Early transcriptional response to aminoglycoside antibiotic suggests alternate pathways leading to apoptosis in sensory hair cells in the mouse inner ear.

Litao Tao,Neil Segil

doi:10.3389/fncel.2015.00190

Abstract

Aminoglycoside antibiotics are “the drug of choice” for treating many bacterial infections, but their administration results in hearing loss in up to one fourth of the patients who receive them. Several biochemical pathways have been implicated in aminoglycoside antibiotic ototoxicity; however, little is known about how hair cells respond to aminoglycoside antibiotics at the transcriptome level. Here we have investigated the genome-wide response to the aminoglycoside antibiotic gentamicin. Using organotypic cultures of the perinatal organ of Corti, we performed RNA sequencing using cDNA libraries obtained from FACS-purified hair cells. Within 3 h of gentamicin treatment, the messenger RNA level of more than three thousand genes in hair cells changed significantly. Bioinformatic analysis of these changes highlighted several known signal transduction pathways, including the JNK pathway and the NF-κB pathway, in addition to genes involved in the stress response, apoptosis, cell cycle control, and DNA damage repair. In contrast, only 698 genes, mainly involved in cell cycle and metabolite biosynthetic processes, were significantly affected in the non-hair cell population. The gene expression profiles of hair cells in response to gentamicin share a considerable similarity with those previously observed in gentamicin-induced nephrotoxicity. Our findings suggest that previously observed early responses to gentamicin in hair cells in specific signaling pathways are reflected in changes in gene expression. Additionally, the observed changes in gene expression of cell cycle regulatory genes indicate a disruption of the postmitotic state, which may suggest an alternate pathway regulating gentamicin-induced apoptotic hair cell death. This work provides a more comprehensive view of aminoglycoside antibiotic ototoxicity, and thus contributes to identifying potential pathways or therapeutic targets to alleviate this important side effect of aminoglycoside antibiotics.

Highlights

Ototoxicity is a well-known side effect limiting the use of aminoglycoside antibiotics, with reported incidence of hearing loss between 2% and 25% of treated patients (Huth et al, 2011)
To investigate the early transcriptional response of hair cells to gentamicin, cultured cochleae were treated with gentamicin for 3 h, and immediately dissociated and FACSsorted to obtain purified hair cell and non-hair cell samples (Figure 1C) for RNA sequencing
Since there is a low level of misexpression of GFP in inner phalangeal cells and probably in border cells, cells with low-level GFP expression were excluded by stringently gating GFP during fluorescence-activated cell sorting (FACS) purification to minimize possible contamination

Summary

Introduction

Ototoxicity is a well-known side effect limiting the use of aminoglycoside antibiotics, with reported incidence of hearing loss between 2% and 25% of treated patients (Huth et al, 2011). Several biochemical mechanisms of aminoglycoside ototoxicity have been investigated, including production of reactive oxygen species (ROS) (Forge and Schacht, 2000), disruption of intracellular calcium storage (Matsui et al, 2004), and inhibition of cytoplasmic protein synthesis (Francis et al, 2013). Some pathways have been identified as signaling pathways mediating aminoglycoside-induced hair cell death, such as the pro-apoptotic JNK pathway (Ylikoski et al, 2002) and the protective NF-κB pathway (Jiang et al, 2005). ROS production is induced shortly after aminoglycoside administration; the JNK pathway responds to drug treatment within 3 h; the intracellular calcium level is increased dramatically at around 4 h; cytochrome C is released into cytoplasm at around 12 h and apoptosis execution steps occur at around 18 h (Matsui et al, 2004). Whole cochlea samples were used in those studies, which consisted largely of many different non-hair cell populations in addition to hair cells, and the heterogeneity of the samples limited the interpretation of the results due to a low signal to noise ratio

Methods

Results

Conclusion