Partial protection against permanent noise-induced threshold deficits can be achieved using glutathione (GSH) based strategies. The GSH peroxidase mimic 2-phenyl-1,2-benzisoselenazol-3(2H)-one (Ebselen) (adapted from Lynch et al., 2004), the GSH precursor N-acetylcysteine (NAC, adapted from Ohinata et al., 2003), and the cysteine precursor D-methionine (D-Met, adapted from Kopke et al., 2002) all provide partial protection. Data are either mean ± SD (NAC) or mean ± SE (D-Met), as plotted in the original publications. Permanent threshold deficits at 4 kHz (A), 8 kHz (B), and 16 kHz (C) were measured either 10 days (NAC) or 3 weeks (Ebselen, D-Met) post-noise. In the investigation performed by Lynch et al. (2004), rats (N=4 animals/group) were exposed to 113 dB SPL noise (4 kHz to 16 kHz) for 4 hours; rats were treated with 4 mg/kg ebselen p.o. BID one day pre- and one day post-noise, as well as one hour pre- and one hour post-noise on the day of exposure. In the investigation performed by Ohinata et al. (2003), guinea pigs (N=16 control animals, N=7 treated animals) were exposed to 115 dB SPL noise (octave band centered at 4 kHz) for 5 hours; guinea pigs were treated with a single dose of 500 mg/kg NAC i.p. 30 minutes pre-noise. In the investigation performed by Kopke et al. (2002), chinchillas (N=6 animals/group) were exposed to 105 dB SPL noise (octave band centered at 4 kHz) for 6 hours; chinchillas were treated with 200 mg/kg D-methionine i.p. BID beginning two days pre-noise and continuing two days post-noise, as well as one hour pre- and one hour post-noise on the day of exposure. Although the utility of comparisons across agents and investigations is compromised by variation in dose schedule and duration, species, and noise exposure, the data clearly illustrate that protection via glutathione-based strategies is typically incomplete. Asterisks indicate statistically reliable differences relative to saline-treated control animals tested in each investigation.

Immunolabeling of byproducts of reactive oxygen species (ROS) and reactive nitrogen species (RNS) varies with time post time. Byproducts of 4-hydroxynonenal (4-HNE, panels A and C), used to mark ROS byproducts, and nitrotyrosine (NT, panels B and D), used to mark RNS byproducts is shown. Little or no labeling was evident in control animals not exposed to noise (panels A and B). Peak ROS and RNS production was observed in cells in the organ of Corti 7–10 days post-noise. Noise exposure was an octave band noise centered at 4 kHz and presented at a level of 120-dB SPL for 5 hours. Significant 4-HNE and NT expression was observed in the outer hair cell bodies, with some labeling in inner hair cells as well (panels C and D). Adapted from Yamashita et al. (2004); see also Miller et al. (2006b).

Byproducts of free radical formation can be measured in the cochlea post-noise. Enzymatic (ELISA) immunoassay for 8-isoprostane-F2α in the cochlea revealed a robust (approximately 30-fold) increase in free radical formation immediately following 5 hours of noise (adapted from Ohinata et al., 2000a). Noise exposure was an octave band noise centered at 4 kHz and presented at a level of 115-dB SPL. Levels of 8-isoprostane-F2α increased almost linearly during the duration of the exposure (based on samples collected immediately following 1, 3, and 5 hour exposure durations; not shown), and decreased rapidly after the termination of noise (8 hours post-noise illustrated, mean ± SE). Reprinted from Miller et al. (2006b).

Pre-noise treatment with calcineurin inhibitors reduces noise-induced threshold deficits. Treatment with the calcineurin inhibitors FK506 (10 μg/ml) or cyclosporine A (10 μg/ml), delivered directly into the cochlear perilymph, reduced noise-induced threshold deficits (average hearing loss at 4, 8, and 16 kHz; mean ± SE). Noise exposure was an octave band noise centered at 4 kHz and presented at a level of 120-dB SPL for 5 hours. Asterisks indicate statistically reliable differences relative to subjects treated with artificial perilymph (control). Adapted from Minami et al. (2004).

Syergistic interactions among agents, while likely, have been challenging to identify. Treatment with antioxidants (salicylate: N=6 ears from 6 animals; Trolox, a synthetic analogue of vitamin E: N=10 ears from 5 animals), a vasodilator (betahistine: N=8 ears from 4 animals), or combinations of these agents (N=8 ears from 4 animals for each combination) reduced NIHL relative to controls (N=28 ears from 18 animals), with no additive effects observed with combinations of these agents. Given the small numbers of animals in some groups, statistical reliability of observed differences in NIHL was not evaluated. Average hearing loss at 4, 8, and 16 kHz (mean ± SE) is illustrated, reprinted from Miller et al. (2006b).

Status of the endogenous antioxidant glutathione (GSH) influences noise-induced hearing loss (NIHL). Treatment with L-buthionine-[S,R]-sulfoximine (BSO), which inhibits endogenous GSH synthesis, increases NIHL whereas 2-oxothiazolidine-4-carboxylate (OTC), a pro-cysteine drug which promotes rapid restoration of GSH, reduces NIHL. Threshold deficits are illustrated 1 day (A), 3 days (B), 5 days (C), and 14 days (D) following noise (average hearing loss at 12 kHz, mean ± SE); effects of OTC were time-dependant in that permanent threshold shift was reduced whereas initial temporary threshold deficits were unchanged (i.e., days 1–3). Noise exposure was a broadband noise presented at a level of 102-dB SPL for 3 hours per day for 5 consecutive days. Asterisks indicate statistically reliable differences relative to saline-treated control. Adapted from Yamasoba et al. (1998).

Partial protection against permanent noise-induced threshold deficits can be achieved using dietary antioxidants. Agents evaluated to date include vitamin A (see Ahn et al., 2005 for click thresholds), vitamin C (adapted from McFadden et al., 2005), vitamin E (adapted from Hou et al., 2003), vitamin E plus salicylate (adapted from Yamashita et al., 2005a), and resveratrol (see Seidman et al., 2003 for data at 3, 6, 9, 12, and 18 kHz), which occurs naturally in red wine and grape skin. Data are either mean ± SD (vitamins A, E) or mean ± SE (vitamin C, vitamin E + salicylate), as plotted in the original publications. Permanent threshold deficits at 4 kHz (A), 8 kHz (B), and 16 kHz (C) shown here were measured either 8 days (vitamin E), 10 days (vitamin E plus salicylate), or 21 days (vitamin C) post-noise. In the investigation performed by McFadden et al. (2005), guinea pigs (N=8 animals/group) were exposed to 115 dB SPL noise (octave band centered at 4 kHz) for 6 hours; beginning 35 days pre-noise exposure, guinea pigs were fed a diet containing 500 mg per kg chow L--2-pascorbylolyphosphate (control diet) or 5,000 mg per kg chow L--2-pascorbylolyphosphate (vitamin C supplement). In the investigation performed by Hou et al. (2003), guinea pigs (N=8 animals/group) were exposed to 100 dB SPL noise (octave band centered at 4 kHz) for 8 hours per day for 3 days; guinea pigs were treated with 10 (not shown) or 50 mg/kg vitamin E (α-tocopherol) i.p. beginning three days pre-noise and continuing three days post-noise, as well as on the three days of exposure. In the investigation performed by Yamashita et al. (2005), guinea pigs (N=6 animals/group) were exposed to 120 dB SPL noise (octave band centered at 4 kHz) for 5 hours; guinea pigs were treated with 50 mg/kg vitamin E (trolox) i.p. and 75 mg/kg salicylate s.c. BID. Here we show data from animals in which treatment began 3 days prior to noise exposure and continued ten days post-noise. Although the utility of comparisons across agents and investigations is reduced by variation in dose schedule and duration, species, and noise exposure, the data clearly illustrate that protection via dietary antioxidants has been incomplete with all combinations tested to date. Asterisks indicate statistically reliable differences relative to saline-treated control animals tested in each investigation.

Antioxidant treatment reduces noise-induced hearing loss even when treatment is delayed relative to the noise insult. Treatment with Trolox (a synthetic analogue of Vitamin E) and salicylate reduced threshold deficits (A, average hearing loss at 4, 8, and 16 kHz, mean ± SE) and hair cell loss (B, 10–16 mm from the apex) 10 days post-noise (adapted from Yamashita et al., 2005a) when treatment was initiated 3 days pre-noise, or up to 3 days post-noise (1 hour, 1 day, or 3 days post-noise). Noise exposure was an octave band noise centered at 4 kHz and presented at a level of 120-dB SPL for 5 hours. Data are mean ± SE, and asterisks indicate statistically reliable differences relative to saline-treated controls. Adapted from Yamashita et al. (2005a); see also Miller et al. (2006b).

Oxidative stress initiates a cascade of reactions leading to cell death via one or more sequences of molecular events. First, calcium-dependant calcineurin/calpain activation can initiate downstream events including dephosphorylation of NFAT and activation of BAD, release of cytochrome c, activation of caspases 9 and 3, and cell death. Second, a caspase-2 dependant pathway to cell death can be triggered by DNA damage. Third, caspase-independent pathways to cell death include release of AIF and EndoG from the mitochondria. Translocation to the cell nucleus results in chromatin condensation and high-molecular mass-chromatin fragments. Fourth, receptor-mediated cell death begins with ligation of death receptors, death inducible signaling complex is formed, which activates pro-caspase-8. Caspase-8 can activate caspase-3, leading to cell death, or cleave Bid, which results in translocation and insertion of Bax and/or Bak into the mitrochondrial membrane and release of cytochrome c, activation of caspases 9 and 3, and cell death. The caspase-2 dependent pathway differs from the caspase-8 and caspase-9 dependent pathways in that pro-caspase-2 is activated by DNA damage. Different cell types vary in the extent to which apoptotic cell death depends on the activation of caspase-2 (Lassus et al., 2002; Robertson et al., 2002), and a specific role for caspase-2 in apoptotic cell death in the inner ear has not been described to date.

A. Treatment with an iron chelator (deferoxamine mesylate, DFO), an antioxidant (mannitol, a hydroxyl scavenger), or a combination of these agents (with or without the neurotrophic factor GDNF) reduced NIHL (average hearing loss at 4, 8, and 16 kHz, mean ± SE, adapted from Yamasoba et al., 1999). Protection via DFO was statistically reliable at 16 kHz but not 4 or 8 kHz. Protection via mannitol was statistically reliable at 4 kHz but not 8 or 16 kHz. The combination of DFO + mannitol did not result in statistically reliable reduction of NIHL at 4, 8, or 16 kHz. The combination of DFO + mannitol + GDNF resulted in statistically reliable protection against NIHL at all of the test frequencies (see asterisk). Panel A reprinted from Miller et al. (2006b). Figure 9B. Although the reliability of threshold protection varied across agents, reductions in outer hair cell loss between 7.6 and 19.0 mm from the apex of the cochlea were statistically reliable for each of the individual agents as well as the combinations of agents (see asterisks). Average outer hair cell loss in rows 1–3 (mean ± SE) is illustrated. Panel B adapted from Yamasoba et al. (1999).