Noise Health Home 

[Download PDF]
Year : 1999  |  Volume : 2  |  Issue : 5  |  Page : 53--69

Evidence of a common pathway in noise-induced hearing loss and carboplatin ototoxicity

Donald Henderson1, Bohua Hu2, Sandra L McFadden1, Xiangyang Zheng1,  
1 Center for Hearing and Deafness, SUNY at Buffalo, 215 Parker Hall, Buffalo, New York 14214, USA
2 Institute of Otolaryngology, Chinese PLA General Hospital, Beijing, P.R., China

Correspondence Address:
Donald Henderson
Center for Hearing and Deafness, 215 Parker Hall, State University of New York at Buffalo, Buffalo, NY 14214


In spite of the differences in the nature of the insult, the hearing loss from ototoxic drugs and noise exposure share a number of similarities in cochlear pathology. This paper explores the common factors between noise-induced hearing loss and ototoxicity by experimentally manipulating cochlear glutathione (GSH). In the first experiment, chinchillas were treated with a drop of saline (50 µl) on the round window of one ear and a drop of buthionine sulfoximine (BSO, 50 µl of 200 mM) on the other ear. BSO is a drug that blocks GSH synthesis and it was hypothesized that GSH-depressed ears would be more vulnerable to noise. Six hours after treatment, the animals were exposed to a 105 dB 4 kHz octave band noise for 4 hours, then a second dose of BSO was applied 2 hours later. The BSO treated ears showed more temporary threshold shifts and reduced GSH staining at day 4 post exposure, but there was no BSO effect in terms of greater permanent threshold shift (PTS) or hair cell loss. In the second experiment, chinchillas were pretreated with BSO and 3 days later were given either a single dose of carboplatin (25 mg/kg i.p.), a double dose (day 3 and 7) or only BSO. Chinchillas that received BSO and the double dose of carboplatin had significantly greater loss of inner and outer hair cells than the carboplatin chinchillas. In addition, the BSO and carboplatin chinchillas also had larger decreases in evoked response amplitudes suggesting that GSH depletion potentiated the ototoxicity of carboplatin. These results are discussed in terms of the role of reactive oxygen species in creating hearing loss and the potential protective role of glutathione.

How to cite this article:
Henderson D, Hu B, McFadden SL, Zheng X. Evidence of a common pathway in noise-induced hearing loss and carboplatin ototoxicity.Noise Health 1999;2:53-69

How to cite this URL:
Henderson D, Hu B, McFadden SL, Zheng X. Evidence of a common pathway in noise-induced hearing loss and carboplatin ototoxicity. Noise Health [serial online] 1999 [cited 2021 Nov 28 ];2:53-69
Available from:

Full Text


The hypothesis of a common pathway underlying the hearing loss caused by noise exposure and certain ototoxic drugs is based on similarities in their cochlear pathology and audiometric profiles. For example, noise exposures create a hearing loss that is related to the spectral characteristics of the traumatizing exposure. For example, many industrial settings are associated with a relatively broad band noise that is enhanced by the resonance of the external auditory meatus to deliver a band of noise centered at approximately 3 kHz. Thus, the typical audiogram of noise induced hearing loss (NIHL) has either a peak at 4 kHz or has a loss at 4 and 8 kHz; aminoglycosides and platinum compounds produce a hearing loss that starts at the highest frequencies. Both noise and drugs typically target the outer hair cells at the base of the cochlea with the stereocilia often being the first to show pathological changes such as "blebs" or fusion. With greater hearing losses the pathology becomes more pervasive including supporting cell damage, lesions in the stria vascularis and, in very severe cases, inner hair cell loss. As Hawkins (1973) pointed out almost 25 years ago, it is difficult to predict the cochlear pathology from knowing the pattern of the audiogram and, as a corollary it is difficult to identify the cause of the cochlear lesions from the pattern of cochlear pathology.

The similarity between the pathology associated with noise-induced hearing loss (NIHL) and drug-induced hearing loss may be a reflection of a common trigger, reactive oxygen species (ROS) leading to the sequela of cochlear pathology. Much of the data on the role of ROS in tissue damage comes from cell or organ culture experiments. Unfortunately, experiments with mammalian cochlear organ cultures have been limited to studies of ototoxic effects in the developing cochlea and there is no viable model for studying NIHL in cultures. Additional problems in studying the direct effects of ROS arise because of the short half life of ROS and the difficulty in accessing the living cochlea.

In spite of the lack of direct evidence, there are compelling reasons and strong circumstantial evidence for ROS and their potential role in hearing loss. Given the energy consuming operation of the cochlea, the generation of ROS is a normal part of homeostasis. However, with high levels of stimulation, the cellular respiratory process of the mitochondria cannot accommodate the higher levels of demand and produce excessive ROS. Thus, the balance between ROS and their normal reactions with cellular antioxidants shifts with an accumulation of ROS. In addition, high levels of stimulation lead to excitotoxic reactions and swelling of the afferent dendrites, as well as ischaemia as reflected in ROS accumulation at the stria vascularis. Yamane et al. (1995) showed the intimate correlation between strial blood flow and the presence of the superoxide radical O2 - in guinea pigs following noise exposure. They reported that immediately after a noise exposure strial blood flow was either greatly reduced or stopped and there was an accumulation of O2­ along the endolymph margin of the stria. Two hours later, blood flow had resumed and the O2'­ concentration was greatly reduced; six hours later blood flow was normal and O2 - was not detectable. The implication of the Yamane et al. (1995) results is that noise exposure creates cochlear ischaemia with an increase in O2­ presence. The toxic O2 - and related ROS are then responsible for assaults on the cell membrane system, structural proteins and cell nucleus.

Additional evidence that ROS contribute to NIHL comes from pharmacological studies of prevention with R-phenylisopropyladenosine (R-PIA). Studies by Hu et al. (1997) and Liu et al. (1999) showed that the hearing loss and hair cell damage from exposure to either a 105 dB SPL 4 kHz octave band (OB) noise or a 150 dB peak SPL impulse noise could be reduced with a prior application of R-PIA to the round window. These results are exciting in terms of their clinical implications but they are ambiguous in terms of the cellular mechanisms involved. R­PIA is a selective adenosine receptor agonist that has been found to upregulate the activity of several antioxidant enzymes, including superoxide dismutase (SOD), glutathione peroxidase, glutathione reductase, and catalase (Maggirwar et al., 1994). R-PIA is also a potential inhibitor of glutamate synthesis and a stimulus for nitrous oxide (NO) production. All three of these effects of R-PIA have positive implications for reducing hearing loss. For example, R-PIA may reduce ROS-mediated damage by stimulating antioxidant production; blocking glutamate might reduce cytotoxicity seen at afferent auditory nerve dendrites following noise (Spoendlin, 1976); or increased NO availability might stimulate blood flow to the cochlea which would partially reduce ischaemic conditions.

The source for an increase in ROS may be quite different for drugs such as cisplatin and carboplatin. Cisplatin and carboplatin are members of a family of platinum-based drugs that are used in the treatment of various types of tumours in humans. One potential side effect of the anti-cancer drugs is hearing loss, particularly at high frequencies (Frelich et al., 1996; McKeage, 1995; Macdonald et al., 1994). Platinum cytotoxicity has been shown to be accelerated or enhanced in the presence of ROS, and decreased by antioxidants, particularly glutathione (GSH), a thiol/sulfhydryl compound consisting of y-glutamate, y-cysteine and glycine (Ahn et al., 1994; Arrick and Nathan, 1984; Babu et al., 1995; de Graeff et al., 1988; Eastman, 1987; Ferguson, 1995; Lai et al., 1989; McGinness et al., 1978; Mistry et al., 1993; Russo et al., 1986; Tonetti et al., 1993; Walker and Gale, 1981). Clear correlations have been found between GSH levels and sensitivity to platinum compounds (Ahn, 1994; Arrick and Nathan, 1984; Meijer et al., 1990, 1992; Mellish et al., 1993; Mistry et al., 1991). Cells with inherent or acquired resistance to platinum compounds have higher levels of GSH than susceptible cells, and depletion of cellular GSH can increase sensitivity to damage. In organ culture, cisplatin has been shown to increase superoxide dismutase (SOD), catalase and malondialdehyde activities, while reducing the activity levels of GSH, GSH-peroxidase and GSH-reductase (Ravi et al., 1995). In vitro experiments are interesting, but it can be risky to simply extrapolate the results to the organ system (ear) because the state of equilibrium between ROS and the set of antioxidant molecules is complex. Consequently, when the ear is poisoned with cisplatin, the normal antioxidant defence system is disrupted and may be unable to keep up with the normal ROS that are generated when the ear is stimulated. By contrast, high levels of noise exposure produce an increase in ROS (Yamane et al., 1995a,b).

The common pathway hypothesis may help explain the particularly traumatic interaction seen with cisplatin and noise. Gratton et al. (1990) reported that cisplatin and noise exposures that were individually non-traumatic could combine to produce substantial hearing loss and hair cell damage in chinchillas. If cisplatin depletes GSH and other components of the antioxidant defence system (Ravi et al., 1995), then the ear would be expected to be even more vulnerable to the ROS generated by the noise exposure. The common pathway with noise and certain ototoxic drugs which leads to cochlear pathology appears to be toxic ROS. However, the processes leading to an accumulation of ROS may be quite different.

The following two experiments continue to explore the commonality between the patho­physiology of noise- and drug-induced hearing loss by explicitly blocking one component of the antioxidant pathway, GSH, and observing the effects on hearing loss and hair cell loss in chinchillas. In both experiments, buthionine­sulfoximine (BSO) was used to inhibit g­glutamyl cysteine synthtase (GCS) which is the rate limiting enzymes for GSH synthesis. In Experiment I, a small dose of BSO was applied to the round window membrane before and after animals were exposed to high-level noise.

In Experiment II, BSO was chronically infused into the cochlea via osmotic pumps before and after animals were treated with carboplatin. Carboplatin is related to cisplatin and is used in treatment with a number of cancers because it has less severe side effects than cisplatin. In people, guinea pigs and rats, carboplatin produces a high-frequency loss with a basal lesion of OHC. Carboplain in the chinchilla produces a loss of IHC with virtually no loss of OHC as decrease in distortion product. The introduction of BSO with carboplatin will provide a perspective on the importance of glutathione with the unique IHC loss of the chinchilla. The results support the notion of a common pathway leading to NIHL and carboplatin ototoxicity.

Experiment I: BSO alters the course of recovery from NIHL

BSO is a specific inhibitor of g-glutamylcysteine synthetase (GCS), the rate-limiting enzyme in the synthesis of GSH from its constituent amino acids (Gao et al., 1994; Grieshaber, 1989; Griffith and Meister, 1979; Hoffman et al., 1988; Meister, 1991, 1995). Because BSO treatment blocks GSH synthesis and leads to reduced levels of GSH both in vitro and in vivo, it has been used as a tool for depleting GSH in a variety of cells and tissues, including the cochlea (Lazenby et al., 1988; Murray et al., 1986). Given the role of GSH in the cellular antioxidant system, the fact that GSH is upregulated with noise exposure (Bobbin et al., 1995), and the facilitative effects of R-PIA on recovery from noise-induced damage (Hu et al., 1997), we hypothesized that BSO-treated ears would have larger noise-induced threshold shifts and greater hair cell (HC) losses than saline-treated control ears.


Sixteen adult chinchillas served as subjects. Animals were divided into noise-exposed (N=12) and non-exposed (N=4) groups. Two noise-exposed animals were sacrificed 4 days after exposure for mercury orange staining to look at GSH distribution in the hair cells, while the other 10 noise-exposed animals were sacrificed along with non-exposed animals at 20 days post-exposure for hair cell counts. Procedures for the care and use of all animals in our studies were approved by the University of Buffalo Institutional Animal Care and Use Committee.

Each animal was anesthetized with an intramuscular injection of ketamine (60 mg/kg) and acepromazine (0.5 mg/kg), and recording electrodes were implanted in each inferior colliculus (IC) and in the rostral cranium. After two or more weeks of recovery, the animals were again anesthetized, and small holes were made in the bullae. Approximately 50 µl of 200 mM BSO (Sigma Chemicals) in physiological saline was dropped onto the right round window, and an equal volume of physiological saline was dropped onto the left round window. Stainless steel tubes were implanted above the round window niches for application of a second dose after the noise exposure. The tubes were sealed to the bulla with dental cement, and the skin incision was sutured. We applied the first dose of BSO at 6 h before noise exposure and a second dose via the stainless steel tubes at 2 h after noise exposure. The purpose of the two dose schedule was to maximise the likelihood of depleting GSH during the early stage of hair cell recovery. After the second BSO application, the tubes were removed and the holes in the bullae were sealed with dental cement.

Animals in the noise-exposed group were exposed to a 4 kHz OB noise at 105 dB SPL (re: 20 mPa) for 4 h. The noise was generated by a D/A converter on a digital signal processing board (DSP) in a personal computer (PC), and routed through a manual attenuator (HP 350D), a filter, and a low-distortion power amplifier (NAD 2200) to an acoustic horn (JBL 2360) suspended directly above the cages in a sound booth. Noise levels were measured using a Type I sound level meter (Larson and Davis 800B) and a ˝" microphone (Larson and Davis, LDL 2559), with the microphone positioned within the cage at the level of an animal's head.

The effects of GSH depletion on cochlear function were assessed by recording auditory evoked potentials from the IC (IC-EVPs) and cubic distortion product otoacoustic emissions (DPOAEs). For IC-EVP testing, stimuli were digitally generated tones (10 ms duration, 5 ms cosine rise/fall) at 1, 2, 4, 6, and 8 kHz, presented at a rate of 10/sec. The stimuli were routed through a computer-controlled attenuator to an insert earphone (Etymotic Research ER-2). The output of the insert earphone was calibrated before each test. Electrical activity from the recording electrode was amplified (20,000 X) and filtered (10-3000 Hz) by a Grass P511K bioamplifier, and directed to an A/D converter on a DSP board in the PC. Stimulus level was varied in 5 dB steps from 0 to 80 dB SPL. One hundred samples were averaged at each level. Threshold was defined as the mid-point between the lowest level at which a clear response was seen and the next lower level where no response could be discerned. The thresholds of IC-EVPs were determined before BSO application and at various times after the noise exposure (15 minutes; 1, 2, 4 and 20 days).

For DPOAE testing, pure tone stimuli (f 1 and f 2 ) were generated by two DSP boards in a PC, and sound levels in the ear canal were measured using a low-noise probe microphone. Input/output (I/O) functions were recorded in 5 dB steps from 0 to 80 dB SPL, using equal-level primaries and an f 2 /f 1 ratio of 1.2. I/O functions were collected at f 1 =2, 4, and 8 kHz, in random order. The animals were held in a custom-built restraining device (Snyder and Salvi, 1994) and tested while awake. At least two sets of I/O functions were obtained prior to the noise exposure, and the average of these measures served as the baseline. DPOAEs were also measured at 15 minutes, and 1, 2, 4, and 20 days after noise exposure.

All noise-induced threshold shifts (TSs) were calculated relative to the threshold measured before the application of BSO or saline. The mean TSs of the BSO- and saline-treated ears were compared using Student t-tests. For comparison of DPOAE amplitudes, mean amplitudes for stimuli at 60 dB SPL (L 1 =L 2 ) were compared as a function of treatment (BSO versus saline) with Student t-tests.

After final auditory tests had been completed, the animals were anesthetized with a lethal dose of sodium pentobarbital and decapitated. Each bulla was quickly removed from the skull, and the cochleas were slowly perfused through the round window with a succinate dehydrogenase (SDH) staining solution. The apex of the cochlea was removed, and the cochlea was immersed in the SDH staining solution for 1 h at 37°C, followed by immersion in 10% formaldehyde for 24 h. The cochlea was dissected and sections of the organ of Corti were mounted in glycerin on glass microscope slides and cover slipped. The specimens were examined for HC losses under a light microscope.

We also examined GSH staining in two animals sacrificed 4 days after noise exposure. Both left and right bullae were quickly removed and the cochleas were exposed. The ossicles were removed, the oval window was opened, and 50 mM mercury orange (1-4-chloro-mercury­phenyl-azo-2-naphthol; Sigma Chemicals) in toluene was gently perfused through the round window membrane with a fine pipette. The cochleas were then immersed in the same solution for 20 minutes at room temperature, rinsed in two changes of toluene, and fixed with 10% formalin in phosphate buffer (pH 7.4) for 1 hour. The cochleas were dissected in phosphate buffer and the organ of Corti sections were mounted in 50% glycol on glass microscope slides for examination with a Bio-Rad MRC 1000 laser scanning confocal unit attached to a Nikon microscope. Optical sections were collected at 0.5 mm intervals through the organ of Corti.


The mean IC-EVP thresholds for right ears (open circles) and left ears (filled circles) of all 16 chinchillas before the BSO and saline treatment are shown in [Figure 1]. There were no significant differences between right and left ears at any frequency prior to drug application. [Figure 2] shows IC-EVP thresholds as a function of time and frequency for the four non-exposed animals.

IC-EVP thresholds remained stable over the 20­day period, indicating that the BSO treatment alone had no effect on auditory sensitivity.

[Figure 3] compares mean TSs of the BSO treated ears and the control ears 15 minutes (0 days; n=12 animals), 4 days (n=12 animals), and 20 days (n=10 animals) after noise exposure at 2, 4, 6 and 8 kHz. The mean TS measured 15 min after the noise exposure was 66.6 dB. There were no significant differences in TSs at this point between BSO-treated and control ears. One to two days after noise exposure, thresholds of the BSO-treated ears had recovered 4.4 to 12.7 dB less than thresholds of the saline-treated ears at 4 and 8 kHz. However, these differences were not statistically significant. Four days after noise exposure, the BSO-treated ears showed 9.4, 16.3, 18.8, and 13.1 dB less recovery at 2, 4, 6 and 8 kHz, respectively. Differences between left and right ears were significant at 4, 6 and 8 kHz (p = 0.018, 0.002 and 0.044, respectively). Although BSO-treated ears showed less recovery between 0 and 4 days after exposure, there were no significant differences between ears 20 days after noise exposure.

[Figure 4] shows DPOAE amplitudes for 2, 4 and 8 kHz stimuli measured in saline-treated ears (solid lines) and BSO-treated ears (dashed lines). The solid lines without symbols show DPOAEs measured prior to drug treatment and noise exposure. Lines with symbols show DPOAE amplitudes measured at 15 minutes (top panel), 4 days (middle panel) and 20 days (bottom panel) following the noise exposure. Amplitudes at all three frequencies decreased significantly 15 minutes after exposure, but the magnitude of loss was not significantly different between the BSO­ treated ears and the saline-treated ears. DPOAE amplitudes recovered substantially between 15 minutes and 4 days, with greater recovery in saline-treated ears than in BSO-treated ears. On Day 4, BSO-treated ears showed 12.0 and 11.5 dB less recovery at 60 dB SPL at 4 and 8 kHz (p = 0.012 and 0.024, respectively). By 20 days after noise exposure, however, the DPOAE amplitudes of both experimental and control ears had essentially recovered to the pre-exposure values and differences between BSO- and saline­ treated ears were negligible.

Two noise-exposed animals were sacrificed 4 days after exposure for mercury orange staining. The IHC were only weakly stained in all four ears, and no differences were seen between the BSO-treated ears and the control ears. In contrast, there were obvious and consistent differences in staining patterns of OHCs. [Figure 5] illustrates mercury orange staining in the region of the cochlea with the largest OHC lesions, at an optical level of 6-8 µm below the cuticular plate. The upper panel shows mercury orange staining in a saline-treated ear. Note that areas where OHCs are missing (arrow heads) are completely devoid of staining, whereas most surviving OHCs show a normal GSH staining pattern, characterised by a ring-like structure with strong central staining. Only a few OHCs showed enlarged cell bodies with weakly-stained central portions (arrows). The BSO-treated ears exhibited GSH staining patterns (lower panel) that were much different from controls. Enlarged OHCs, which usually lacked the strong bolus of central staining (arrows), appeared to be more numerous. Normal-sized OHCs also showed different staining patterns. The normal ring-like structure was replaced by relatively even mercury orange staining throughout the cytoplasm.

[Figure 6] shows mercury orange staining in an area of the cochlea where relatively little noise­ induced HC loss occurred. The upper panel shows staining in a saline-treated ear and the lower panel shows staining in a BSO-treated ear. Note that enlarged OHCs are more numerous in the BSO-treated ear than in the saline-treated ear. Despite differences between BSO- and saline-treated ears in GSH staining at 4 days post­exposure, there were no differences in HC loss at 20 days post-exposure.


The dose of BSO used in this study did not interfere with normal function or survival of cochlear HCs. However, application of BSO to the round window membrane altered the pattern of recovery of IC-EVP thresholds and DPOAE amplitudes following noise exposure. There were no differences between the BSO- and saline-treated ears immediately after the exposure, suggesting that GSH depletion did not potentiate the initial noise-induced damage to the hair cells. However, 4 days after noise exposure, physiologic measurements showed significantly less recovery in BSO-treated ears. Consistent with this, mercury orange staining at 4 days post­exposure showed reduced GSH levels in BSO­ treated ears.

The reason why the effects of BSO were evident 4 days after noise exposure and not earlier or later is not clear. The simplest explanations are related to the dose and time course of BSO application. It seems likely that the small dose of BSO used in this experiment produced only modest decreases in GSH levels, and that GSH levels were able to recover relatively rapidly, before permanent damage could occur. Another possibility to consider is that the acute application of BSO may have caused a compensatory increase of other antioxidant enzymes. If this is the case, then more general suppression of the antioxidant system should result in greater permanent deficits than we observed here.

In the saline-treated ears, areas of missing OHCs were devoid of staining. The surviving OHCs generally showed normal shapes and a normal staining pattern, characterised by staining concentrated in the central portion and around the perimeter of the cells. In contrast, the BSO ­treated ears showed variable staining patterns. Some OHCs had enlarged cell bodies, suggesting that swelling had occurred. Others, even those of apparently normal size, lost the normal pattern of staining, and mercury orange staining became diffusely distributed throughout the cytoplasm. In addition, in less damaged areas of the cochlea, the number of swollen OHCs appeared to be greater in BSO-treated ears than in saline-treated ears.

One limitation of the mercury orange staining method is a lack of specificity, because it reacts rapidly with other low molecular weight thiols such as cysteine (Murray et al., 1986). Therefore, the residual staining seen in BSO ­treated ears cannot be attributed specifically to GSH. However, because BSO is a specific inhibitor of GSH synthesis, it is reasonable to attribute any differences in staining patterns to BSO-induced depletion of GSH. While staining was not eliminated in BSO-treated ears, its distribution was altered dramatically, and the alteration was correlated with an altered pattern of recovery after noise exposure.

 Summary and Conclusions

The results of Experiment I suggest that ears with decreased levels of GSH are more vulnerable to noise trauma than normal ears of the same animals. Increased vulnerability was evident in two separate physiological measures, IC-EVPs and DPOAEs, and correlated with diminished mercury orange staining of GSH and other cellular thiols. The effects were modest and temporary, appearing within 4 days after exposure and dissipating by 20 days, probably as a consequence of our BSO dosing parameters. It is likely that more profound and permanent effects would have been seen with chronic administration of BSO throughout the recovery period. In the next experiment, we used osmotic pumps to infuse BSO directly into the cochlea for an extended period of time, in order to test the hypothesis that GSH depletion potentiates HC damage from carboplatin.

Experiment II: GSH Depletion Potentiates Carboplatin Ototoxicity

The ototoxic effects of carboplatin have been studied extensively in guinea pigs and chinchillas. In guinea pigs, the pattern of HC loss caused by carboplatin represents the typical response of the mammalian cochlea to ototoxic drugs and noise, i.e., greater vulnerability of OHCs versus IHCs, and greater vulnerability of basal HCs versus apical HCs (Schweitzer et al., 1986; Takeno et al., 1994a,b). Consequently, guinea pigs treated with carboplatin develop a hearing loss that is primarily high-frequency in nature (Taudy et al., 1992). The chinchilla, by contrast, has an atypical response to carboplatin (but not to other ototoxic drugs or noise). For reasons that are not yet known, carboplatin selectively destroys IHCs throughout the cochlea of the chinchilla, leaving OHCs intact except at high doses (Hofstetter et al., 1997a,b; Takeno et al., 1994a,b; Wake et al., 1993, 1994). The ability to selectively destroy IHCs while leaving OHCs intact provides us with unprecedented opportunities to study the effects of IHC loss independent of OHC loss. Studies using the unique carboplatin-treated chinchilla model of IHC loss have confirmed that the IHCs make little or no contribution to DPOAEs (Hofstetter et al., 1997a; Jock et al., 1996; Trautwein et al., 1996), and have led to the unexpected finding that thresholds of cochlear potentials and IC­EVPs remain normal with moderate-to-large losses of IHCs as long as OHCs remain intact (Burkard et al., 1997; Jock et al., 1996; McFadden et al., 1998; Trautwein et al., 1996).

Previous studies have implicated GSH and GSH ­dependent enzymes in platinum toxicity. With respect to ototoxicity, Ryback et al. (1995) showed that administration of diethyldithiocar­bamate, a compound that increases GSH and GSH-peroxidase activity, significantly reduced threshold shifts in rats treated with cisplatin, and Ravi et al. (1995) demonstrated that cochlear GSH is down-regulated with application of cisplatin. The purpose of the present study was to determine if carboplatin ototoxicity in the chinchilla cochlea is enhanced when GSH synthesis is inhibited. The severity of carboplatin ototoxicity was assessed using the same physiological measures and procedures used in Experiment I, and these measurements were complemented by cochleogram showing IHC and OHC losses. Since previous studies have shown that carboplatin preferentially damages the IHCs in chinchilla, we were particularly interested to see if GSH depletion would increase the amount of damage to the OHCs.


Twelve chinchillas were randomly divided into three groups of 4 animals each: a single-dose carboplatin group that received BSO in the right ear, followed 3 days later by a single dose of carboplatin (25 mg/kg i.p.); a double-dose carboplatin group that received BSO in the right ear followed by two doses of carboplatin (25 mg/kg i.p. X 2), at 3 and 7 days after the beginning of BSO treatment; and a drug control group, that only received BSO in the right ear without any carboplatin.

Each subject was anesthetized with a mixture of ketamine (60 mg/kg) and acepromazine (0.5 mg/kg), and tungsten recording electrodes were stereotaxically implanted into the left and right IC as described for Experiment I. Osmotic pumps (2ML4, Alza Corporation) (Brown et al., 1993; Schindler et al., 1995) were also implanted in each animal. The right bulla was exposed through a posterior approach and a small hole was drilled in the bony wall of the basal turn of the cochlea. A stainless steel tube was inserted into scala tympani and sealed to the cochlea with silicone glue. Polyethylene tubing connected the stainless steel tube to the osmotic pump. The pump was implanted under the skin on the back of the neck, and the incision was sutured closed. The pump, pre-filled with 15 mM BSO in Hanks balanced salt solution, had an infusion rate of 5 µl/hr. The infusion period lasted 14 days.

Three days after implanting the osmotic pump, animals in the two carboplatin groups were injected with carboplatin. Animals in the double-dose group received a second dose of carboplatin 4 days after the first.


Thresholds and I/O functions of IC-EVPs and DPOAEs were examined before and at various times during the BSO infusion. Drug control subjects showed small losses of sensitivity (5-10 dB on average for IC-EVPs) at 2 days after beginning the BSO treatment, but complete recovery of thresholds and response amplitudes by 7-14 days. When animals in the drug control group were sacrificed for histology after 14 days of BSO infusion, no IHC or OHC losses were observed. Thus, both physiological and anatomical measures show that GSH depletion causes no direct cochlear damage.

Three animals in the single-dose carboplatin group showed relatively minor IHC losses from carboplatin, and no significant threshold shifts. As shown in [Figure 7]A, the average IHC loss of these animals was only 10-30% in the 1-10 kHz region of the cochlea, and there were no significant differences between treated and non treated ears. In contrast, one animal in the single-dose group showed large IHC losses [Figure 7B], and a significant difference (p <0.001) between the BSO-treated ear and the normal ear. The average IHC loss for this animal was 18.74% ± 4.6 in the control ear, versus 29.97% ± 2.74 in the BSO-treated ear. OHC loss was negligible in all ears.

The most pronounced effects of GSH depletion were seen in the double-dose carboplatin group. Both BSO-treated and control ears showed a significant decrease in IC-EVP response amplitudes after carboplatin treatment, with slightly greater decreases in BSO-treated ears. Interestingly, control ears showed no significant change in DPOAE amplitude after carboplatin treatment, while BSO-treated ears exhibited decreased amplitudes at 2, 4, and 8 kHz. At primary levels of 60 dB SPL, the BSO-treated ears had amplitude shifts of 4-12.3 dB relative to control ears and pre-drug values.

The most dramatic differences between ears were seen in HC losses. As shown in [Figure 8], the double dose of carboplatin not only produced more IHC loss in BSO-treated ears (top panel), but also significant OHC loss (bottom panel). Carboplatin produced an average IHC loss of 18.1% ± 4.62 in control ears versus 59.1% ± 2.71 in BSO-pretreated ears. Control ears showed scattered OHC losses along the organ of Corti, with an average OHC loss of only 1.9% ± 8.8. In contrast, the average OHC loss for the BSO ­treated ears was 36.7% ± 6.0. Both IHC and OHC differences were statistically significant (p <0.001).


The results show that infusion of BSO (15 mM) directly into the chinchilla cochlea potentiates the ototoxicity of carboplatin. Potentiation was seen in one of four animals treated with a single low dose of carboplatin, and in all four animals in the double-dose group. The differences between the BSO-treated ears and the control ears of animals in the double-dose carboplatin group were dramatic. BSO-treated ears showed approximately 41% more IHC loss than control ears, and significantly more OHC loss (37%, versus 2%) as well.

The results are consistent with the idea that carboplatin treatment itself reduces cellular GSH, as has been shown to occur following cisplatin treatment (Ravi et al., 1995) and that the combined effects of BSO and carboplatin depleted GSH levels beyond a critical level for producing permanent damage. Presumably, permanent damage occurs only when the scavenging capacity of the antioxidant defence system falls below a critical level relative to ROS production. The hypersensitivity of the one chinchilla in the single-dose carboplatin group could be an indication of a generally less efficient antioxidant defence system or a heightened production of ROS in this animal. In future studies, it will be useful to develop techniques for assaying cochlear levels of GSH and other antioxidant enzymes and correlating them with individual susceptibility to address these issues directly.

The most intriguing finding of this study was the potentiation of OHC loss in BSO-treated ears, since previous studies have clearly established that OHCs in the chinchilla are much less susceptible to carboplatin than IHCs. In the current study, OHC loss was minor in all control ears and BSO-treated ears of animals in the single-dose group. Consistent with this, DPOAEs, which rely on OHCs for their generation (Trautwein et al., 1996), were normal in these ears. In contrast, BSO-treated ears of animals in the double-dose group had OHC losses scattered throughout the cochlea, resulting in reduced DPOAE amplitudes. One interpretation of these results is that OHCs in the chinchilla cochlea are susceptible to carboplatin only when antioxidant enzyme levels fall below a critical level.

Previous studies have shown that the sensitivity of tumour cells to platinum drugs is related to intracellular levels of GSH and related enzymes (e.g., Meijer et al., 1992; Mellish et al., 1993; Mistry et al., 1991). Inherent and acquired resistance to platinum drugs through up­regulation of intracellular GSH poses an important problem for cancer treatment. Although studies show that decreasing cellular

GSH levels by pre-treatment with BSO can increase the platinum sensitivity of some resistant tumour cell lines (Lee et al., 1992; Meijer et al., 1992; Mistry et al., 1991; Singh et al., 1995), our results suggest that this approach may potentiate ototoxic damage. Developing techniques for increasing the sensitivity of tumour cells to chemotherapy while avoiding increased damage to cochlear hair cells is an important challenge for both basic and clinical research.

 Summary and Conclusions

Since BSO is a specific and irreversible inhibitor of g-GCS, the effects of BSO on susceptibility to both noise and carboplatin are most likely mediated by inhibition of cochlear GSH synthesis (Kera et al., 1989; Kisera et al., 1995). The results of the two experiments described above provide strong circumstantial evidence of a common pathway for damage from noise and carboplatin.


This research was supported by the Center for Hearing and Deafness at the University of Buffalo, and by grants 5RO1-DC-0123703 (D. Henderson), RO1 DC 00166-13 (R.J. Salvi) and DAMD17-96-1-6330 (S.L. McFadden).[57]


1Ahn, H., Lee, E., Kim, K., & Lee, C. (1994) Effect of glutathione and its related enzymes on chemosensitivity of renal cell carcinoma and bladder carcinoma cell lines. J. Urol. 151:263-267.
2Arrick, B.A., & Nathan, C.F. (1984) Glutathione metabolism as a determinant of therapeutic efficacy: a review. Cancer Res. 44:4224-4232.
3Babu, E., Gopalakrishnan, V.K., Sriganth, I.N., Gopalakrishnan, R., & Sakthisekaran, D. (1995) Cisplatin induced nephrotoxicity and the modulating effect of glutathione ester. Mol. Cell Biochem. 144:7-11.
4Bobbin, R.P., Fallon, M., LeBlanc, C., & Baber, A. (1995) Evidence that glutathione is the unidentified amine released by high potassium into cochlear fluids. Hear Res 87:49-54.
5Brown, J.N., Miller, J.M., Altschuler, R.A., &Nuttall, A.L. (1993) Osmotic pump implant for chronic infusion of drugs into the inner ear. Hear. Res. 70:167-172.
6Burkard, R., Trautwein, P., & Salvi, R. (1997) The effects of click level, click rate and level of background masking noise on the inferior colliculus potential (ICP) in the normal and carboplatin-treated chinchilla. J. Acout. Soc. Am. 102:3620-3627.
7de Graeff, A., Slebos, R.J., & Rodenhuis, S. (1988) Resistance to cisplatin and analogues: mechanisms and potential clinical implications. Cancer Chemother. Pharmacol. 22:325-332.
8Eastman, A. (1987) Glutathione-mediated activation of anticancer platinum(IV) complexes. Biochem. Pharmacol. 36:4177-4178.
9Ferguson, P.J. (1995) Mechanisms of resistance of human tumours to anticancer drugs of the platinum family: a review. J. Otolaryngol. 24:242-252.
10Freilich, R.J., Kraus, D.H., Budnick, A.S., Bayer, L.A., & Finlay, J.L. (1996) Hearing loss in children with brain tumors treated with cisplatin and carboplatin-based high­dose chemotherapy with autologous bone marrow rescue. Med. Pediatr. Oncol. 26:95-100.
11Gao, G., Ollinger, K., & Brunk U.T. (1994) Influence of intracellular glutathione concentration of lipofuscin accumulation in cultured neonatal rat cardiac myocytes. Free Rad Biol Med 16:187-194.
12Gratton M.A., Salvi R.J., Kamen B.A., & Saunders S.S. (1990) Interaction of cisplatin and noise on the peripheral auditory system. Hear. Res. 50:211-23.
13Grieshaber, C.K. (1989) Pharmacokinetics of buthionine sulfoximine (NSC 326231) and its effect on melphalan­induced toxicity in mice. Cancer Res. 49:5385-5391.
14Griffith, O.W., & Meister, A. (1979) Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocysteine sulfoximine). J Biol Chem. 254:7558-7560.
15Hawkins, J.E., Jr. (1973) Comparative otopathology: aging, noise, and ototoxic drugs. Adv Otorhinolaryngol 20, 125-141.
16Hoffman, D.W., Whitworth, C.A., Jones-King, K.L., & Rybak, L.P. (1988) Potentiation of ototoxicity by glutathione depletion. Ann. Otol. Rhinol. Laryngol. 97:36­41.
17Hofstetter, P., Ding, D., Powers, N., & Salvi, R.J. (1997a) Quantitative relationship between carboplatin dose, inner and outer hair cell loss and reduction in distortion product otoacoustic emission amplitude chinchillas. Hear. Res. 112:199-215.
18Hofstetter, P., Ding, D., & Salvi, R.J. (1997b) Magnitude and pattern of inner and outer hair cell loss in chinchilla as a function of carboplatin dose. Audiol. 36:301-311.
19Hu, B.H., Zheng, X.Y., McFadden, S.L., Kopke, R.D., & Henderson, D. (1997) R-phenylisopropyladenosine attenuates noise-induced hearing loss in the chinchilla. Hear. Res. 113:198-206.
20Jock, B.M., Hamernik, R.P., Aldrich, L.G., Ahroon, W.A., Petriello, K.-L., & Johnson, A.R. (1996) Evoked-potential thresholds and cubic distortion product otoacoustic emissions in the chinchilla following carboplatin treatment and noise exposure. Hear. Res. 96:179-190.
21Kera, Y., Ohbora, Y., & Komura, S. (1989) Buthionine sulfoximine inhibition of glutathione biosynthesis enhances hepatic lipid peroxidation in rats during acute ethanol intoxication. Alcohol 24:519-524.
22Kisara, S., Furusawa, S., Takayanagi, Y., & Sasaki, K. (1995) Effect of glutathione depletion by buthionine sulfoximine on doxorubicin toxicity in mice. Res. Comm. Mol. Pathol. Pharmacol. 89:401-410.
23Lai, G.M., Ozols, R.F., Young, R.C., & Hamilton, T.C. (1989) Effect of glutathione on DNA repair in cisplatin­resistant human ovarian cancer cell lines. J. Natl. Cancer Inst. 81:535-539.
24Lazenby, C.M., Lee, S.J., Harpur, E.S., & Gescher, A. (1988) Glutathione depletion in the guinea pig and its effect on the acute cochlear toxicity of ethacrynic acid. Biochem. Pharmacol. 37:3743-3747.
25Lee, K.S., Kim, H.K., Moon, H.S., Hong, Y.S., Kang, J.H., Kim, D.J., & Park, J.G. (1992) Effects of buthionine sulfoximine treatment on cellular glutathione levels and cytotoxicities of cisplatin, carboplatin and radiation in human stomach and ovarian cancer cell lines. Korean. J. Intern. Med. 7:111-117.
26Leeuwenburgh, C, & Ji, LL (1995) Glutathione depletion in rested and exercised mice: biochemical consequences and adaptation. Arch. Biochem. Biophys. 316:941-949.
27Liu, C.C. et al. (1999) Assoc. Res. Otolaryngol. Abstr
28Macdonald, M.R., Harrison, R.V., Wake, M., Bliss, B., & Macdonald, R.E. (1994) Ototoxicity of carboplatin: comparing animal and clinical models at the hospital for sick children. J. Otolaryngol. 23:151-159.
29Maggirwar, SB, Dhanraj, DN, Somani, SM, & Ramkumar, V. (1994) Adenosine acts as an endogenous activator of the cellular antioxidant defence system. Biochem. Biophys. Res. Comm. 201:508-515.
30McFadden, S.L., Kasper, C., Ostrowski, J., Ding, D., & Salvi, R.J. (1998) Effects of inner hair cell loss on inferior colliculus evoked potential thresholds, amplitudes and forward masking functions in chinchillas. Hear. Res. 120:121-132.
31McGinness, J.E., Proctor, P.H., Demopoulos, H.B., Hokanson, J.A., & Kirkpatrick, D.S. (1978) Amelioration of cis-platinum nephrotoxicity by orgotein (superoxide dismutase). Physiol. Chem. Phys. 10:267-277.
32McKeage, M.J. (1995) Comparative adverse effect profiles of platinum drugs. Drug Saf. 13:228-244.
33Meijer, C., Mulder, N.H., Hospers, G.A., Uges, D.R., & de Vries, E.G. (1990) The role of glutathione in resistance to cisplatin in a human small cell lung cancer cell line. Br. J. Cancer 62:72-77.
34Meijer, C., Mulder, N.H., Timmer Bosscha, H., Sluiter, W.J., Meersma, G.J., & de Vries, E.G. (1992) Relationship of cellular glutathione to the cytotoxicity and resistance of seven platinum compounds. Cancer Res. 52:6885-6889.
35Meister, A. (1991) Glutathione deficiency produced by inhibition of its synthesis, and its reversal: Applications in research and therapy. Pharmacol. Ther. 51:155-194.
36Meister, A. (1995) Mitochondrial changes associated with glutathione deficiency. Biochmica Biophysica Acta 1271:35-42.
37Mellish, K.J., Kelland, L.R., & Harrap, K.R. (1993) In vitro platinum drug chemosensitivity of human cervical squamous cell carcinoma cell lines with intrinsic and acquired resistance to cisplatin. Br. J. Cancer 68:240-250.
38Mistry, P., Kelland, L.R., Abel, G., Sidhar, S., & Harrap, K.R. (1991) The relationships between glutathione, glutathione-S-transferase and cytotoxicity of platinum drugs and melphalan in eight human ovarian carcinoma cell lines. Br. J. Cancer 64:215-220.
39Mistry, P., Loh, S.Y., Kelland, L.R., & Harrap, K.R. (1993) Effect of buthionine sulfoximine on PtII and PtIV drug accumulation and the formation of glutathione conjugates in human ovarian-carcinoma cell lines. Int. J. Cancer 55:848-856.
40Murray, G.I., Burke, M.D., & Ewen, S.W.B. (1986) Glutathione localization by a novel o-phtalaldehyde histofluorescence method. Histochem. J.18:434-440.
41Ravi, R., Somani, S.M., & Rybak, L.P. (1995) Mechanism of cisplatin ototoxicity: antioxidant system. Pharmacol. Toxicol. 76:386-394.
42Russo, A., Carmichael, J., Friedman, N., DeGraff, W., Tochner, Z., Glatstein, E., & Mitchell, J.B. (1986) The roles of intracellular glutathione in antineoplastic chemotherapy. Int. J. Radiat. Oncol. Biol. Phys. 12:1347-­1354.
43Rybak, L.P., Ravi, R., & Somani, S.M. (1995) Mechanism of protection by diethyldithiocarbamate against cisplatin ototoxicity: antioxidant system. Fund. Appl. Toxicol. 26:293-300.
44Schindler, R.A., Gladstone, H.B., Scott, N., Hradek, G.T., Williams, H., & Shah, S.B. (1995) Enhanced preservation of the auditory nerve following cochlear perfusion with nerve growth factors. Am. J. Otol. 16:304-309.
45Schweitzer, V., Rarey, K., Dolon, D., Abrams, G., Litterst, C., & Sheridan, C. (1986) Ototoxicity of cisplatin versus platinum analogues CBDCA (JM-8) and CHIP (JM-9). Otolaryngol. Head Neck Surgery 94:458-470.
46Singh, S.V., Xu, B.H., Jani, J.P., Emerson, E.O., Backes, M.G., Rihn, C., Scalamogna, D., Stemmler, N., Specht, S., Blanock, K.., et al. (1995) Mechanism of cross-resistance to cisplatin in a mitomycin C-resistant human bladder cancer cell line. Int. J. Cancer 6:431-436.
47Spoendlin, H. (1976) Anatomical changes following various noise exposures. In Effects of Noise on Hearing. Henderson, D., Hamernik, R.P. & Mills, J.H.. eds. Raven Press, New York, pp69-90.
48Takeno, S., Harrison, R.V., Ibrahim, D., Wake, M., & Mount, R.J. (1994a) Cochlear function after selective inner hair cell degeneration induced by carboplatin. Hear. Res. 75:93-102.
49Takeno, S., Harrison, R.V., Mount, R.J., Wake, M., & Harada, Y. (1994b) Induction of selective inner hair cell damage by carboplatin. Scanning Microsc. 8:97-106.
50Taudy, M., Syka, J., Popelar, J., & Ulehlova, L. (1992) Carboplatin and cisplatin ototoxicity in guinea pigs. Audiol. 31:293-299.
51Tonetti, M., Giovine, M., Gasparini, A., Benatti, U., & De Flora, A. (1993) Enhanced formation of reactive species from cis-diammine-(1,1-cyclobutanedicarboxylato)­platinum(II) (carboplatin) in the presence of oxygen free radicals. Biochem. Pharmacol. 46:1377-1383.
52Trautwein, P., Hofstetter, P., Wang, J., Salvi, R., & Nostrant, A. (1996) Selective inner hair cell loss does not alter distortion product otoacoustic emissions. Hear. Res. 96:71-82.
53Wake, M., Takeno, S., Ibrahim, D., & Harrison, R (1994) Selective inner hair cell ototoxicity induced by carboplatin. Laryngosc. 104:488-493.
54Wake, M., Takeno, S., Ibrahim, D., Harrison, R., & Mount, R. (1993) Carboplatin ototoxicity: an animal model. J. Laryngol. Otol. 107:585-589.
55Walker, E.M.J., Gale, G.R. (1981) Methods of reduction of cisplatin nephrotoxicity. Ann. Clin. Lab. Sci. 11:397-410.
56Yamane, H., Nakai, Y., Takayama, M., Konishi, K., Iguchi, H., Nakagawa, T., Shibata, S., Kato, A., Sunami, K., & Kawakatsu, C. (1995a) The emergence of free radicals after acoustic trauma and strial blood flow. Acta Oto­Laryngologica-Suppl. 519:87-92.
57Yamane, H., Nakai, Y., Takayama, M., Iguchi, H., Nakagawa, T., & Kojima, A. (1995b) Appearance of free radicals in the guinea pig inner ear after noise-induced acoustic trauma. Eur. Arch. Oto-Rhino-Laryngol. 252:504-8, 1995.