Mouse otocyst transuterine gene transfer restores hearing in mice with connexin 30 deletion-associated hearing loss

Abstract

Although numerous causative genes for hereditary hearing loss have been identified, there are no fundamental treatments for this condition. Herein, we describe a novel potential treatment for genetic hearing loss. Because mutations or deletions in the connexin (Cx) genes are common causes of profound congenital hearing loss in both humans and mice, we investigated whether gene supplementation therapy using the wild-type Cx gene could cure hearing loss. We first generated inner ear-specific connexin 30 (Cx30)-deficient mice via the transuterine transfer of Cx30-targeted short hairpin RNA (shRNA-Cx30) into otocysts. The inner ear-specific Cx30-deficient mice mimicked homozygous Cx30-deficient mice both histologically and physiologically. Subsequently, we cotransfected the shRNA-Cx30 and the wild-type Cx30 gene. The cotransfected mice exhibited Cx30 expression in the cochleae and displayed normal auditory functions. Next, we performed the transuterine transfer of the wild-type Cx30 gene into the otocysts of homozygous Cx30-deficient mice, thereby rescuing the lack of Cx30 expression in the cochleae and restoring auditory functioning. These results demonstrate that supplementation therapy with wild-type genes can restore postnatal auditory functioning. Moreover, this is the first report to show that Cx-related genetic hearing loss is treatable by in vivo gene therapy.

Figure 1

EUGO of the enhanced green fluorescent protein (EGFP) plasmid into the otocysts. (a) A cross-sectional image of a typical adult cochlea. The adult mammalian cochlea is divided into three compartments: the scala vestibuli, the scala tympani, and the scala media. Depicted here is a cross-section of the scala media, which contains the organ of Corti (OC). The OC contains three types of cell populations: inner hair cells (IHCs), outer hair cells (OHCs), and supporting cells (SCs). The two types of auditory hair cells (IHCs and OHCs) play critical roles in hearing as mechanoelectrical transducers. The auditory hair cells are overlaid by the tectorial membrane (TM). The stria vascularis (SV), in the lateral wall of the scala media, is responsible for the secretion of K⁺ into the endolymph and for endocochlear potential production. (b) The arrow indicates an E11.5 embryo otocyst. Fast Green dye was microinjected into the otocysts, and the uterine wall was then removed. (c) The OC exhibited normal cochlear morphology at postnatal day 30 (P30). (d) At P30, there were no significant differences in auditory thresholds between the EGFP-transfected, EGFP-fused Cx30-transfected, and untreated mice. The data are presented as the means ± SEM. Two-tailed t-tests were performed for statistical comparisons. Bar = 20 μm. EUGO, electroporation-mediated transuterine gene transfer into otocyst; SG, spiral ganglion; SL, spiral limbus; SLi, spiral ligament.

Figure 2

EUGO of the enhanced green fluorescent protein (EGFP) plasmid into otocysts and the endogenous Cx30 expression in wild-type mouse cochleae. (a,b) After EUGO of the EGFP plasmid into the otocysts, EGFP (green) was detected in the (a) medial wall (MW), prosensory lesion (PL), lateral wall (LW), and spiral ganglion (SG) at E18.5 and in the (b) spiral limbus (SL), organ of Corti (OC), stria vascularis (SV), SG, and a portion of the spiral ligament (SLi) at postnatal day 30 (P30). (c) Endogenous Cx30 expression (red) was detected in the nascent spiral limbus (NSL) and LW in E18.5 wild-type mouse otocysts. (d) Endogenous Cx30 expression was detected in the SL, SLi, and basal cell areas of the SV and OC in wild-type P30 mice. Bars = 100 μm (in a,c) and 200 μm (b,d). EUGO, electroporation-mediated transuterine gene transfer into otocyst.

Figure 3

The endogenous Cx30 expression in the cochleae was downregulated by short hairpin RNA (shRNA)-Cx30 transfection into the otocysts, and the downregulation of Cx30 expression was restored by the cotransfection of shRNA-Cx30 and resistant-Cx30 at postnatal day 30 (P30). (a) The transfection of shRNA-Cx30 markedly decreased the endogenous expression of Cx30 (red). (b) shRNA-Cx30 (green) was detected in the spiral limbus (SL), organ of Corti (OC), stria vascularis (SV), spiral ligament (SLi), and spiral ganglion (SG). (c) The cotransfection of shRNA-Cx30 and resistant-Cx30 restored the Cx30 expression (red) in the SL, OC, SV, SLi, and SG. (d) A merged image of the Cx30, shRNA-Cx30, resistant-Cx30, and Hoechst-stained images is presented. shRNA-Cx30 (green) and resistant-Cx30 (green) were detected in the SL, OC, SV, SLi, SG, osseous spiral lamina, and lateral bony wall of the cochleae. Bar = 200 μm.

Figure 4

Short hairpin RNA (shRNA)-Cx30 transfection worsened the auditory function in wild-type mice, whereas the cotransfection of shRNA-Cx30 and resistant-Cx30 restored auditory function at postnatal day 30 (P30). (a) The auditory thresholds of the shRNA-Cx30–transfected mice were significantly worse than those of the mice transfected with scrambled plasmid or cotransfected with shRNA-Cx30 and resistant-Cx30 (n = 5 each). There were no significant differences in auditory thresholds between the cotransfected and scrambled plasmid-transfected mice (n = 5 each). (b) The endocochlear potentials (EPs) of the shRNA-Cx30–transfected mice were significantly lower than those of the scrambled plasmid-transfected and cotransfected mice (n = 5 each). There were no significant differences between the EPs of the scrambled plasmid-transfected and cotransfected mice (n = 5 each). (c) The western blot analysis showed that the Cx30 protein expression level was lower in the shRNA-Cx30–transfected mice than in the scrambled plasmid-transfected and cotransfected mice. The data are presented as the means ± SEM. Two-tailed t-tests were performed for statistical comparisons.

Figure 5

The surface morphology of the organ of Corti in the mice that underwent short hairpin RNA (shRNA)-Cx30 transfection into the otocysts was restored by the cotransfection of shRNA-Cx30 and resistant-Cx30 at postnatal day 30 (P30). (a) The surface morphology of the organ of Corti in the mice that underwent a scrambled plasmid transfection was normal. (b) The outer hair cells (OHCs) were partially missing in the shRNA-Cx30–transfected mice at P30. (c) The surface morphology of the organ of Corti in the mice that underwent shRNA-Cx30 and resistant-Cx30 cotransfection was almost normal. The arrowheads in b and c indicate missing hair cells. Bar = 20 μm (in a–c). (d) There were no significant differences in the numbers of inner hair cells (IHCs) among the groups. (e) The number of OHCs in the shRNA-Cx30–transfected mice was significantly lower than those in the scrambled plasmid-transfected and cotransfected mice (n = 5 each). There were no significant differences between the numbers of OHCs in the scrambled plasmid-transfected and cotransfected mice (n = 5 each).

Figure 6

Cx30 transfection restored cochlear function in the postnatal day 30 Cx30^−/− mice. (a) The auditory thresholds of the Cx30^−/− mice were significantly worse than those of the Cx30^+/+ mice (n = 5 each). (b) The auditory thresholds of the treated sides were significantly better than those of the untreated sides (n = 4 each). There were no significant differences between the auditory thresholds of the treated sides and those of the Cx30^+/+ mice (treated mice: n = 4, Cx30^+/+ mice: n = 5). (c) The endocochlear potential (EP) results of the Cx30^−/− mice that underwent Cx30 transfection into the otocysts are presented. The EPs of the treated sides were significantly higher than those of the untreated sides (n = 5 each). The data are presented as the means ± SEM. Two-tailed t-tests were performed for statistical comparisons.

Figure 7

Cx30 transfection restored cochlear Cx30 expression in E18.5 and postnatal day 30 (P30) Cx30^−/− mice. (a) The transfection of enhanced green fluorescent protein (EGFP)-fused Cx30 (green) into the otocyst cells of the left ears of Cx30^−/− mice induced Cx30 expression (red) in the medial wall (MW), prosensory lesion (PL), lateral wall (LW), and spiral ganglion (SG) in the treated left cochlea at E18.5. (b) A merged image of Cx30, EGFP-fused Cx30, and Hoechst staining is shown. (c) The transfection of EGFP-fused Cx30 (green) into the otocyst cells of the left ears of Cx30^−/− mice induced Cx30 expression (red) in the spiral limbus (SL), organ of Corti (OC), stria vascularis (SV), spiral ligament (SLi), and SG of the left cochleae at P30. (d) A merged image of Cx30, EGFP-fused Cx30, and Hoechst staining is shown. (e,f) There was no Cx30 expression in the untreated right cochleae of the Cx30^−/− mice that underwent Cx30 gene transfection into the left ears at (e) E18.5 and (f) P30. Bars = 100 μm (in a,b,e) and 200 μm (in c,d,f).

Figure 8

The auditory response results of the Cx30^−/− mice that underwent Cx30 transfection into otocysts at postnatal day 30 are presented. The auditory thresholds of the treated side were 48 dB at 4 kHz, 37 dB at 12 kHz, and 58 dB at 20 kHz. The auditory thresholds of the untreated side exceeded the maximum scale value at 4 kHz and 20 kHz and were 72 dB at 12 kHz. Thus, ABR waves appeared in the treated ears, and the auditory thresholds of the treated side were better than those of the untreated side. ABR, auditory brainstem response.