Tmprss3 Gene Expression and Gene Therapy in a Mouse Model of Human Deafness

Abstract

To investigate gene therapy as a potential biologic strategy for restoration of auditory function in patients with genetic hearing loss, we tested a gene augmentation approach in mouse models of deafness resulting from mutations in the transmembrane protease serine 3 gene (Tmprss3). Tmprss3 mutations cause non-syndromic autosomal recessive deafness in childhood (DFNB8) and early infancy (DFNB10), and in severe cases cause vestibular symptoms. The genuine structural and functional role of TMPRSS3 in the inner ear is still poorly understood. In this study, I hypothesized the protein TMPRSS3 was important for inner ear development and may be amenable to gene therapy in Tmprss3Y260X mice, and potentially humans. I found that (1) Tmprss3 is diffusely expressed in the inner ear, particularly in hair cells, the organ of Corti, spiral ganglion cells, the inner and outer spiral sulcus, and interdental cells, (2) lack of TMPRSS3 results in structural degeneration of hair cells and severe auditory impairment relative to wild-type mouse controls, and (3) Tmprss3 point mutations (Y260X) lead to altered hair cell function without hair bundle structural disruption perceivable on electron morphology prior to postnatal day 10. Tmprss3 expression was driven in vivo using gene therapy AAV2/Anc80-CMV-Tmprss3 vectors and led to partial hearing recovery (dB <70 to 100) in one homozygous mutant mouse compared with uninjected homozygous mutant controls (dB > 110+). The contralateral ear in the animal showed hearing thresholds > 110 dB at all frequencies, demonstrating true recovery of function in the injected ear. However, there was no functional recovery in DPOAEs, suggesting minimal rescue of outer hair cells. On histology, cochlear tissue of the homozygous mutant mouse injected with AAV2/Anc80-CMV-Tmprss3 showed recovery of hair cell morphology of inner hair cells and intact hair bundles with robust intracellular TMPRSS3 expression, similar to wild-type controls. Outer hair cells in the injected the homozygous Tmprss3Y260X mouse showed disrupted cell morphology with TMPRSS3 localization in the outer cell surface membrane. Histology findings were consistent with hearing function tests. On vestibular testing, homozygous mutant and heterozygous mice showed a similar latency to fall on rotarod testing, which was statistically significant from wild-type counterparts (p < 0.0001). There were no significant differences in open field performance. Our data may establish a foundation for development of gene therapy to treat DFNB8/10 in humans.