative Stress
Oxidative stress was studied in the whole cochlea homogenates as previously described . Catalase and SOD activities were measured as previously described by Marklund . Lipid peroxidation was assessed using the thiobarbituric acid-reactive substances method and was expressed in nanomoles per milligram malondialdehyde (MDA) . Protein concentrations were measured using the BCA protein assay kit (Pierce #23250). All experiments were performed in triplicate.
Cellular Localization of DNA Damage Foci and Foci Counting
Immunocytochemistry was employed to localize DNA damage foci in cultured organ of Corti using mouse monoclonal antibody against phospho-H2AX (1/500, Ser139, Merck Millipore #05-636) or rabbit polyclonal antibody against 53BP1 (1/100, Novus Biologicals #NB100-305). The cochlear samples were counterstained with a rabbit polyclonal antibody against myosin 7A or mouse monoclonal antibody against parvalbumin (1/500, SWANT #PV235) to label hair cells. All secondary antibodies were used at a dilution of 1/1000. The Hoechst 33342 dye (0.002% wt:vol in PBS 1×, Thermo Fisher Scientific #62249) was used to stain DNA. Fluorescent tags were visualized using a confocal microscope (LSM 5 Live Duo, Zeiss). No fluorescent signal was detected in the control specimens without primary antibodies.
The number of foci (i.e., γH2AX or 53BP1) per nucleus was computed using algorithms with Matlab custom-made software (MathWorks Company) that allow 3D rendering and visualization of “isosurfaces” enveloping all pixel clusters with intensities greater than a user-defined criterion value in each corresponding image channel (for more details, see ). All experiments were performed in triplicate.
Expression of Apoptosis, Oxidative Stress, and Autophagy-Related Proteins, DNA Damage Response, and Senescence Markers
Evaluation of apoptosis, oxidative stress, autophagy, DNA damage responses, and cellular senescence-like state in the cochlear protein extracts from H2O2-exposed and cultured whole cochleae was performed using Western blotting technique as described previously . Antibodies used included those recognizing Bax (1/1000, Abcam #7977), BCl2 (1/1000, Santa Cruz #sc-492), cleaved caspase 3 (1/1000, Asp175 Cell Signaling #9661), SOD2 (1/1000, Abcam #ab13533), catalase (1/1000, Abcam #16731), p66Shc (1/1000, Abcam #ab33770), phospho- p66Shc (S36, 1/1000, Abcam #54518), p-Beclin 1 Ser93 (1/1000, Cell Signaling #14717), LC3-II (1/800, Cell Signaling #2775), Rab7 (1/800, Santa Cruz Biotechnology #sc-376,362), p62 (1/1000, MBL International #PM045), phospho-H2AX (1/1500, Ser139, Cell Signaling #2577), 53BP1 (1/1000, Novus Biologicals #100-305), DDB2 (1/800, Invitrogen #PA5-37361), phospho-Chk1 (1/1000, Ser345, Cell Signaling #2348), phospho-Chk2 (1/1000, Thr68, Cell Signaling #2661), p53 (1/1500, Cell Signaling #2524), p21 (1/2000, Cell Signaling #2946), p38 (1/500, Elabscience #E-AB-32459), p-p38 (Thr180/Tyr182, 1/1000, Cell Signaling #9211), p16 (1/1000, BD Pharmingen #551154), p19 (1/1000, Abcam #ab80), BubR1 (1/1000, Abcam #ab183496), FOXO3a (1/1000, Cell Signaling #2497), and nuclear factor erythroid 2–related factor 2 (Nrf2) (1/1000, Santa Cruz # sc-365949). β-Actin (1/10,000, Sigma-Aldrich #A1978) served as a loading control. Secondary antibodies used were horseradish peroxidase-conjugated goat anti-mouse IgG antibodies (1/3000, Jackson ImmunoResearch #115-001-003) or goat anti-rabbit IgG antibodies (1/3000, Jackson ImmunoResearch #111-001-003). All experiments were performed in triplicate. Image scans of Western blots were used for semiquantitative analysis.
Senescence-Associated Beta-Galactosidase Activity Assay
Senescence-associated beta-galactosidase (SA-β-gal) activity was measured according to the manufacturer’s protocol (Cell Signaling #9860). Briefly, cultured cochlear explants and cochleae from 6-month-old SAMR1 and SAMP8 mice were fixed in 4% paraformaldehyde in phosphate-buffered saline 1× (PBS) for 30 min at room temperature and washed twice with PBS. The fixed SAMR1 and SAMP8 cochleae were then processed for either cryosectioning or surface preparation. All cochlear samples were then incubated overnight at 37 °C in a dry incubator with fresh β-galactosidase staining solution at pH 6.0. Cochlear explants were observed using light microscopy, while cryosections and surface preparations were scanned using a NanoZoomer (Hamamatsu Photonics, Hamamatsu City, Japan) with a ×40 objective.
Pharmacological Mitigation of ROS
To probe the impact of ROS mitigation on H2O2-induced DNA damage and senescence-like phenotype, whole cochleae were exposed to either culture medium alone or medium containing 10 μM EUK-207 for 24 h in a humidified incubator (37 °C, 5% CO2). The culture medium was then replaced with fresh medium containing H2O2 at a concentration of 0.5 mM for 5 h. The cochlear samples were further grown for three additional days in culture medium either alone or with EUK-207. All control samples were maintained in culture medium and were run concurrently with experimental cultures. The samples were then collected for the evaluation of oxidative stress, DNA damage, and cellular senescence-like state using Western blotting and immunolabeling techniques.
Validation in Adult SAMP8 In Vivo
Our previous study demonstrated that SAMP8 mice displayed premature ARHL and early onset cochlear cell degeneration when compared with the normally aging control senescence-accelerated mouse-resistant 1 (SAMR1) mice .
The auditory function was assessed by recording auditory brainstem responses (ABRs) in SAMP8 and SAMR1 mice at the age of 1, 3, 6, and 12 months (n = 14 animals per age and per strain). ABRs reflect the synchronous activation of auditory neurons from the cochleae up to the colliculi in response to incoming sound. The recordings were carried out under anesthesia with Rompun 2% (3 mg/kg) and Zoletil 50 (40 mg/kg) in a Faraday-shielded anechoic soundproof cage. Rectal temperature was measured with a thermistor probe and maintained at 38.5 ± 1 °C using a heated under blanket. ABRs were recorded from three subcutaneous needle electrodes placed at the vertex (active) and on the pinna of the tested ear, as well as in the neck muscles (ground) of the mice. The acoustic stimuli generated by a NI PXI-4461 signal generator (National Instruments) consisted of 10 ms tone bursts with a 1-ms rise and fall time delivered at a rate of 10/s. Sound was delivered by a JBL 075 loudspeaker (James B. Lansing Sound) positioned at 10 cm from the tested ear, under calibrated free-field conditions. Cochlear amplification (20,000) was achieved via a Grass P511 differential amplifier, averaged 1000 times (Dell Dimensions). Amplitude-intensity functions of the ABRs were obtained at each frequency tested (2, 4, 6.3, 8, 10, 12.5, 16, 20, 25, and 32 kHz) by varying the intensity of the tone bursts from 0 to 100 dB SPL, in 5 dB incremental steps. The ABR thresholds were defined as the minimum sound intensity necessary to elicit a well-defined and reproducible wave II. Recordings and analysis were performed blindly.
Evaluation of oxidative stress, DNA damage responses, and cellular senescence-like state in the cochlear protein extracts from the cochleae of SAMP8 and SAMR1 mice aged 1, 6, and 12 months (n = 16 cochleae per age and per strain) was performed using Western blotting technique as described previously . Antibodies used included those recognizing Nrf2 (1/1000, Santa Cruz #sc-365949), SOD2 (1/1000, Abcam #ab13533), p66Shc (1/1000, Abcam #ab33770), phospho-p66Shc (S36, 1/1000, Abcam #54518), phospho-Chk2 (1/1000, Thr68, Cell Signaling #2661), p53 (1/1500, Cell Signaling #2524), phospho-p53 (1/1500, Ser15, Cell Signaling #9289), p21 (1/2000, Cell Signaling #2946), p16 (1/1000, BD Pharmingen #551154), BubR1 (1/1000, Abcam #ab183496), and p19 (1/1000, Abcam #ab80). β-Actin (1/10,000, Sigma-Aldrich #A1978) served as a loading control. Secondary antibodies used were horseradish peroxidase-conjugated goat anti-mouse IgG antibodies (1/3000, Jackson ImmunoResearch #115-001-003) or goat anti-rabbit IgG antibodies (1/3000, Jackson ImmunoResearch #111-001-003). All experiments were performed in triplicate. Image scans of Western blots were used for semiquantitative analysis.
Age-Related Hearing Impairment and Molecular Correlation
To compare the time course of the hearing impairments found in SAMP8 and SAMR1 mice, the mean ABR threshold evoked by the range of tested frequencies from 2 to 32 kHz for each strain and at each time point was calculated. A linear regression was then performed to determine the threshold elevation per month. To decipher the molecular determinants of the accelerated age-related hearing impairment in SAMP8 mice, we used the mean values of ABR thresholds and protein levels in SAMR1 as a reference and subtracted these from the values for SAMP8 mice. The differences in threshold and protein level (SAMP8 − mean SAMR1) were then averaged and used for statistical analysis and to calculate SEM.
Pharmacological Counteraction of ROS
Finally, the impact of ROS mitigation on premature ARHL was also investigated in SAMP8 mice. The 6-month-old male and female SAMP8 mice were randomly divided into two groups: (i) control 5% mannitol (vehicle, n = 10) and (ii) EUK-207 (n = 10). The EUK-207 was dissolved in 5% mannitol to a final concentration of 1.5 mM, filter-sterilized, and administered at a dose of 0.2 mg/kg/day with the use of a subcutaneous Alzet micro-osmotic pump, continuing for up to 84 days. This dose of EUK-207 was selected according to the effective EUK-207 dose range reported previously . Control mice received pumps filled with 5% mannitol. Mice received new pumps every month.
The ABRs were recorded prior to pump implantation and 2 and 3 months afterwards. Sensory hair cell loss was evaluated using scanning electron microscopy (Hitachi S4000). The cochleae from the different groups (n = 7 per group) were processed and evaluated using previously reported standard techniques [25, 26]. Hair cell counting was performed in four different 300-μm-long segments of the organ of Corti, centered at 1.1, 2.6, 3.5, or 4.1 mm from the cochlear apical end and corresponding to the frequencies of 8, 16, 25, and 32 kHz, respectively .
Data are expressed as mean ± SEM. Normality of the variables was tested by the Shapiro–Wilks test. The significance of the group differences was assessed with a one-way ANOVA; once the significance of the group differences (P < 0.05) was established, Tukey’s post hoc tests were subsequently used for pairwise comparisons. Significance of linear regressions was determined using the Pearson correlation coefficient (P < 0.001). For in vivo mouse studies, based on data from our previous reports [28, 29] or from preliminary experiments, we calculated the sample size using G*Power 220.127.116.11 to ensure adequate power of key experiments for detecting prespecified effect sizes.
In Vitro Studies
To probe a direct causal relationship between increased ROS and premature occurrence of cochlear cell senescence-like phenotype, we adapted an available in vitro model of oxidative stress [30, 31], which consists of exposing cochlear explants or whole cochleae to a series of H2O2 concentrations in culture.
H2O2 Induces Dose-Dependent Outer Hair Cell Apoptosis
To determine the nature of the H2O2-induced hair cell loss, we probed for apoptotic markers at 3 days after H2O2 exposure. Large amounts of OHCs and some IHCs displayed TUNEL-positive nuclei (Fig. 1i, j) after incubation with the higher H2O2 concentration (≥ 0.75 mM), when compared with controls or the sublethal concentration (0.5 mM, Fig. 1g, h). Concomitantly, Western blots from protein extracts of H2O2-exposed cochleae showed significant caspase 3 activation in the cochleae exposed to ≥ 0.5 mM of H2O2 (Fig. 1k, l) (0.5 mM of H2O2: F1:6 = 24.8, P = 0.003; 0.75 mM: F1:6 = 986, P < 0.001). In addition, higher levels of Bax protein were observed in those exposed to ≥ 0.75 mM of H2O2 (Fig. 1k, l) (F1:6 = 53.6, P < 0.001). In contrast, Bcl-2 protein levels remained unaltered (Fig. 1k). Taken together, these data demonstrate that H2O2 induces a concentration-dependent hair cell loss with massive OHC apoptosis after exposure to high concentrations (≥ 0.75 mM). In an in vivo situation, ROS levels are presumed very low due to their short half-life and the existence of various antioxidant enzymes . On this basis, we chose H2O2 concentrations ≤ 0.5 mM for subsequent experiments.
Increased Oxidative Stress and Upregulation of Autophagy
Considering the link between oxidative stress and autophagy , we examined autophagy induction by monitoring the levels of proteins involved in autophagosome formation and autophagic flux, the dynamic process of autophagosome synthesis, and degradation. H2O2-exposed cochleae displayed significantly increased levels of phosphorylated Beclin 1, a major player in the autophagic initiation process  (0.4 mM: F1:6 = 24.7, P = 0.003; 0.5 mM: F1:6 = 134.1, P < 0.001); LC3-II (0.4 mM: F1:6 = 10.3, P = 0.018; 0.5 mM: F1:6 = 53.7, P < 0.001), a hallmark of autophagosome formation; and Rab7, a small GTP-binding protein that plays a role in maturation of late autophagic vacuoles  (0.4 mM: F1:6 = 57.4, P < 0.001; 0.5 mM: F1:6 = 10.2, P = 0.029) (Fig. 2g, h). Identically, a significant increase of p62, a ubiquitin-binding scaffold protein and a subtract of autophagy, was observed in 0.5 mM H2O2-exposed cochleae (F1:6 = 255, P < 0.001) (Fig. 2g, h). Taken together, these results indicate that H2O2 induced cochlear cell oxidative stress and triggered an autophagic response.
DNA Damage Responses
DNA double-strand breaks are potent inducers of DDR, characterized by activation of DDR proteins, including the cell cycle checkpoint proteins like checkpoint kinase 1 (p-Chk1), checkpoint kinase 2 (p-Chk1), and p53 . Once activated, p53 plays an important role in modulating distinct cell fate decisions [40, 41, 42]. It enhances DNA repair pathways through the upregulation of several repair proteins like the damage-specific DNA binding protein 2 (DDB2), which is required for nucleotide excision repair . We found significant increases in the levels of DDB2 (0.4 mM of H2O2: F1:6 = 6.0, P = 0.049; 0.5 mM: F1:6 = 12.1, P = 0.013), p-Chk2 (0.5 mM: F1:6 = 30, P = 0.002), and p53 (0.4 mM: F1:6 = 9.2, P = 0.023; 0.5 mM: F1:6 = 25.8, P = 0.002) in H2O2-exposed cochleae when compared with the control condition (Fig. 3i, k). In contrast, the increase in the level of p-Chk1 after H2O2 challenge did not reach significance (Fig. 3i, k). These results indicate that H2O2 exposure induces DNA damage and the activation of the Chk2–p53 pathway in postmitotic cochlear cells.
Senescence-Like Phenotype in Postmitotic Cochlear Cells
Pharmacological Mitigation of ROS with EUK-207
In Vivo Validation in Adult SAMP8 Mice
Molecular Pathway Validation
As expected, SAMP8 mice displayed a significant increase in phosphorylated Beclin 1 (6 months: F1:6 = 114.8, P < 0.0001; 12 months: F1:6 = 78, P = 0.006) and Rab7 (6 months: F1:6 = 82, P = 0.0001; 12 months: F1:6 = 85.3, P = 0.0001) from 6 months of age. In contrast, SAMR1 mice only showed such increases (p-Beclin 1: F1:6 = 167, P < 0.0001; Rab7: F1:6 = 26.3, P = 0.0022) at 12 months of age. SAMP8 mice also showed significantly higher levels of phosphorylated Beclin 1 (6 months: F1:6 = 108, P < 0.0001; 12 months: F1:6 = 73.3, P = 0.0001) and Rab7 (6 months: F1:6 = 76, P = 0.0001; 12 months: F1:6 = 21.2, P = 0.0036) than SAMR1 mouse strains at the same age (Fig. 6d, e). Not surprisingly, a significant higher level of p62 was observed in SAMP8 mice at 6 months compared with SAMR1 of the same age (F1:6 = 9.48, P = 0.022). While p62 levels maintained in SAMP8 mice during aging, these levels decreased significantly in SAMR1 mice at 6 months when compared with 1 month (F1:6 = 8.98, P = 0.0241). Taken together, these results indicate an increased autophagy with impaired autophagic flux in SAMP8 mice.
To further explore the potential appearance of DDR in SAMP8 mice, we assessed DDR proteins. Western blot analysis demonstrated dramatic increases in p-Chk2 levels from as early as 6 months in SAMP8 mouse cochleae (6 months: F1:6 = 161.9, P < 0.001; 12 months: F1:6 = 292, P < 0.001), while such an increase was only seen in 12-month-old SAMR1 mice (F1:6 = 209.5, P < 0.001) compared to 1-month olds (Fig. 6f, g). In addition, significantly higher levels of p-Chk2 were observed in SAMP8 than SAMR1 mice of the same age (6 months: F1:6 = 151.4, P < 0.001; 12 months: F1:6 = 199, P < 0.001) (Fig. 6f, g). Similarly, SAMP8 mice showed significantly increased p53 (6 months: F1:6 = 11.2, P = 0.029; 12 months: F1:6 = 358.6, P < 0.001) and p-p53 levels (6 months: F1:6 = 11.1, P = 0.029; 12 months: F1:6 = 129.8, P < 0.001) from 6 months (Fig. 6f, g). The same phenomenon occurred in SAMR1 but at a later stage (at 12 months, p53: F1:6 = 10.9, P = 0.029; p-p53: F1:6 = 70.7, P < 0.001, Fig. 6f, g). However, changes were more pronounced in SAMP8 than in SAMR1 mice (p53: 6 months: F1:6 = 11, P = 0. 029, 12 months: F1:6 = 181.6, P < 0.001; p-p53: 6 months: F1:6 = 11.2, P = 0.029, 12 months: F1:6 = 84.1.2, P < 0.001, Fig. 6f, g). Together, these results point toward the activation of a DNA damage response during cochlear aging, with more severe effects in SAMP8 compared to SAMR1 mice.
Correlation Between Molecular Events and Hearing Loss
The SAMP8 strain displayed a deficit in antioxidant proteins Nfr2 and SOD2 and an excess of the pro-oxidant molecule p66Shc at all ages. The levels of pro-oxidant molecules p66Shc and p-p66Shc increased concomitantly with the degradation of hearing in SAMP8, while the deficit in antioxidants remained relatively stable until 6 months of age (Fig. 8d). The differences between the strains in terms of levels of the antioxidant molecules Nrf2 and SOD2 decreased at 12 months of age (Fig. 8d), probably due to the age-related repression of these molecules in both strains. While both strains displayed comparable levels of DDR proteins (p-Ch2, p53, p-p53) at 1 month, only SAMP8 mice thereafter showed a dramatic increase in levels to reach a plateau at 6 months (Fig. 8e). Finally, prosenescence proteins (p21, p16) increased continually with age in SAMP8 mice (Fig. 8f). By contrast, the mitotic checkpoint protein BubR1 and the cell cycle regulator p19 remained close to zero, except at 6 months where p19 had increased before falling back to zero at 12 months with a net result of no significant difference between strains at 12 months (Fig. 8f).
Collectively, these in vivo data support our in vitro results indicating that an early increase in the level of ROS triggers the activation of DNA damage responses and premature occurrence of senescence-like state in postmitotic cochlear cells and an accelerated hearing loss in SAMP8 mice.
EUK-207 Slows Down Age-Related Hearing Loss in SAMP8 Mice
ARHL is the third most prevalent chronic medical condition affecting older adults . Although the etiology of ARHL remains unclear, oxidative stress is proposed as one of the most important risk factors [4, 6]. Consistent with the results obtained from other cell types [30, 31], cochlear sensory hair cells displayed dose-dependent cytotoxic effects in cochlear explants exposed to H2O2. In order to examine the potential ROS-related mechanisms leading to cochlear aging, we set the level of ROS to resemble as closely as possible the in vivo conditions. Even at such low concentrations, H2O2 led to oxidative stress–related events in cochlear tissues 3 days after exposure. Considering the common link between oxidative stress and autophagy , we assessed autophagy induction by monitoring autophagosome formation and autophagic flux. As suspected, H2O2-induced oxidative stress initiated autophagy, as illustrated by the activation or increased expression of several autophagy-related proteins, such as Beclin 1, LC3-II, Rab7, and p62. The increased level of the autophagic substrate p62 in particular indicated an impaired autophagic flux . Collectively, these results demonstrate that at low concentrations (below EC50), H2O2 can induce oxidative stress and trigger an impaired autophagic response, and they thereby validate our in vitro model of oxidative stress.
Unlike the necrosis or apoptosis observed in cells subjected to cytotoxic doses of H2O2, we have here discovered that exposure to low concentrations of H2O2 slightly alters cochlear cell viability within 3 days after exposure but importantly induces DNA damage, mainly in sensory hair cells. These results are consistent with our previous study in which genotoxic stress was induced by cisplatin, a DNA-damaging anticancer drug . In the present study, H2O2 exposure induced DNA damage responses within cochlear cells, as shown by significant increases in DDB2, a protein involved in nucleotide excision repair , p-Chk2, and p53. Depending on the severity of the DNA damage, the cell microenvironment, and the cell type, activated p53 can orchestrate different cellular outcomes such as repair of DNA damage, apoptosis, or senescence .
DNA damage and DDR are known to be involved in the induction of replicative senescence and DNA damaging agent-induced premature senescence in cycling cells [51, 52]. Here, we have demonstrated for the first time that the ROS-induced DDR coincided with the premature occurrence of senescence-like features in postmitotic cochlear cells, as shown by increased levels of p21, a substrate of p53, and a necessary signal transducer between DDR and senescence in neurons . Consistent with the increase in p21, H2O2-exposed cochleae displayed high levels of p38 and phospho-p38. These latter two proteins can be activated by p21 and may be implicated in senescence-associated secretory phenotypes . Together with increased activity of SA-β-gal following H2O2 exposure, our results indicate that the activation of the p53–p21 pathway might mediate the low concentration H2O2-induced senescence-like phenotype in postmitotic cochlear cells in vitro. To shed more light on these postmitotic cochlear cell senescence-like phenotypes, we quantified the levels of p16, named also p16INK4A, which can be induced independently of DDR and p53. Above a certain threshold, p16 blocks G1–S progression within cycling cells  and may act as an effector of senescence in cultured cells . However, our H2O2-exposed cochleae exhibited no significantly increased levels of p16 arguing against the engagement of the p16 pathway in our in vitro conditions.
We also assessed the level of the mitotic checkpoint protein BubR1, the level of which decreases with age in various mouse tissues . BubR1 insufficiency is associated with the occurrence of multiple progeroid and aging-associated phenotypes in mice  and in children with mosaic variegated aneuploidy syndrome . Consistent with these reports, we found a significantly decreased expression of BubR1 after H2O2 challenge suggesting its implication in ROS-induced premature cochlear cell senescence-like phenotypes. We also found a significantly decreased expression of the cell cycle regulator p19 in H2O2-exposed cochleae. This protein likely plays a role in the attenuation of senescence and aging in BubR1-insufficient mice , yet the precise relationship between its decreased level and increased cochlear cell senescence-like phenotypes after ROS challenge requires further in-depth investigation.
To probe the impact of ROS mitigation on H2O2-induced DNA damage and senescence-like phenotypes, we tested a synthetic SOD/catalase mimetic, EUK-207. This molecule possesses both SOD and catalase activity and scavenges ROS generated in the cytoplasm or organelles, including the mitochondria . In addition, EUK-207 suppresses oxidative modifications of proteins, lipids, and nucleic acids . Here, we found that cochleae treated with EUK-207, alone or in combination with H2O2, displayed increased levels of FOXO3a and Nrf2. These two transcription factors have been shown to positively regulate cellular resistance to oxidative stress [47, 62]. Increasing their levels of expression would be expected therefore to account for the observed protective effect of EUK-207 against H2O2-induced DNA damage and senescence-like phenotype in cochlear cells in culture.
To extend these in vitro data to the system level, we used SAMP8 mice. This strain of mouse develops a premature ARHL and early onset cochlear cell degeneration . Here, we showed that, compared to control, SAMP8 cochleae displayed an early increased oxidative stress (1 month of age), highlighted by significantly lower levels of both Nrf2, a main regulator of cellular resistance to oxidants, and the antioxidant enzyme SOD2, concomitant with a large increase in the levels of p66Shc and its phosphorylated form. These two latter proteins are involved in downregulating the synthesis of antioxidant enzymes and the generation of ROS in the mitochondria . These in vivo data indicate that the events related to oxidative stress occur early (1 month of age), before the onset of hearing loss in SAMP8 mice (3 months) . Here, we also found significantly higher levels of phosphorylated Beclin 1, Rab7, and p62 in SAMP8 mice during aging compared with SAMR1. Together with our previous study demonstrating the upregulation of LC3-II with concomitant accumulation of lipofuscin in SAMP8 mouse cochleae, we suggest an increased autophagy with impaired and/or insufficient capacity for autophagy flux in SAMP8 mice. These results are consistent with some recent studies showing that autophagy plays an important role in maintaining adult hearing in response to auditory stress or during aging. Indeed, activation of autophagy by rapamycin diminished noise-  or cisplatin-induced  hair cell and hearing loss. Moreover, the weaker autophagy gene expression in 1-year-old cochleae of Igf1−/− mice coincides with age-related hearing loss [66, 67].
Consistent with our in vitro results, the SAMP8 mice displayed activation of the Chk2–p53 pathway from 6 months of age. From the same age, they also displayed similar senescence signatures, characterized by decreased levels of BubR1 and increased levels of p21 and p16, together with increased activity of SA-β-gal in the region of spiral ganglion neurons, OHCs, and IHCs of SAMP8 mice. These discoveries are consistent with recent studies in which activation of DNA damage responses and the appearance of certain cell senescence markers were implicated in postmitotic neuron senescence [68, 69]. Based on the significant correlation between early increased oxidative stress together with later accumulation of DDR and senescence-like features, and severe hearing loss in SAMP8 mice, we suggest that oxidative stress may be one of the main culprits behind age-related cochlear cell degeneration and auditory function decline.
Having established the causal relationship between ROS and age-related cochlear cell degeneration, we set out to test the efficiency of systemically administering EUK-207 to mitigate the pathological effects of uncontrolled ROS production on hearing function. Our results reveal that systemic treatment with EUK-207 effectively prevented or slowed down ARHL in SAMP8 mice. These results are consistent with previous studies showing successful suppression of age-related cognitive impairment in mice with EUK-207 [15, 16]. They also represent a proof-of-concept that pharmacological scavenging of superoxide and hydrogen peroxide can mitigate ARHL in SAMP8 mice. They thereby provide a strong rationale for the clinical development of EUK-207 for use in the prevention or slowing down of ARHL in humans.
The in vitro and in vivo data presented here provide evidence that (i) ROS is one of the main culprits behind age-related sensory hair cell degeneration, (ii) ROS-induced DDR driving senescence-like features may account for premature cochlear aging, and (iii) pharmacological scavenging of superoxide and hydrogen peroxide can mitigate ARHL in SAMP8 mice.
The authors would like to thank Dr. Jérôme Bourien and Adrien Caplot for statistical analyses and Dr. Alexandre Pattyn, Dr. Régis Nouvian, and Dr. Patrick Carroll for helpful discussions. All confocal and electron-microscopic analyses were performed at the Montpellier RIO Imaging-INM core facility. The manuscript has been revised by an independent scientific English language editing service (Angloscribe).
This work was supported by the Fondation de l’Avenir (Et2-675) and the Fondation Gueules Cassées.
Compliance with Ethical Standards
Experiments were carried out in accordance with the animal welfare guidelines 2010/63/EC of the European Communities Council Directive and French Ethical Committee (agreements 00447.01), regarding the care and use of animals for experimental procedures.
Conflict of Interest
The authors declare that they have no conflicts of interest.
- 1.Lethbridge-Cejku M, Schiller JS, Bernadel L (2004) Summary health statistics for U.S. adults: National Health Interview Survey, 2002. Vital Health Stat Series 10, Data from the National Health Survey (222):1–151.Google Scholar
- 4.Menardo J, Tang Y, Ladrech S, Lenoir M, Casas F, Michel C, Bourien J, Ruel J et al (2012) Oxidative stress, inflammation, and autophagic stress as the key mechanisms of premature age-related hearing loss in SAMP8 mouse cochlea. Antioxid Redox Signal 16(3):263–274. https://doi.org/10.1089/ars.2011.4037 CrossRefPubMedGoogle Scholar
- 5.Jiang H, Talaska AE, Schacht J, Sha SH (2007) Oxidative imbalance in the aging inner ear. Neurobiol Aging 28(10):1605–1612. https://doi.org/10.1016/j.neurobiolaging.2006.06.025 CrossRefPubMedGoogle Scholar
- 6.Someya S, Xu J, Kondo K, Ding D, Salvi RJ, Yamasoba T, Rabinovitch PS, Weindruch R et al (2009) Age-related hearing loss in C57BL/6J mice is mediated by Bak-dependent mitochondrial apoptosis. Proc Natl Acad Sci U S A 106(46):19432–19,437. https://doi.org/10.1073/pnas.0908786106 CrossRefPubMedPubMedCentralGoogle Scholar
- 14.Liesa M, Luptak I, Qin F, Hyde BB, Sahin E, Siwik DA, Zhu Z, Pimentel DR et al (2011) Mitochondrial transporter ATP binding cassette mitochondrial erythroid is a novel gene required for cardiac recovery after ischemia/reperfusion. Circulation 124(7):806–813. https://doi.org/10.1161/CIRCULATIONAHA.110.003418 CrossRefPubMedPubMedCentralGoogle Scholar
- 15.Clausen A, Doctrow S, Baudry M (2010) Prevention of cognitive deficits and brain oxidative stress with superoxide dismutase/catalase mimetics in aged mice. Neurobiol Aging 31(3):425–433. https://doi.org/10.1016/j.neurobiolaging.2008.05.009 CrossRefPubMedGoogle Scholar
- 16.Liu R, Liu IY, Bi X, Thompson RF, Doctrow SR, Malfroy B, Baudry M (2003) Reversal of age-related learning deficits and brain oxidative stress in mice with superoxide dismutase/catalase mimetics. Proc Natl Acad Sci U S A 100(14):8526–8531. https://doi.org/10.1073/pnas.1332809100 CrossRefPubMedPubMedCentralGoogle Scholar
- 48.Dinkova-Kostova AT, Abramov AY (2015) The emerging role of Nrf2 in mitochondrial function. Free Radic Biol Med 88(Pt B):179–188. https://doi.org/10.1016/j.freeradbiomed.2015.04.036 CrossRefPubMedPubMedCentralGoogle Scholar
- 59.Matsuura S, Matsumoto Y, Morishima K, Izumi H, Matsumoto H, Ito E, Tsutsui K, Kobayashi J et al (2006) Monoallelic BUB1B mutations and defective mitotic-spindle checkpoint in seven families with premature chromatid separation (PCS) syndrome. Am J Med Genet A 140(4):358–367. https://doi.org/10.1002/ajmg.a.31069 CrossRefPubMedGoogle Scholar
- 60.Baker DJ, Perez-Terzic C, Jin F, Pitel KS, Niederlander NJ, Jeganathan K, Yamada S, Reyes S et al (2008) Opposing roles for p16Ink4a and p19Arf in senescence and ageing caused by BubR1 insufficiency. Nat Cell Biol 10(7):825–836. https://doi.org/10.1038/ncb1744 CrossRefPubMedPubMedCentralGoogle Scholar
- 61.Doctrow SR, Lopez A, Schock AM, Duncan NE, Jourdan MM, Olasz EB, Moulder JE, Fish BL et al (2013) A synthetic superoxide dismutase/catalase mimetic EUK-207 mitigates radiation dermatitis and promotes wound healing in irradiated rat skin. J Invest Dermatol 133(4):1088–1096. https://doi.org/10.1038/jid.2012.410 CrossRefPubMedGoogle Scholar
- 63.Galimov ER (2010) The role of p66shc in oxidative stress and apoptosis. Acta Nat 2(4):44–51Google Scholar
- 68.Jurk D, Wang C, Miwa S, Maddick M, Korolchuk V, Tsolou A, Gonos ES, Thrasivoulou C et al (2012) Postmitotic neurons develop a p21-dependent senescence-like phenotype driven by a DNA damage response. Aging Cell 11(6):996–1004. https://doi.org/10.1111/j.1474-9726.2012.00870.x CrossRefPubMedPubMedCentralGoogle Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.