Abstract
Paraquat, a superoxide generator, can damage the cochlea causing an ototoxic hearing loss. The purpose of the study was to determine if deletion of Bak, a pro-apoptotic gene, would reduce paraquat ototoxicity or if deletion of Sirt3, which delays age-related hearing loss under caloric restriction, would increase paraquat ototoxicity. We tested these two hypotheses by treating postnatal day 3 cochlear cultures from Bak±, Bak-/-, Sirt3±, Sirt3-/-, and WT mice with paraquat and compared the results to a standard rat model of paraquat ototoxicity. Paraquat damaged nerve fibers and dose-dependently destroyed rat outer hair cells (OHCs) and inner hair cells (IHCs). Rat hair cell loss began in the base of the cochlea with a 10 μM dose and as the dose increased from 50 to 500 μM, the hair cell loss increased near the base of the cochlea and spread toward the apex of the cochlea. Rat OHC losses were consistently greater than IHC losses. Unexpectedly, in all mouse genotypes, paraquat-induced hair cell lesions were maximal near the apex of the cochlea and minimal near the base. This unusual damage gradient is opposite to that seen in paraquat-treated rats and in mice and rats treated with other ototoxic drugs. However, paraquat always induced greater OHC loss than IHC loss in all mouse strains. Contrary to our hypothesis, Bak deficient mice were more vulnerable to paraquat ototoxicity than WT mice (Bak-/- > Bak± > WT), suggesting that Bak plays a protective role against hair cell stress. Also, contrary to expectation, Sirt3-deficient mice did not differ significantly from WT mice, possibly due to the fact that Sirt3 was not experimentally upregulated in Sirt3-expressing mice prior to paraquat treatment. Our results show for the first time a gradient of ototoxic damage in mice that is greater in the apex than the base of the cochlea.