Transduction of Adeno-Associated Virus Vectors Targeting Hair Cells and Supporting Cells in the Neonatal Mouse Cochlea

Abstract

Adeno-associated virus (AAV) is the preferred vector for gene therapy of hereditary deafness, and different viral serotypes, promoters and transduction pathways can influence the targeting of AAV to different types of cells and the expression levels of numerous exogenous genes. To determine the transduction and expression patterns of AAV with different serotypes or promoters in hair cells and supporting cells in the neonatal mouse cochlea, we examined the expression of enhanced green fluorescent protein (eGFP) for five different types of AAV vectors [serotypes 2, 9, and Anc80L65 with promoter cytomegalovirus (CMV)-beta-Globin and serotypes 2 and 9 with promoter chicken beta-actin (CBA)] in in vitro cochlear explant cultures and we tested the transduction of AAV2/2-CBA, AAV2/9-CBA, and AAV2/Anc80L65-CMV by in vivo microinjection into the scala media of the cochlea. We found that each AAV vector had its own transduction and expression characteristics in hair cells and supporting cells in different regions of the cochlea. There was a tonotopic gradient for the in vitro transduction of AAV2/2-CBA, AAV2/9-CBA, AAV2/2-CMV, and AAV2/9-CMV in outer hair cells (OHCs), with more OHCs expressing eGFP at the base of the cochlea than at the apex. AAV2/2-CBA in vitro and AAV2/Anc80L65-CMV in vivo induced more supporting cells expressing eGFP at the apex than in the base. We found that AAV vectors with different promoters had different expression efficacies in hair cells and supporting cells of the auditory epithelium. The CMV-beta-Globin promoter could drive the expression of the delivered construct more efficiently in hair cells, while the CBA promoter was more efficient in supporting cells. The in vitro and in vivo experiments both demonstrated that AAV2/Anc80L65-CMV was a very promising vector for gene therapy of deafness because of its high transduction rates in hair cells. These results might be useful for selecting the appropriate vectors for gene delivery into different types of inner ear cells and thus improving the effectiveness of gene therapy.

Keywords: adeno-associated virus vectors; cochlea; gene therapy; hair cell; promoter; supporting cell.

FIGURE 1

Comparison of the in vitro transduction efficiencies in targeting HCs between different serotypes of AAV with the same promoters. (A) Representative confocal images of transduction of five AAV vectors (serotypes 2 and 9 with the CBA promoter and serotypes 2, 9, and Anc80L65 with the CMV-beta-Globin promoter) in HCs of the apical, middle, and basal turns of the cochlea. The arrows indicate the eGFP⁺ HCs. (B) Comparison of the transduction efficiencies between AAV serotypes 2 and 9 with the CBA promoter in targeting IHCs at a titer of 1 × 10¹¹ VG/ml. (C) Comparison of the transduction efficiencies between AAV2/2-CBA and AAV2/9-CBA in targeting OHCs at a titer of 1 × 10¹¹ VG/ml. AAV2/2-CBA clearly transduced more OHCs than AAV2/9-CBA in the apical, middle, and basal turns of the cochlea. (D) Comparison of the transduction efficiencies between serotypes Anc80L65, 2, and 9 with the same CMV-beta-Globin promoter in targeting IHCs at a titer of 1 × 10¹¹ VG/ml. Overall, the transduction efficiency of AAV2/Anc80L65 was statistically significantly greater than the transduction efficiency of AAV2/2. There was no statistically significant difference in the percentages of eGFP⁺ IHCs between AAV2/2 and AAV2/9. (E) Comparison of transduction efficiencies between the three AAV vectors with the CMV-beta-Globin promoter when targeting OHCs at a titer of 1 × 10¹¹ VG/ml. AAV2/Anc80L65 transduced more OHCs in the apical and middle turns than AAV2/9, while AAV2/2 transduced more OHCs in the apical turn than AAV2/9. ^∗p < 0.0167, ^∗∗p < 0.0033, n = 5. Scale bar: 20 μm.

FIGURE 2

Comparison of the in vitro expression efficacy in HCs between AAV vectors with different promoters. (A) Representative confocal images of transduction of AAV2/2-CBA, AAV2/2-CMV-beta-Globin, AAV2/9-CBA, and AAV2/9-CMV-beta-Globin in HCs of the apical, middle, and basal turns of the cochlea. The arrows indicate the eGFP⁺ HCs. (B1) Comparison of the expression efficiencies between AAV2/2-CBA and AAV2/2-CMV-beta-Globin in IHCs at a titer of 1 × 10¹¹ VG/ml. (B2) Comparison of the expression efficiencies between AAV2/2-CBA and AAV2/2-CMV-beta-Globin in OHCs at a titer of 1 × 10¹¹ VG/ml. AAV2/2-CMV-beta-Globin induced more IHCs in the middle and basal turns and more OHCs in the apical and middle turns of the cochlea to express eGFP than AAV2/2-CBA. (C1) Comparison of the expression efficiencies between AAV2/9-CBA and AAV2/9-CMV-beta-Globin in IHCs at a titer of 1 × 10¹¹ VG/ml. (C2) Comparison of the expression efficiencies between AAV2/9-CBA and AAV2/9-CMV-beta-Globin in OHCs at a titer of 1 × 10¹¹ VG/ml. AAV2/9-CMV-beta-Globin induced more IHCs throughout the cochlea to express eGFP than AAV2/9-CBA. ^∗p < 0.0167, ^∗∗p < 0.0033, n = 5. Scale bar: 20 μm.

FIGURE 3

Transduction of AAV2/Anc80L65-CMV-beta-Globin in HCs and SCs at a titer of 2 × 10¹¹ VG/ml. (A) In the HC layer, the arrows indicate the Myo7a⁺/eGFP⁺ HCs. (B) In the SC layer, the arrows indicate the Sox2⁺/eGFP⁺ SCs. The square frames indicate the regions of SCs expressing eGFP.

FIGURE 4

Comparison of the expression efficacy in SCs of the cochlea between AAV vectors with different promoters. (A) Representative confocal images of transduction of AAV2/2-CBA, AAV2/2-CMV-beta-Globin, AAV2/9-CBA and AAV2/9-CMV-beta-Globin in targeting SCs of the apical and middle turns of the cochlea in vitro. There were many Sox2⁺/eGFP⁺ SCs (arrows) in the apical and middle turns of the cochlea when transduced with AAV2/9-CBA. (B) Comparison of the expression efficiencies between AAV2/2-CBA and AAV2/2-CMV-beta-Globin in SCs of the apical, middle, and basal turns of the cochlea in vitro at a working titer of 1 × 10¹² VG/ml. AAV2/2-CBA had greater expression efficacy in SCs than AAV2/2-CMV-beta-Globin on average, although there was no statistically significant difference between the expression efficiencies of AAV2/2-CBA and AAV2/2-CMV-beta-Globin in the apical, middle, or basal turn, respectively. (C) Comparison of the expression efficiencies between AAV2/9-CBA and AAV2/9-CMV-beta-Globin in SCs of the apical, middle, and basal turns of the cochlea in vitro at a working titer of 1 × 10¹² VG/ml. AAV2/9-CBA had greater expression efficacy in SCs than AAV2/9-CMV-beta-Globin on average. n = 4. Scale bar: 20 μm.

FIGURE 5

The confocal images and ABR test results of AAV2/9-CBA transduction into the cochlea after in vivo injection. (A) Transduction of AAV2/9-CBA in HCs of the apical, middle, and basal turns of the cochlea after in vivo injection. The white arrows indicate the Myo7a⁺/eGFP⁺ HCs. (B) Transduction of AAV2/9-CBA in SCs of the apical, middle, and basal turns of the cochlea after in vivo injection. The transduction efficiency of AAV2/9-CBA in SCs gradually decreased from the apex to the base. The tonotopic gradient with more eGFP⁺ SCs at the apex than in the base was similar to what was seen in vitro. The red arrows indicate the Sox2⁺/eGFP⁺ SCs. (C) The ABR thresholds of the injected ears were not statistically significantly different compared to the un-injected control ears 1 month after injection of AAV2/9-CBA into the inner ears (p > 0.05, n = 3). Scale bar: 20 μm.

FIGURE 6

The confocal images of AAV2/Anc80L65-CMV-beta-Globin transduction into the cochlea after in vivo injection and comparison of the in vivo transduction efficiencies in IHCs, OHCs, and SCs between AAV2/9-CBA and AAV2/Anc80L65-CMV-beta-Globin. (A) Transduction of AAV2/Anc80L65-CMV-beta-Globin in HCs of the apical, middle, and basal turns of the cochlea after in vivo injection. (B) Transduction of AAV2/Anc80L65-CMV-beta-Globin in SCs of the apical, middle, and basal turns of the cochlea after in vivo injection. The red arrows indicate the Sox2⁺/eGFP⁺ SCs. (C) Comparison of the in vivo transduction efficiencies in IHCs between AAV2/9-CBA and AAV2/Anc80L65-CMV-beta-Globin. (D) Comparison of the in vivo transduction efficiencies in OHCs between AAV2/9-CBA and AAV2/Anc80L65-CMV-beta-Globin. Overall, the in vivo transduction efficiency of AAV2/Anc80L65-CMV-beta-Globin in IHCs and OHCs was statistically significantly greater than the transduction efficiency of AAV2/9-CBA. (E) Comparison of the in vivo transduction efficiencies in SCs between AAV2/9-CBA and AAV2/Anc80L65-CMV-beta-Globin. AAV2/Anc80L65-CMV-beta-Globin induced more SCs to express eGFP than AAV2/9-CBA on average (24.33 ± 1.83% vs. 17.87 ± 1.88%, p = 0.0130). ^∗p < 0.0167, ^∗∗p < 0.0033, n = 3. Scale bar: 20 μm.