Lower ototoxicity and absence of hidden hearing loss point to gentamicin C1a and apramycin as promising antibiotics for clinical use

Abstract

[Background]: Spread of antimicrobial resistance and shortage of novel antibiotics have led to an urgent need for new antibacterials (Maura et al. 2016, Curr Opin Microbiol 33: 41-46; Tacconelli et al. 2018, Lancet Infect Dis 18: 318-327). Although aminoglycoside antibiotics (AGs) exhibit potent antimicrobial activity, their use has been largely restricted due to serious sideeffects, mainly nephrotoxicity and ototoxicity (Forge and Schacht 2000, Audiol Neurootol 5: 3-22; Huth et al. 2011, Int J Otolaryngol 2011: 937861). It is therefore of great importance to identify AGs of strong antibacterial activity that lack their most harmful side effects.

[Methods]: A large number of AGs were tested against a series of multidrug-resistant clinical isolates of the ESKAPE panel; of these, five AGs showing strong antibacterial activity were selected to evaluate their ototoxicity. A stepwise approach was followed, aiming at setting up a protocol that could be used in future high-throughput screenings. In vitro tests were initially conducted by assessing the viability of two established otic cell lines following AG treatment, and subsequently on murine cochlear organotypic cultures, by analysing hair cell survival. In vivo work was then carried out on a guinea pig model, following local round window application of the AGs.

[Results]: Commercial gentamicin mixture (GM), the GM congener gentamicin C1a (GM C1a), apramycin (Apra), paromomycin (Paro) and neomycin (Neo) were selected for ototoxicity testing. In vitro analyses confirmed GM and Neo as the most toxic of the tested AGs, and Apra and Paro as those with the lowest toxicity; interestingly, GM C1a appeared to be less toxic than GM. Regarding the in vivo work, a dose-dependent effect of AGs on outer hair cell (OHC) survival and compound action potentials (CAPs) showed that GM C1a and Apra were the least toxic. Strikingly, although no changes were observed in CAP thresholds and OHC survival following treatment with low concentrations of Neo, GM and Paro, the number of inner hair cell (IHC) synaptic ribbons and the CAP amplitudes were reduced. This indication of hidden hearing loss was not observed with GM C1a or Apra at such concentrations.

[Conclusion]: These findings have: (a) validated our screening approach, approach that will now be used for high-throughput testing of newly isolated AG congeners, (b) revealed the IHCs as the inner ear’;s most vulnerable element to AG treatment, and (c) identified GM C1a and Apra as promising bases for the development of clinically useful antibiotics.