- Email: [email protected]
Glutathione limits noise-induced hearing loss
Hearing Research 146 (2000) 28^34 www.elsevier.com/locate/heares
Glutathione limits noise-induced hearing loss Yoshimitsu Ohinata a
a;b
, Tatsuya Yamasoba c , Jochen Schacht a , Josef M. Miller
a;
*
Kresge Hearing Research Institute, University of Michigan, 1301 East Ann Street, Ann Arbor, MI, 48109-0506, USA b Department of Otolaryngology, Osaka Medical College, Takatsuki, Osaka 569-8686, Japan c Department of Otolaryngology, University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-8655, Japan Received 15 October 1999; accepted 24 April 2000
Abstract The generation of reactive oxygen species (ROS) is thought to be part of the mechanism underlying noise-induced hearing loss (NIHL). Glutathione (GSH) is an important cellular antioxidant that limits cell damage by ROS. In this study, we investigated the effectiveness of a GSH supplement to protect GSH-deficient animals from NIHL. Pigmented guinea pigs were exposed to a 4 kHz octave band noise, 115 dB SPL, for 5 h. Group 1 had a normal diet, while groups 2, 3 and 4 were fed a 7% low protein diet (leading to lowered tissue levels of GSH) for 10 days prior to noise exposure. One hour before, immediately after and 5 h after noise exposure, subjects received either an intraperitoneal injection of 5 ml/kg body weight of 0.9% NaCl (groups 1 and 2), 0.4 M glutathione monoethyl ester (GSHE; group 3) or 0.8 M GSHE (group 4). Auditory thresholds were measured by evoked brain stem response at 2, 4, 8, 12, 16 and 20 kHz before and after noise exposure. Ten days post exposure, group 1 showed noise-induced threshold shifts of approximately 20 dB at 2, 16 and 20 kHz and 35 to 40 dB at other frequencies. Threshold shifts in group 2 were significantly greater than baseline at 2, 4, 16 and 20 kHz. GSHE supplementation in a dose-dependent fashion attenuated the threshold shifts in the low protein diet animals. Hair cell loss, as evaluated with cytocochleograms, was consistent with the auditory-evoked brainstem response results. Group 2 exhibited significantly more hair cell loss than any of the other groups; hair cell loss in group 3 was similar to that seen in group 1; group 4 showed less loss than group 1. These results indicate that GSH is a significant factor in limiting noiseinduced cochlear damage. This is compatible with the notion that ROS generation plays a role in NIHL and that antioxidant treatment may be an effective prophylactic intervention. ß 2000 Elsevier Science B.V. All rights reserved. Key words: Reactive oxygen species ; Antioxidant; Noise; Hearing loss; Diet; Guinea pig
1. Introduction A variety of mechanisms have been proposed to account for hearing loss following high intensity sound exposure. They are classi¢ed into two main categories: (1) direct mechanical trauma to the organ of Corti and (2) `metabolic stress' through increased oxidative metabolism in the inner ear. One important factor associated with tissue destruction through metabolic stress is the generation of reactive oxygen species (ROS). ROS formation results from increased mitochondrial activity
* Corresponding author. Tel.: +1 (734) 764 8110; Fax: +1 (734) 764 0014; E-mail: [email protected]
and may also follow changes in cochlear blood £ow associated with intense sound exposure (Lamm and Arnold, 1996; Scheibe et al., 1990; Thorne and Nuttall, 1987). The importance of ROS in noise-induced hearing loss (NIHL) and tissue damage is supported by ¢ndings that: (1) superoxide anion radicals emerge in the stria vascularis after intense sound exposure (Yamane et al., 1995); (2) hydroxyl radical levels signi¢cantly increase in the cochlea following intense sound exposure (Ohlemiller and Dugan, 1998); (3) conditioning noise exposure which reduces NIHL increases the activity of some antioxidant enzymes (Jacono et al., 1997), (4) reduction in the endogeneous antioxidant, glutathione (GSH, Lglutamyl-L-cysteinylglycine), increases NIHL (Yamasoba et al., 1998b), (5) GSH is upregulated in the lateral wall following noise exposure (Yamasoba et al., 1998a) and (6) NIHL is attenuated by a variety of antioxidant
0378-5955 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 0 0 ) 0 0 0 9 6 - 4
HEARES 3474 17-7-00
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
interventions. The latter include: superoxide dismutasepolyethylene glycol, a scavenger of ROS, and allopurinol, a blocker of ROS production and potential scavenger of ROS (Seidman et al., 1993); lazaroids, lipid peroxidation inhibitors and ROS scavengers (Quirk et al., 1994); R-phenylisopropyl adenosine, an adenosine analogue e¡ective in upregulating antioxidant enzyme activity levels (Hu et al., 1996); and the antioxidant mannitol and the iron chelator deferoxamine mesylate (Yamasoba et al., 1999). GSH is the most abundant non-protein thiol in mammalian cells. It serves as an antioxidant by reacting with ROS either directly or in reactions catalyzed by GSH peroxidases or GSH transhydrogenases (Meister and Anderson, 1983 ; Meister, 1991; Orrenius and Molde¨us, 1984). GSH may be especially important for organs that are exposed to exogenous toxins or oxidative stress. In the inner ear, substantial amounts of GSH are normally present (Lautermann et al., 1997). GSH immunoreactivity is preferentially distributed in the stria vascularis, spiral ligament and vestibular end organs (Usami et al., 1996). This GSH distribution overlaps with that of GSH S-transferase, whose activity is highest in the lateral wall tissues, intermediate in the neurosensory epithelium, and lowest in the modiolus (El Barbary et al., 1993). Compared to brain tissues GSH Stransferase activity is much higher in cochlear lateral wall and neurosensory epithelium. Increasing cochlear GSH may protect against gentamicin ototoxicity (Garetz et al., 1994a, 1994b), whereas reduction of GSH enhances hearing loss induced by gentamicin (Lautermann et al., 1995a), cisplatin (Lautermann and Schacht, 1996) or the combination of aminoglycosides and ethacrynic acid (Ho¡man et al., 1987, 1988 ; Lautermann and Schacht, 1996). Cisplatin-induced tissue damage in the inner ear correlates with a decrease in GSH levels following cisplatin administration (Ravi et al., 1995). Glutathione monoethyl ester (GSHE) is an e¡ective delivery agent for GSH. It is transported into cells and hydrolyzed, presumably by intracellular esterases, to GSH and ethyl alcohol (Puri and Meister, 1983 ; Anderson and Meister, 1989). In contrast, GSH itself does not readily enter cells. After intraperitoneal injections into mice, GSHE e¡ectively increased the GSH levels in liver and kidney (Puri and Meister, 1983). Likewise, GSH levels in the cochlea that were decreased by a low protein diet could be restored by supplementary administration of GSHE (Lautermann et al., 1995b). We examine here the in£uence of a low protein diet on auditory threshold shifts following noise exposure and the e¡ect of dietary GSHE supplementation.
29
2. Materials and methods 2.1. Experimental groups Pigmented guinea pigs (250 to 300 g) were obtained from Murphy Breeding Laboratories (Plain¢eld, NJ, USA). Since sex di¡erences have been associated with a di¡ering ability to detoxify ROS (Julicher et al., 1984) and varying degrees of GSH S-transferase activity in the cochlea (El Barbary et al., 1993), only male guinea pigs were used. The animals were on a normal day/night cycle. The experimental protocol was approved by the Animal Care and Use Committee at the University of Michigan and conforms to the NIH Guidelines for the Care and Use of Laboratory Animals. 2.2. Dietary and experimental schedule Animals with a normal Preyer's re£ex were assigned to one of four groups (¢ve animals/group). Group 1 was kept on a regular guinea pig diet (18.5% protein: Purina 5025; St. Louis, MO, USA), while groups 2, 3 and 4 were fed a low protein diet (7% protein : TD 93177 ; Harlan Teklad, Madison, WI, USA) for the duration of the study. Auditory-evoked brainstem responses (ABR) were measured for both ears. The ¢rst (pre-exposure) ABR was taken on day 8 of the feeding schedule. Noise exposure was on day 11. A second (post-exposure) ABR was taken on day 21 and animals were subsequently sacri¢ced for assessment of cochlear pathology. 2.3. Auditory brainstem responses Animals were anesthetized with a mixture of xylazine (10 mg/kg) and ketamine (40 mg/kg) given intramuscularly. A di¡erential active needle electrode was placed subcutaneously below the test ear and a reference electrode at the vertex. A ground electrode was positioned below the contralateral ear. The sound stimulus consisted of a 15 ms tone burst, with a rise^fall time of 1 ms at frequencies of 2, 4, 8, 12, 16 and 20 kHz. The sound intensity was varied in 5 dB intervals. One thousand and twenty-four tone presentations given at the rate of 9 s31 were averaged using a microcomputer and custom software to obtain a waveform. Hearing threshold was de¢ned by the consistent appearance of ABR peaks 3 or 4. Each ear was monitored separately. The threshold was averaged between ears for each frequency for each animal. The average from both ears was used for all analyses. Animals demonstrating an initial threshold of more than 25 dB SPL were excluded from this study.
HEARES 3474 17-7-00
30
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
mize GSH at critical times of ROS formation. Noiseinduced ROS formation can be detected immediately and 2 h after a 3 h exposure (Yamane et al., 1995) and endogenous GSH levels increase 4 h after a 5 h noise exposure (Yamasoba et al., 1998a). Levels of GSH in liver and kidney peak 2 h after the intraperitoneal administration of GSHE and return to baseline 8 h thereafter (Meister, 1991). 2.6. Histological examination
Fig. 1. Power analysis of 4 kHz octaveband noise used in every 1/3 octave.
2.4. Noise exposure Animals were exposed to noise in a lighted and ventilated sound exposure chamber while having free access to food and water. The sound chamber was ¢tted with speakers driven by a noise generator and a power ampli¢er. Using a 12 inch Bruel and Kjear condenser microphone and Fast Fourier Transform analyzer, sound levels were measured and calibrated at multiple locations within the sound chamber to ensure uniformity of the stimulus. The stimulus intensity varied by a maximum of 3 dB across measured sites within the exposure chamber. The spectral power analysis of noise, in 1/3 octave steps, is shown in Fig. 1. Animals were individually subjected to 4 kHz octave band noise, 115 dB SPL for 5 h. From each group, two animals were exposed beginning at 9 a.m. and three animals beginning at 4 p.m. to reduce the in£uence of circadian variations in GSH levels on the resulting threshold shifts. 2.5. Administration of GSHE GSHE (Kyowa Hakko Kogyo Co.) was administered 1 h before, immediately after and 5 h after noise exposure. At these times, control subjects received an intraperitoneal injection of 5 ml of 0.9% NaCl/kg body weight (groups 1 and 2), experimental subjects received 5 ml of 0.4 M GSHE/kg (group 3) or 5 ml of 0.8 M GSHE/kg (group 4). GSHE was dissolved in 0.9% NaCl. The GSHE dosing schedule was designed to maxi-
Animals were killed under deep anesthesia after the ¢nal ABR measurements. The temporal bones were immediately removed and the perilymphatic spaces perfused and then immersed for 1 h in 4% paraformaldehyde in 0.01 M phosphate-bu¡ered saline (pH 7.4). After permeabilization with 0.3% Triton X-100 for 5 min, whole-mounts of the organ of Corti were stained for actin with £uorescent phalloidin for 30 min to outline hair cells and their stereocilia for a quantitative assessment (Raphael and Altschuler, 1991). We note that phalloidin does not stain bodies of the hair cells. The slide preparations were observed under £uorescence microscopy and missing (as shown by scar formation) and present (as shown by intact stereocilia) hair cells in the sensory epithelium were counted from the apex to the base in 0.19 mm segments. The percentage of hair cell loss was calculated at each observation point in each animal and the average at each observation point was plotted along the cochlea to create an average cytocochleogram. This average represents the percent (ratio of scar to intact cells divided by total cells) per 0.19 mm. Statistical di¡erences were evaluated for signi¢cance on the basis of data from the region of 0^14 mm from apex in the cochlea. 2.7. Statistical analysis ABR measurements were obtained from both ears of each animal and were averaged to yield one data point (threshold shift) per animal. The threshold shifts at each frequency, as well as the percentage of missing outer and inner hair cells, were compared among the groups using a one-way analysis of variance (ANOVA). Post-ANOVA comparisons of means were performed using the Fisher method. Statistical signi¢cance was de¢ned as P 6 0.05. 3. Results 3.1. Weight gain There was no signi¢cant di¡erence in the animals' body weight at the start of the experiment. On a full
HEARES 3474 17-7-00
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
31
3.2. Threshold shifts following noise exposure
Fig. 2. Weight gain of experimental animals. There was no signi¢cant di¡erence in the animals' body weight at the start of the experiment. Values represent mean þ S.D. (n = 5). The weight of subjects on a low protein diet was signi¢cantly di¡erent from group 1 after 10 days (P 6 0.05). GSHE did not reverse the e¡ects of low protein diet on weight. *Signi¢cantly di¡erent from group 1 (P 6 0.05).
protein diet (group 1) body weight increased signi¢cantly throughout the course of the study. The weight of subjects on a low protein diet (groups 2, 3 and 4) remained stable and was signi¢cantly di¡erent from group 1 after 10 days (P 6 0.05). GSHE did not reverse the e¡ects of low protein diet on weight (Fig. 2).
There was no signi¢cant di¡erence in the auditory thresholds between any of the groups before treatment. Thresholds averaged 15 to 25 dB over the range of tested frequencies. Ten days after noise exposure, group 1 exhibited an average threshold shift of approximately 20 dB at 2, 16 and 20 kHz and 35 to 40 dB at other frequencies. Threshold shifts in animals on the low protein diet (group 2) were signi¢cantly greater than baseline at 2, 4, 16 and 20 kHz (P 6 0.05). GSHE supplementation reduced the extent of threshold shifts in a dose-dependent fashion. There was signi¢cant attenuation at 2, 4 and 16 kHz in group 3 and at all frequencies in group 4 (P 6 0.05 ; Fig. 3). 3.3. Histological ¢ndings Whole-mount analysis of the organ of Corti showed that peak damage to the outer (Fig. 4a) and inner (Fig. 4b) hair cells was at a region 10 to 12 mm from apex. Hair cell loss in group 2 was signi¢cantly higher than in the other groups (Fig. 5a,b; P 6 0.05) and also more wide-spread, with a secondary peak of loss at around 5 mm from the apex (Fig. 4a,b). There were no di¡erences among groups 1, 3 and 4.
Fig. 3. E¡ect of low-protein diet and GSHE on threshold shifts. Animals were treated and ABR thresholds measured as described in Section 2. Values represent mean þ S.D. (n = 5). The threshold shifts were signi¢cantly di¡erent (P 6 0.05) between groups 1 and 2 at 2, 4, 16 and 20 kHz. Threshold shifts in group 3 (at 2, 4 and 16 kHz) and group 4 (at all frequencies) were signi¢cantly smaller than in group 2 (P 6 0.05)*.
HEARES 3474 17-7-00
32
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
4. Discussion GSH is an important factor in the modi¢cation of inner ear trauma following noise exposure. This is shown in this study by two complimentary approaches. Under conditions where cochlear GSH is lowered (lowprotein diet) threshold shift and hair cell loss is aggravated. When GSH levels were restored by dietary supplementation, threshold shifts and hair cell loss were reduced. A low-protein diet is a convenient way to lower GSH levels while maintaining the general health of the animals. The diet provides less of the amino acid L-cysteine, a rate-limiting substrate in GSH synthesis (DeLeve and Kaplowitz, 1991). The diet decreases the level of GSH in the cochlear sensory epithelium by about 50% compared to animals on a normal protein diet (Lautermann et al., 1995b). At an operational level it
Fig. 5. Hair cell loss in the organ of Corti after noise exposure (a, outer hair cells; b, inner hair cells). (Group 1: n = 5, group 2: n = 4, group 3: n = 3, group 4: n = 5.) *Signi¢cantly di¡erent from group 2 (P 6 0.05).
Fig. 4. Average cytocochleogram in each group after noise exposure. Animals were treated and cytocochleograms produced as described in Section 2. Hair cell loss at 10 days after noise exposure is averaged for outer (a) and inner (b) hair cells. Predominant damage was observed in the region 10 to 12 mm from apex. Hair cell loss in group 2 is greater and more wide-spread than in the other groups. (Group 1: n = 5, group 2: n = 4, group 3: n = 3, group 4: n = 5.)
is appropriate to interpret the current data to indicate that GSH reverses the e¡ect of low-protein diet. However, GSHE does not reduce other general systemic e¡ects including unchanged body weight. It does reverse GSH levels in the inner ear (Lautermann et al., 1995a). Therefore, we believe these data indicate that GSH limits NIHL. The pattern of threshold shift and hair cell loss re£ects the exposure parameters. In animals on a normal diet, damage was greater at 4, 8 and 12 kHz than at 2, 16 and 20 kHz. The threshold shifts and hair cell loss in the animals on the low protein diet were greater and broader across frequencies. This suggests an increased sensitivity that extends from peak frequencies to structures at the `edge' of the noise exposure ¢eld. The supplement of GSH in the animals on the low protein diet was most e¡ective at the frequencies furthest from the 4 kHz center frequency. The result may indicate that the GSH supplement is inadequate to prevent damage if the level of stress and ROS exceed certain limits. Alternatively, for tissues at the center frequency of the exposure, other factors, in addition to
HEARES 3474 17-7-00
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
ROS formation, may have contributed to the level of destruction. While the noise exposure produced a permanent threshold shift of 20 to 35 dB in groups 1, 3 and 4, the extent of missing inner and outer hair cells was limited and mild. Such a discrepancy is not unusual. Yamasoba et al. (1999) have also found signi¢cant threshold shifts in the presence of mild hair cell loss in guinea pigs exposed to noise. Liberman and Gao (1995) likewise reported that guinea pigs with approximately 40 dB hearing loss showed minimal hair cell loss. Hunter-Duvar and Elliot (1973) even found no observable hair cell loss in squirrel monkey cochleae despite threshold shifts greater than 40 dB. However, species di¡erences may exist in the relationship between hearing loss and hair cell damage. For example, Hamernik et al. (1989) suggested that missing outer hair cells were a major contributor to the ¢rst 30^40 dB of threshold shift in chinchillas and that loss of outer hair cells was almost complete when permanent threshold shifts were more than 40 dB. Subtle changes in the structure of hair cells and other cochlear tissues that are not evident from hair cell counts may have contributed to noise-induced threshold shifts. For example, acoustic overstimulation leads to excessive release of extracellular excitatory amino acid and destruction of the dendrites beneath the inner hair cells (Puel et al., 1996). Such potential changes were not evaluated in the present study. In any case, the increased threshold shifts and hair cell loss at low GSH concentrations are consistent with our previous ¢ndings that drug-inhibition of GSH synthesis enhances NIHL (Yamasoba et al., 1998a). They are also strongly supportive of an involvement of ROS in noise trauma, as suggested by dramatically increased lipid peroxidation under similar conditions (Ohinata et al., 1999). The source of the ROS remains speculative. As in other tissues, ROS in the overstimulated inner ear may be produced by intensi¢ed oxidative processes in mitochondria. Acoustic stimulation indeed increases cochlear energy metabolism (Canlon and Schacht, 1983 ; Ryan et al., 1982). In addition, ROS are generated in tissues in response to prolonged hypoxia or after reperfusion following ischemic periods. Studies utilizing laser Doppler £owmetry (Lamm and Arnold, 1996 ; Scheibe et al., 1990; Thorne and Nuttall, 1987) and intravital microscopy (Lamm and Arnold, 1996; Quirk et al., 1991 ; Quirk et al., 1992; Quirk and Seidman, 1995) have demonstrated reduced cochlear blood £ow during intense sound exposure. Intense sound exposure may also cause vascular alterations such as vasoconstriction, aggregations of red blood cells, increased vascular permeability and localized edema (Axelsson and Dengerink, 1987; Hawkins, 1971; Hultcrantz et al.,
33
1979 ; Yamane et al., 1991 ; Yamane et al., 1995). The combined facts support the idea that localized ischemia and the subsequent release of ROS contribute to the cochlear damage in noise trauma. In conclusion, GSHE reverses the e¡ects of the low protein diet on NIHL. It elevates GSH in the inner ear tissues, reversing the diet-induced reduction in the inner ear tissue GSH. These ¢ndings indicate that cochlear GSH level is involved in NIHL and is consistent with the hypothesis that the generation of ROS is a mechanism of NIHL. Acknowledgements A preliminary report was presented at the 22nd Midwinter Meeting of the Association for Research in Otolaryngology (February 13^18, 1999, at St. Petersburg Beach, FL, USA). We thank Dr. Richard A. Altschuler, Dr. David F. Dolan and Ms. J. Nadine Brown for their valuable help. This work was supported by NIH research grant DC-03685 from the National Institute of Deafness and Communication Disorders, National Institutes of Health.
References Anderson, M.E., Meister, A., 1989. Glutathione monoesters. Anal. Biochem. 183, 16^20. Axelsson, A., Dengerink, H., 1987. The e¡ects of noise on histological measures of the cochlear vasculature and red blood cells: a review. Hear. Res. 31, 183^192. Canlon, B., Schacht, J., 1983. Acoustic stimulation alters deoxyglucose uptake in the mouse cochlea and inferior colliculus. Hear. Res. 10, 217^226. DeLeve, L.D., Kaplowitz, N., 1991. Glutathione metabolism and its role in hepatotoxicity. Pharmacol. Ther. 52, 287^305. El Barbary, A., Altschuler, R., Schacht, J., 1993. Glutathione S-transferases in the organ of Corti of the rat: enzymatic activity, subunit composition and immunohistological localization. Hear. Res. 71, 80^90. Garetz, S.L., Altschuler, R.A., Schacht, J., 1994a. Attenuation of gentamicin ototoxicity by glutathione in the guinea pig in vivo. Hear. Res. 77, 81^87. Garetz, S.L., Rhee, D.J., Schacht, J., 1994b. Sulfhydryl compounds and antioxidants inhibit cytotoxicity to outer cells of a gentamicin metabolite in vitro. Hear. Res. 77, 75^80. Hamernik, R.P., Patterson, J.H., Turrentine, G.A., Ahroon, W.A., 1989. The quantitative relationship between sensory cell loss and hearing thresholds. Hear. Res. 38, 199^212. Hawkins, J.E., 1971. The role of vasoconstriction in noise-induced hearing loss. Ann. Otol. Rhinol. Laryngol. 80, 903^913. Ho¡man, D.W., Whitworth, C.A., Jones, K.L., Rybak, L.P., 1987. Nutritional status, glutathione levels and ototoxicity of loop diuretics and aminoglycoside antibiotics. Hear. Res. 31, 217^222. Ho¡man, D.W., Whitworth, C.A., Jones-King, K.L., Rybak, L.P., 1988. Potentiation of ototoxicity by glutathione depletion. Ann. Otol. Rhinol. Laryngol. 97, 36^41. Hu, B.H., Kopke, R., Zheng, X.Y., Henderson, D., 1996. E¡ect of
HEARES 3474 17-7-00
34
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
R-phenylisopropyl adenosine on noise-induced hearing loss. Assoc. Res. Otolaryngol. Abstr. 19, 802. Hultcrantz, E., Angelborg, C., Beausant-Linder, M., 1979. Noise and cochlear blood £ow. Arch. Otorhinolaryngol. 224, 103^106. Hunter-Duvar, I.M., Elliot, D.N., 1973. E¡ects of intense auditory stimulation: hearing losses and inner ear changes in the squirrel monkey II. J. Acoust. Soc. Am. 54, 1179^1183. Jacono, A.A., Kopke, R., Vugmeyster, L., Steinman, H., Rowe, J., Hu, B.H., Van De Water, T.R., Henderson, D., 1997. Changes in cochlear antioxidant enzyme activity levels after conditioning noise exposure in the chinchilla. Assoc. Res. Otolaryngol. Abstr. 20, 132. Julicher, R.H., Sterrenberg, L., Haenen, G.R., Bast, A., Noordhoek, J., 1984. Sex di¡erences in the cellular defense system against free radicals from oxygen or drug metabolites in rat. Arch. Toxicol. 56, 83^86. Lamm, K., Arnold, W., 1996. Noise-induced cochlear hypoxia is intensity dependent, correlates with hearing loss and precedes reduction of cochlear blood £ow. Audiol. Neurootol. 1, 148^160. Lautermann, J., McLaren, J., Schacht, J., 1995a. Glutathione protection against gentamicin ototoxicity depends on nutritional status. Hear. Res. 86, 15^24. Lautermann, J., Schacht, J., 1996. A sensitive animal model to assess acute and chronic ototoxic e¡ects. Arch. Otolaryngol. Head Neck Surg. 122, 837^840. Lautermann, J., Song, B., McLaren, J., Schacht, J., 1995b. Diet is a risk factor in cisplatin ototoxicity. Hear. Res. 88, 47^53. Lautermann, J., Crann, S.A., McLaren, J., Schacht, J., 1997. Glutathione-dependent antioxidant systems in the mammalian inner ear: e¡ects of aging, ototoxic drugs and noise. Hear. Res. 114, 75^82. Liberman, M.C., Gao, W.Y., 1995. Chronic cochlear de-e¡erentation and susceptibility to permanent acoustic injury. Hear. Res. 90, 158^168. Meister, A., 1991. Glutathione de¢ciency produced by inhibition of its synthesis, and its reversal: applications in research and therapy. Pharm. Ther. 51, 155^194. Meister, A., Anderson, M.E., 1983. Glutathione. Ann. Rev. Biochem. 52, 711^760. Ohinata, Y., Miller, J.M., Brown, N., Altschuler, R.A., Yamasoba, T., Schacht, J., 1999. Role of reactive oxygen molecules (ROM) in noise induced hearing loss (NIHL). Inner Ear Biol. Abstr. 36, C11. Ohlemiller, K.K., Dugan, L.L., 1998. In vivo measurement of cochlear reactive oxygen species (ROS) in mice: E¡ects of noise exposure and cochlear ischemia. Assoc. Res. Otolaryngol. Abstr. 21, 518. Orrenius, S., Molde¨us, P., 1984. The multiple roles of glutathione in drug metabolism. Trends. Pharmacol. Sci. 6, 432^435. Puel, J.-L., Gervais d'Aldin, C., Sa¤edine, S., Eybalin, M., Pujol, R., 1996. Excitotoxicity and plasticity of IHC-auditory nerve contributes to both temporary and permanent threshold shift. In: Axelsson, A., Borchgrevink, H., Hamernik, R., Hellstro«m, P., Hender-
son, D., Salvi, R. (Eds.), Scienti¢c Basis of Noise-Induced Hearing Loss. Thieme Medical Press, New York, pp. 36^42. Puri, R.N., Meister, A., 1983. Transport of glutathione, as gammaglutamylcysteinylglycyl ester, into liver and kidney. Proc. Natl. Acad. Sci. USA 80, 5258^5260. Quirk, W.S., Avanish, G., Nuttall, A.L., Miller, J.M., 1992. The in£uence of loud sound on red blood cell velocity and blood cell diameter in the cochlea. Hear. Res. 63, 102^107. Quirk, W.S., Seidman, M.D., 1995. Cochlear vascular changes in response to loud noise. Am. J. Otol. 16, 322^325. Quirk, W.S., Shapiro, B.D., Miller, J.M., Nuttall, A.L., 1991. Noiseinduced changes in red blood cell velocity in lateral wall vessels of rat cochlea. Hear. Res. 52, 217^224. Quirk, W.S., Shivapuja, B.G., Schwimmer, C.L., Seidman, M.D., 1994. Lipid peroxidation inhibitor attenuates noise-induced temporary threshold shifts. Hear. Res. 74, 217^220. Raphael, Y., Altschuler, R.A., 1991. Scar formation after drug-induced cochlear insult. Hear. Res. 31, 173^184. Ravi, R., Somani, S.M., Rybak, L.P., 1995. Mechanism of cisplatin ototoxicity: antioxidant system. Pharmacol. Toxicol. 76, 386^394. Ryan, A.F., Goodwin, P., Woolf, N.K., Sharp, F.R., 1982. Auditory stimulation alters the pattern of 2-deoxyglucose uptake in the inner ear. Brain Res. 234, 213^225. Scheibe, F., Haupt, H., Nuttall, A.L., Ludwig, C., 1990. Laser Doppler measurements of cochlear blood £ow during loud sound presentation. Arch. Otorhinolaryngol. 247, 84^88. Seidman, M.D., Shivapuja, B.G., Quirk, W.S., 1993. The protective e¡ects of allopurinol and superoxide dismutase on noise-induced cochlear damage. Otolaryngol. Head Neck Surg. 109, 1052^1056. Thorne, P., Nuttall, A.L., 1987. Laser Doppler measurements of cochlear blood £ow during loud sound exposure in the guinea pig. Hear. Res. 27, 1^10. Usami, S., Hjelle, O.P., Ottersen, O.P., 1996. Di¡erential cellular distribution of glutathione, an endogenous antioxidant, in the guinea pig inner ear. Brain Res. 743, 337^340. Yamane, H., Nakai, Y., Konishi, K., Sakamoto, H., Matsuda, Y., Iguchi, H., 1991. Strial circulation impairment due to acoustic trauma. Acta Otolaryngol. (Stockh.) 111, 85^93. Yamane, H., Nakai, Y., Takayama, M., Iguchi, H., Nakagawa, T., Kojima, A., 1995. Appearance of free radicals in the guinea pig inner ear after noise-induced acoustic trauma. Eur. Arch. Otorhinolaryngol. 252, 504^508. Yamasoba, T., Harris, C., Shoji, F., Lee, R.J., Nuttall, A.L., Miller, J.M., 1998a. In£uence of intense sound exposure on glutathione synthesis in the cochlea. Brain Res. 804, 72^78. Yamasoba, T., Nuttall, A.L., Harris, C., Raphael, Y., Miller, J.M., 1998b. Role of glutathione in protection against noise-induced hearing loss. Brain Res. 784, 82^90. Yamasoba, T., Schacht, J., Shoji, F., Miller, J.M., 1999. Attenuation of cochlear damage from noise trauma by an iron chelator, a free radical scavenger and glial cell line-derived neurotrophic factor in vivo. Brain Res. 815, 317^325.
HEARES 3474 17-7-00
Glutathione limits noise-induced hearing loss Yoshimitsu Ohinata a
a;b
, Tatsuya Yamasoba c , Jochen Schacht a , Josef M. Miller
a;
*
Kresge Hearing Research Institute, University of Michigan, 1301 East Ann Street, Ann Arbor, MI, 48109-0506, USA b Department of Otolaryngology, Osaka Medical College, Takatsuki, Osaka 569-8686, Japan c Department of Otolaryngology, University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-8655, Japan Received 15 October 1999; accepted 24 April 2000
Abstract The generation of reactive oxygen species (ROS) is thought to be part of the mechanism underlying noise-induced hearing loss (NIHL). Glutathione (GSH) is an important cellular antioxidant that limits cell damage by ROS. In this study, we investigated the effectiveness of a GSH supplement to protect GSH-deficient animals from NIHL. Pigmented guinea pigs were exposed to a 4 kHz octave band noise, 115 dB SPL, for 5 h. Group 1 had a normal diet, while groups 2, 3 and 4 were fed a 7% low protein diet (leading to lowered tissue levels of GSH) for 10 days prior to noise exposure. One hour before, immediately after and 5 h after noise exposure, subjects received either an intraperitoneal injection of 5 ml/kg body weight of 0.9% NaCl (groups 1 and 2), 0.4 M glutathione monoethyl ester (GSHE; group 3) or 0.8 M GSHE (group 4). Auditory thresholds were measured by evoked brain stem response at 2, 4, 8, 12, 16 and 20 kHz before and after noise exposure. Ten days post exposure, group 1 showed noise-induced threshold shifts of approximately 20 dB at 2, 16 and 20 kHz and 35 to 40 dB at other frequencies. Threshold shifts in group 2 were significantly greater than baseline at 2, 4, 16 and 20 kHz. GSHE supplementation in a dose-dependent fashion attenuated the threshold shifts in the low protein diet animals. Hair cell loss, as evaluated with cytocochleograms, was consistent with the auditory-evoked brainstem response results. Group 2 exhibited significantly more hair cell loss than any of the other groups; hair cell loss in group 3 was similar to that seen in group 1; group 4 showed less loss than group 1. These results indicate that GSH is a significant factor in limiting noiseinduced cochlear damage. This is compatible with the notion that ROS generation plays a role in NIHL and that antioxidant treatment may be an effective prophylactic intervention. ß 2000 Elsevier Science B.V. All rights reserved. Key words: Reactive oxygen species ; Antioxidant; Noise; Hearing loss; Diet; Guinea pig
1. Introduction A variety of mechanisms have been proposed to account for hearing loss following high intensity sound exposure. They are classi¢ed into two main categories: (1) direct mechanical trauma to the organ of Corti and (2) `metabolic stress' through increased oxidative metabolism in the inner ear. One important factor associated with tissue destruction through metabolic stress is the generation of reactive oxygen species (ROS). ROS formation results from increased mitochondrial activity
* Corresponding author. Tel.: +1 (734) 764 8110; Fax: +1 (734) 764 0014; E-mail: [email protected]
and may also follow changes in cochlear blood £ow associated with intense sound exposure (Lamm and Arnold, 1996; Scheibe et al., 1990; Thorne and Nuttall, 1987). The importance of ROS in noise-induced hearing loss (NIHL) and tissue damage is supported by ¢ndings that: (1) superoxide anion radicals emerge in the stria vascularis after intense sound exposure (Yamane et al., 1995); (2) hydroxyl radical levels signi¢cantly increase in the cochlea following intense sound exposure (Ohlemiller and Dugan, 1998); (3) conditioning noise exposure which reduces NIHL increases the activity of some antioxidant enzymes (Jacono et al., 1997), (4) reduction in the endogeneous antioxidant, glutathione (GSH, Lglutamyl-L-cysteinylglycine), increases NIHL (Yamasoba et al., 1998b), (5) GSH is upregulated in the lateral wall following noise exposure (Yamasoba et al., 1998a) and (6) NIHL is attenuated by a variety of antioxidant
0378-5955 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 0 0 ) 0 0 0 9 6 - 4
HEARES 3474 17-7-00
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
interventions. The latter include: superoxide dismutasepolyethylene glycol, a scavenger of ROS, and allopurinol, a blocker of ROS production and potential scavenger of ROS (Seidman et al., 1993); lazaroids, lipid peroxidation inhibitors and ROS scavengers (Quirk et al., 1994); R-phenylisopropyl adenosine, an adenosine analogue e¡ective in upregulating antioxidant enzyme activity levels (Hu et al., 1996); and the antioxidant mannitol and the iron chelator deferoxamine mesylate (Yamasoba et al., 1999). GSH is the most abundant non-protein thiol in mammalian cells. It serves as an antioxidant by reacting with ROS either directly or in reactions catalyzed by GSH peroxidases or GSH transhydrogenases (Meister and Anderson, 1983 ; Meister, 1991; Orrenius and Molde¨us, 1984). GSH may be especially important for organs that are exposed to exogenous toxins or oxidative stress. In the inner ear, substantial amounts of GSH are normally present (Lautermann et al., 1997). GSH immunoreactivity is preferentially distributed in the stria vascularis, spiral ligament and vestibular end organs (Usami et al., 1996). This GSH distribution overlaps with that of GSH S-transferase, whose activity is highest in the lateral wall tissues, intermediate in the neurosensory epithelium, and lowest in the modiolus (El Barbary et al., 1993). Compared to brain tissues GSH Stransferase activity is much higher in cochlear lateral wall and neurosensory epithelium. Increasing cochlear GSH may protect against gentamicin ototoxicity (Garetz et al., 1994a, 1994b), whereas reduction of GSH enhances hearing loss induced by gentamicin (Lautermann et al., 1995a), cisplatin (Lautermann and Schacht, 1996) or the combination of aminoglycosides and ethacrynic acid (Ho¡man et al., 1987, 1988 ; Lautermann and Schacht, 1996). Cisplatin-induced tissue damage in the inner ear correlates with a decrease in GSH levels following cisplatin administration (Ravi et al., 1995). Glutathione monoethyl ester (GSHE) is an e¡ective delivery agent for GSH. It is transported into cells and hydrolyzed, presumably by intracellular esterases, to GSH and ethyl alcohol (Puri and Meister, 1983 ; Anderson and Meister, 1989). In contrast, GSH itself does not readily enter cells. After intraperitoneal injections into mice, GSHE e¡ectively increased the GSH levels in liver and kidney (Puri and Meister, 1983). Likewise, GSH levels in the cochlea that were decreased by a low protein diet could be restored by supplementary administration of GSHE (Lautermann et al., 1995b). We examine here the in£uence of a low protein diet on auditory threshold shifts following noise exposure and the e¡ect of dietary GSHE supplementation.
29
2. Materials and methods 2.1. Experimental groups Pigmented guinea pigs (250 to 300 g) were obtained from Murphy Breeding Laboratories (Plain¢eld, NJ, USA). Since sex di¡erences have been associated with a di¡ering ability to detoxify ROS (Julicher et al., 1984) and varying degrees of GSH S-transferase activity in the cochlea (El Barbary et al., 1993), only male guinea pigs were used. The animals were on a normal day/night cycle. The experimental protocol was approved by the Animal Care and Use Committee at the University of Michigan and conforms to the NIH Guidelines for the Care and Use of Laboratory Animals. 2.2. Dietary and experimental schedule Animals with a normal Preyer's re£ex were assigned to one of four groups (¢ve animals/group). Group 1 was kept on a regular guinea pig diet (18.5% protein: Purina 5025; St. Louis, MO, USA), while groups 2, 3 and 4 were fed a low protein diet (7% protein : TD 93177 ; Harlan Teklad, Madison, WI, USA) for the duration of the study. Auditory-evoked brainstem responses (ABR) were measured for both ears. The ¢rst (pre-exposure) ABR was taken on day 8 of the feeding schedule. Noise exposure was on day 11. A second (post-exposure) ABR was taken on day 21 and animals were subsequently sacri¢ced for assessment of cochlear pathology. 2.3. Auditory brainstem responses Animals were anesthetized with a mixture of xylazine (10 mg/kg) and ketamine (40 mg/kg) given intramuscularly. A di¡erential active needle electrode was placed subcutaneously below the test ear and a reference electrode at the vertex. A ground electrode was positioned below the contralateral ear. The sound stimulus consisted of a 15 ms tone burst, with a rise^fall time of 1 ms at frequencies of 2, 4, 8, 12, 16 and 20 kHz. The sound intensity was varied in 5 dB intervals. One thousand and twenty-four tone presentations given at the rate of 9 s31 were averaged using a microcomputer and custom software to obtain a waveform. Hearing threshold was de¢ned by the consistent appearance of ABR peaks 3 or 4. Each ear was monitored separately. The threshold was averaged between ears for each frequency for each animal. The average from both ears was used for all analyses. Animals demonstrating an initial threshold of more than 25 dB SPL were excluded from this study.
HEARES 3474 17-7-00
30
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
mize GSH at critical times of ROS formation. Noiseinduced ROS formation can be detected immediately and 2 h after a 3 h exposure (Yamane et al., 1995) and endogenous GSH levels increase 4 h after a 5 h noise exposure (Yamasoba et al., 1998a). Levels of GSH in liver and kidney peak 2 h after the intraperitoneal administration of GSHE and return to baseline 8 h thereafter (Meister, 1991). 2.6. Histological examination
Fig. 1. Power analysis of 4 kHz octaveband noise used in every 1/3 octave.
2.4. Noise exposure Animals were exposed to noise in a lighted and ventilated sound exposure chamber while having free access to food and water. The sound chamber was ¢tted with speakers driven by a noise generator and a power ampli¢er. Using a 12 inch Bruel and Kjear condenser microphone and Fast Fourier Transform analyzer, sound levels were measured and calibrated at multiple locations within the sound chamber to ensure uniformity of the stimulus. The stimulus intensity varied by a maximum of 3 dB across measured sites within the exposure chamber. The spectral power analysis of noise, in 1/3 octave steps, is shown in Fig. 1. Animals were individually subjected to 4 kHz octave band noise, 115 dB SPL for 5 h. From each group, two animals were exposed beginning at 9 a.m. and three animals beginning at 4 p.m. to reduce the in£uence of circadian variations in GSH levels on the resulting threshold shifts. 2.5. Administration of GSHE GSHE (Kyowa Hakko Kogyo Co.) was administered 1 h before, immediately after and 5 h after noise exposure. At these times, control subjects received an intraperitoneal injection of 5 ml of 0.9% NaCl/kg body weight (groups 1 and 2), experimental subjects received 5 ml of 0.4 M GSHE/kg (group 3) or 5 ml of 0.8 M GSHE/kg (group 4). GSHE was dissolved in 0.9% NaCl. The GSHE dosing schedule was designed to maxi-
Animals were killed under deep anesthesia after the ¢nal ABR measurements. The temporal bones were immediately removed and the perilymphatic spaces perfused and then immersed for 1 h in 4% paraformaldehyde in 0.01 M phosphate-bu¡ered saline (pH 7.4). After permeabilization with 0.3% Triton X-100 for 5 min, whole-mounts of the organ of Corti were stained for actin with £uorescent phalloidin for 30 min to outline hair cells and their stereocilia for a quantitative assessment (Raphael and Altschuler, 1991). We note that phalloidin does not stain bodies of the hair cells. The slide preparations were observed under £uorescence microscopy and missing (as shown by scar formation) and present (as shown by intact stereocilia) hair cells in the sensory epithelium were counted from the apex to the base in 0.19 mm segments. The percentage of hair cell loss was calculated at each observation point in each animal and the average at each observation point was plotted along the cochlea to create an average cytocochleogram. This average represents the percent (ratio of scar to intact cells divided by total cells) per 0.19 mm. Statistical di¡erences were evaluated for signi¢cance on the basis of data from the region of 0^14 mm from apex in the cochlea. 2.7. Statistical analysis ABR measurements were obtained from both ears of each animal and were averaged to yield one data point (threshold shift) per animal. The threshold shifts at each frequency, as well as the percentage of missing outer and inner hair cells, were compared among the groups using a one-way analysis of variance (ANOVA). Post-ANOVA comparisons of means were performed using the Fisher method. Statistical signi¢cance was de¢ned as P 6 0.05. 3. Results 3.1. Weight gain There was no signi¢cant di¡erence in the animals' body weight at the start of the experiment. On a full
HEARES 3474 17-7-00
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
31
3.2. Threshold shifts following noise exposure
Fig. 2. Weight gain of experimental animals. There was no signi¢cant di¡erence in the animals' body weight at the start of the experiment. Values represent mean þ S.D. (n = 5). The weight of subjects on a low protein diet was signi¢cantly di¡erent from group 1 after 10 days (P 6 0.05). GSHE did not reverse the e¡ects of low protein diet on weight. *Signi¢cantly di¡erent from group 1 (P 6 0.05).
protein diet (group 1) body weight increased signi¢cantly throughout the course of the study. The weight of subjects on a low protein diet (groups 2, 3 and 4) remained stable and was signi¢cantly di¡erent from group 1 after 10 days (P 6 0.05). GSHE did not reverse the e¡ects of low protein diet on weight (Fig. 2).
There was no signi¢cant di¡erence in the auditory thresholds between any of the groups before treatment. Thresholds averaged 15 to 25 dB over the range of tested frequencies. Ten days after noise exposure, group 1 exhibited an average threshold shift of approximately 20 dB at 2, 16 and 20 kHz and 35 to 40 dB at other frequencies. Threshold shifts in animals on the low protein diet (group 2) were signi¢cantly greater than baseline at 2, 4, 16 and 20 kHz (P 6 0.05). GSHE supplementation reduced the extent of threshold shifts in a dose-dependent fashion. There was signi¢cant attenuation at 2, 4 and 16 kHz in group 3 and at all frequencies in group 4 (P 6 0.05 ; Fig. 3). 3.3. Histological ¢ndings Whole-mount analysis of the organ of Corti showed that peak damage to the outer (Fig. 4a) and inner (Fig. 4b) hair cells was at a region 10 to 12 mm from apex. Hair cell loss in group 2 was signi¢cantly higher than in the other groups (Fig. 5a,b; P 6 0.05) and also more wide-spread, with a secondary peak of loss at around 5 mm from the apex (Fig. 4a,b). There were no di¡erences among groups 1, 3 and 4.
Fig. 3. E¡ect of low-protein diet and GSHE on threshold shifts. Animals were treated and ABR thresholds measured as described in Section 2. Values represent mean þ S.D. (n = 5). The threshold shifts were signi¢cantly di¡erent (P 6 0.05) between groups 1 and 2 at 2, 4, 16 and 20 kHz. Threshold shifts in group 3 (at 2, 4 and 16 kHz) and group 4 (at all frequencies) were signi¢cantly smaller than in group 2 (P 6 0.05)*.
HEARES 3474 17-7-00
32
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
4. Discussion GSH is an important factor in the modi¢cation of inner ear trauma following noise exposure. This is shown in this study by two complimentary approaches. Under conditions where cochlear GSH is lowered (lowprotein diet) threshold shift and hair cell loss is aggravated. When GSH levels were restored by dietary supplementation, threshold shifts and hair cell loss were reduced. A low-protein diet is a convenient way to lower GSH levels while maintaining the general health of the animals. The diet provides less of the amino acid L-cysteine, a rate-limiting substrate in GSH synthesis (DeLeve and Kaplowitz, 1991). The diet decreases the level of GSH in the cochlear sensory epithelium by about 50% compared to animals on a normal protein diet (Lautermann et al., 1995b). At an operational level it
Fig. 5. Hair cell loss in the organ of Corti after noise exposure (a, outer hair cells; b, inner hair cells). (Group 1: n = 5, group 2: n = 4, group 3: n = 3, group 4: n = 5.) *Signi¢cantly di¡erent from group 2 (P 6 0.05).
Fig. 4. Average cytocochleogram in each group after noise exposure. Animals were treated and cytocochleograms produced as described in Section 2. Hair cell loss at 10 days after noise exposure is averaged for outer (a) and inner (b) hair cells. Predominant damage was observed in the region 10 to 12 mm from apex. Hair cell loss in group 2 is greater and more wide-spread than in the other groups. (Group 1: n = 5, group 2: n = 4, group 3: n = 3, group 4: n = 5.)
is appropriate to interpret the current data to indicate that GSH reverses the e¡ect of low-protein diet. However, GSHE does not reduce other general systemic e¡ects including unchanged body weight. It does reverse GSH levels in the inner ear (Lautermann et al., 1995a). Therefore, we believe these data indicate that GSH limits NIHL. The pattern of threshold shift and hair cell loss re£ects the exposure parameters. In animals on a normal diet, damage was greater at 4, 8 and 12 kHz than at 2, 16 and 20 kHz. The threshold shifts and hair cell loss in the animals on the low protein diet were greater and broader across frequencies. This suggests an increased sensitivity that extends from peak frequencies to structures at the `edge' of the noise exposure ¢eld. The supplement of GSH in the animals on the low protein diet was most e¡ective at the frequencies furthest from the 4 kHz center frequency. The result may indicate that the GSH supplement is inadequate to prevent damage if the level of stress and ROS exceed certain limits. Alternatively, for tissues at the center frequency of the exposure, other factors, in addition to
HEARES 3474 17-7-00
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
ROS formation, may have contributed to the level of destruction. While the noise exposure produced a permanent threshold shift of 20 to 35 dB in groups 1, 3 and 4, the extent of missing inner and outer hair cells was limited and mild. Such a discrepancy is not unusual. Yamasoba et al. (1999) have also found signi¢cant threshold shifts in the presence of mild hair cell loss in guinea pigs exposed to noise. Liberman and Gao (1995) likewise reported that guinea pigs with approximately 40 dB hearing loss showed minimal hair cell loss. Hunter-Duvar and Elliot (1973) even found no observable hair cell loss in squirrel monkey cochleae despite threshold shifts greater than 40 dB. However, species di¡erences may exist in the relationship between hearing loss and hair cell damage. For example, Hamernik et al. (1989) suggested that missing outer hair cells were a major contributor to the ¢rst 30^40 dB of threshold shift in chinchillas and that loss of outer hair cells was almost complete when permanent threshold shifts were more than 40 dB. Subtle changes in the structure of hair cells and other cochlear tissues that are not evident from hair cell counts may have contributed to noise-induced threshold shifts. For example, acoustic overstimulation leads to excessive release of extracellular excitatory amino acid and destruction of the dendrites beneath the inner hair cells (Puel et al., 1996). Such potential changes were not evaluated in the present study. In any case, the increased threshold shifts and hair cell loss at low GSH concentrations are consistent with our previous ¢ndings that drug-inhibition of GSH synthesis enhances NIHL (Yamasoba et al., 1998a). They are also strongly supportive of an involvement of ROS in noise trauma, as suggested by dramatically increased lipid peroxidation under similar conditions (Ohinata et al., 1999). The source of the ROS remains speculative. As in other tissues, ROS in the overstimulated inner ear may be produced by intensi¢ed oxidative processes in mitochondria. Acoustic stimulation indeed increases cochlear energy metabolism (Canlon and Schacht, 1983 ; Ryan et al., 1982). In addition, ROS are generated in tissues in response to prolonged hypoxia or after reperfusion following ischemic periods. Studies utilizing laser Doppler £owmetry (Lamm and Arnold, 1996 ; Scheibe et al., 1990; Thorne and Nuttall, 1987) and intravital microscopy (Lamm and Arnold, 1996; Quirk et al., 1991 ; Quirk et al., 1992; Quirk and Seidman, 1995) have demonstrated reduced cochlear blood £ow during intense sound exposure. Intense sound exposure may also cause vascular alterations such as vasoconstriction, aggregations of red blood cells, increased vascular permeability and localized edema (Axelsson and Dengerink, 1987; Hawkins, 1971; Hultcrantz et al.,
33
1979 ; Yamane et al., 1991 ; Yamane et al., 1995). The combined facts support the idea that localized ischemia and the subsequent release of ROS contribute to the cochlear damage in noise trauma. In conclusion, GSHE reverses the e¡ects of the low protein diet on NIHL. It elevates GSH in the inner ear tissues, reversing the diet-induced reduction in the inner ear tissue GSH. These ¢ndings indicate that cochlear GSH level is involved in NIHL and is consistent with the hypothesis that the generation of ROS is a mechanism of NIHL. Acknowledgements A preliminary report was presented at the 22nd Midwinter Meeting of the Association for Research in Otolaryngology (February 13^18, 1999, at St. Petersburg Beach, FL, USA). We thank Dr. Richard A. Altschuler, Dr. David F. Dolan and Ms. J. Nadine Brown for their valuable help. This work was supported by NIH research grant DC-03685 from the National Institute of Deafness and Communication Disorders, National Institutes of Health.
References Anderson, M.E., Meister, A., 1989. Glutathione monoesters. Anal. Biochem. 183, 16^20. Axelsson, A., Dengerink, H., 1987. The e¡ects of noise on histological measures of the cochlear vasculature and red blood cells: a review. Hear. Res. 31, 183^192. Canlon, B., Schacht, J., 1983. Acoustic stimulation alters deoxyglucose uptake in the mouse cochlea and inferior colliculus. Hear. Res. 10, 217^226. DeLeve, L.D., Kaplowitz, N., 1991. Glutathione metabolism and its role in hepatotoxicity. Pharmacol. Ther. 52, 287^305. El Barbary, A., Altschuler, R., Schacht, J., 1993. Glutathione S-transferases in the organ of Corti of the rat: enzymatic activity, subunit composition and immunohistological localization. Hear. Res. 71, 80^90. Garetz, S.L., Altschuler, R.A., Schacht, J., 1994a. Attenuation of gentamicin ototoxicity by glutathione in the guinea pig in vivo. Hear. Res. 77, 81^87. Garetz, S.L., Rhee, D.J., Schacht, J., 1994b. Sulfhydryl compounds and antioxidants inhibit cytotoxicity to outer cells of a gentamicin metabolite in vitro. Hear. Res. 77, 75^80. Hamernik, R.P., Patterson, J.H., Turrentine, G.A., Ahroon, W.A., 1989. The quantitative relationship between sensory cell loss and hearing thresholds. Hear. Res. 38, 199^212. Hawkins, J.E., 1971. The role of vasoconstriction in noise-induced hearing loss. Ann. Otol. Rhinol. Laryngol. 80, 903^913. Ho¡man, D.W., Whitworth, C.A., Jones, K.L., Rybak, L.P., 1987. Nutritional status, glutathione levels and ototoxicity of loop diuretics and aminoglycoside antibiotics. Hear. Res. 31, 217^222. Ho¡man, D.W., Whitworth, C.A., Jones-King, K.L., Rybak, L.P., 1988. Potentiation of ototoxicity by glutathione depletion. Ann. Otol. Rhinol. Laryngol. 97, 36^41. Hu, B.H., Kopke, R., Zheng, X.Y., Henderson, D., 1996. E¡ect of
HEARES 3474 17-7-00
34
Y. Ohinata et al. / Hearing Research 146 (2000) 28^34
R-phenylisopropyl adenosine on noise-induced hearing loss. Assoc. Res. Otolaryngol. Abstr. 19, 802. Hultcrantz, E., Angelborg, C., Beausant-Linder, M., 1979. Noise and cochlear blood £ow. Arch. Otorhinolaryngol. 224, 103^106. Hunter-Duvar, I.M., Elliot, D.N., 1973. E¡ects of intense auditory stimulation: hearing losses and inner ear changes in the squirrel monkey II. J. Acoust. Soc. Am. 54, 1179^1183. Jacono, A.A., Kopke, R., Vugmeyster, L., Steinman, H., Rowe, J., Hu, B.H., Van De Water, T.R., Henderson, D., 1997. Changes in cochlear antioxidant enzyme activity levels after conditioning noise exposure in the chinchilla. Assoc. Res. Otolaryngol. Abstr. 20, 132. Julicher, R.H., Sterrenberg, L., Haenen, G.R., Bast, A., Noordhoek, J., 1984. Sex di¡erences in the cellular defense system against free radicals from oxygen or drug metabolites in rat. Arch. Toxicol. 56, 83^86. Lamm, K., Arnold, W., 1996. Noise-induced cochlear hypoxia is intensity dependent, correlates with hearing loss and precedes reduction of cochlear blood £ow. Audiol. Neurootol. 1, 148^160. Lautermann, J., McLaren, J., Schacht, J., 1995a. Glutathione protection against gentamicin ototoxicity depends on nutritional status. Hear. Res. 86, 15^24. Lautermann, J., Schacht, J., 1996. A sensitive animal model to assess acute and chronic ototoxic e¡ects. Arch. Otolaryngol. Head Neck Surg. 122, 837^840. Lautermann, J., Song, B., McLaren, J., Schacht, J., 1995b. Diet is a risk factor in cisplatin ototoxicity. Hear. Res. 88, 47^53. Lautermann, J., Crann, S.A., McLaren, J., Schacht, J., 1997. Glutathione-dependent antioxidant systems in the mammalian inner ear: e¡ects of aging, ototoxic drugs and noise. Hear. Res. 114, 75^82. Liberman, M.C., Gao, W.Y., 1995. Chronic cochlear de-e¡erentation and susceptibility to permanent acoustic injury. Hear. Res. 90, 158^168. Meister, A., 1991. Glutathione de¢ciency produced by inhibition of its synthesis, and its reversal: applications in research and therapy. Pharm. Ther. 51, 155^194. Meister, A., Anderson, M.E., 1983. Glutathione. Ann. Rev. Biochem. 52, 711^760. Ohinata, Y., Miller, J.M., Brown, N., Altschuler, R.A., Yamasoba, T., Schacht, J., 1999. Role of reactive oxygen molecules (ROM) in noise induced hearing loss (NIHL). Inner Ear Biol. Abstr. 36, C11. Ohlemiller, K.K., Dugan, L.L., 1998. In vivo measurement of cochlear reactive oxygen species (ROS) in mice: E¡ects of noise exposure and cochlear ischemia. Assoc. Res. Otolaryngol. Abstr. 21, 518. Orrenius, S., Molde¨us, P., 1984. The multiple roles of glutathione in drug metabolism. Trends. Pharmacol. Sci. 6, 432^435. Puel, J.-L., Gervais d'Aldin, C., Sa¤edine, S., Eybalin, M., Pujol, R., 1996. Excitotoxicity and plasticity of IHC-auditory nerve contributes to both temporary and permanent threshold shift. In: Axelsson, A., Borchgrevink, H., Hamernik, R., Hellstro«m, P., Hender-
son, D., Salvi, R. (Eds.), Scienti¢c Basis of Noise-Induced Hearing Loss. Thieme Medical Press, New York, pp. 36^42. Puri, R.N., Meister, A., 1983. Transport of glutathione, as gammaglutamylcysteinylglycyl ester, into liver and kidney. Proc. Natl. Acad. Sci. USA 80, 5258^5260. Quirk, W.S., Avanish, G., Nuttall, A.L., Miller, J.M., 1992. The in£uence of loud sound on red blood cell velocity and blood cell diameter in the cochlea. Hear. Res. 63, 102^107. Quirk, W.S., Seidman, M.D., 1995. Cochlear vascular changes in response to loud noise. Am. J. Otol. 16, 322^325. Quirk, W.S., Shapiro, B.D., Miller, J.M., Nuttall, A.L., 1991. Noiseinduced changes in red blood cell velocity in lateral wall vessels of rat cochlea. Hear. Res. 52, 217^224. Quirk, W.S., Shivapuja, B.G., Schwimmer, C.L., Seidman, M.D., 1994. Lipid peroxidation inhibitor attenuates noise-induced temporary threshold shifts. Hear. Res. 74, 217^220. Raphael, Y., Altschuler, R.A., 1991. Scar formation after drug-induced cochlear insult. Hear. Res. 31, 173^184. Ravi, R., Somani, S.M., Rybak, L.P., 1995. Mechanism of cisplatin ototoxicity: antioxidant system. Pharmacol. Toxicol. 76, 386^394. Ryan, A.F., Goodwin, P., Woolf, N.K., Sharp, F.R., 1982. Auditory stimulation alters the pattern of 2-deoxyglucose uptake in the inner ear. Brain Res. 234, 213^225. Scheibe, F., Haupt, H., Nuttall, A.L., Ludwig, C., 1990. Laser Doppler measurements of cochlear blood £ow during loud sound presentation. Arch. Otorhinolaryngol. 247, 84^88. Seidman, M.D., Shivapuja, B.G., Quirk, W.S., 1993. The protective e¡ects of allopurinol and superoxide dismutase on noise-induced cochlear damage. Otolaryngol. Head Neck Surg. 109, 1052^1056. Thorne, P., Nuttall, A.L., 1987. Laser Doppler measurements of cochlear blood £ow during loud sound exposure in the guinea pig. Hear. Res. 27, 1^10. Usami, S., Hjelle, O.P., Ottersen, O.P., 1996. Di¡erential cellular distribution of glutathione, an endogenous antioxidant, in the guinea pig inner ear. Brain Res. 743, 337^340. Yamane, H., Nakai, Y., Konishi, K., Sakamoto, H., Matsuda, Y., Iguchi, H., 1991. Strial circulation impairment due to acoustic trauma. Acta Otolaryngol. (Stockh.) 111, 85^93. Yamane, H., Nakai, Y., Takayama, M., Iguchi, H., Nakagawa, T., Kojima, A., 1995. Appearance of free radicals in the guinea pig inner ear after noise-induced acoustic trauma. Eur. Arch. Otorhinolaryngol. 252, 504^508. Yamasoba, T., Harris, C., Shoji, F., Lee, R.J., Nuttall, A.L., Miller, J.M., 1998a. In£uence of intense sound exposure on glutathione synthesis in the cochlea. Brain Res. 804, 72^78. Yamasoba, T., Nuttall, A.L., Harris, C., Raphael, Y., Miller, J.M., 1998b. Role of glutathione in protection against noise-induced hearing loss. Brain Res. 784, 82^90. Yamasoba, T., Schacht, J., Shoji, F., Miller, J.M., 1999. Attenuation of cochlear damage from noise trauma by an iron chelator, a free radical scavenger and glial cell line-derived neurotrophic factor in vivo. Brain Res. 815, 317^325.
HEARES 3474 17-7-00