AAVR Expression is Essential for AAV Vector Transduction in Sensory Hair Cells

Figures

• 

  Figure 1. Expression of AAVR in inner ear hair cells is significantly decreased in adult mice compared to neonatal mice. A) Representative immunoblot images using whole inner ear homogenates of wild‐type C57BL/6 mice (WT) revealed a decreased AAVR band at 150 kDa at the age of 30 days (P30) compared to postnatal 3 days (P3). GAPDH (36 kDa) serves as the sample loading control. B) Semi‐quantification of AAVR band density confirmed a significant decrease at P30 compared to P3. Data are presented as means + SD, n = 3 per group with 4 ears pooled into one sample, **p < 0.01, analyzed by unpaired t‐test. C) Immunolabeling for AAVR (green) in the organ of Corti (OC) showed strong labeling in both IHCs and OHCs at P3. In contrast, AAVR labeling was weak in P30 mice with IHCs showing stronger labeling than OHCs, as outlined by the white box (upper panel). The lower panel shows reconstructed images from the side view, confirming that AAVR is localized in stereocilia of both IHCs and OHCs with strong labeling at P3. D) Immunolabeling for AAVR in the vestibular system (crista and utricle) also show a decrease in expression in hair cells at P30 compared to P3. Myosin 7a was co‐labeled to visualize…

• 

  Figure 2. Efficiency of GFP expression in outer hair cells in the organ of Corti and vestibular hair cells in the crista and utricle via AAV2 transduction is significantly lower in adult mice than neonatal mice. A) Images indicate LSC micro‐injection of AAV vectors in neonatal and adult mice. PSC: posterior semicircular canal; LSC: lateral semicircular canal. B) Representative images revealed GFP expression in OHCs and IHCs in the apex, middle, and basal turns of the OC and in the vestibular hair cells of the crista and utricle of wild‐type C57BL/6 mice 20 days post AAV2 vector injection with the same number of viral particles via the lateral semicircular canal (LSC) either at P3 or P30. Hair cells were counter‐labeled with myosin 7a for visualization. Scale bar = 50 µm. C) Counting GFP‐positive OHCs in the OC and hair cells from the crista and utricle showed significantly lower GFP‐positive cells in P30 AAV‐injected mice compared to P3 AAV‐injected mice, while in IHCs there was no difference between P30 and P3 mice. *p < 0.05, ***p < 0.001, ****p < 0.0001, ns: not significant. n = 4 in each group, analyzed by unpaired t‐tests.

• 

  Figure 3. AAV2 viral particles form complexes with AAVR in in vitro models. Immunolabeling of intact AAV2 particles (A20, green) and AAVR (red) in A) HEI‐OC1 cells and B) P3 cochlear explants 20 min after AAV2 administration. Both HEI‐OC1 cells and stereocilia of IHC and OHCs revealed almost exact co‐localization of AAV2 and AAVR. The enlarged images from the white boxes are for better visualization. Arrows in the enlarged images indicate the axis for analysis of the plot profiles presented below (C). NC: negative control, blue: DAPI staining for nuclei; gray: phalloidin staining for hair cell structure. Scale bar = 10 µm. C) Intact AAV2 particle (A20) fluorescence and AAVR fluorescence signal plot profiles from the narrow band regions indicated by the arrows in the enlarged images (A and B) confirmed their colocalization in both HEI‐OC1 cells and hair cells of the cochlear explants. D) Formation of AAV2‐AAVR complexes was confirmed by co‐immunoprecipitation. Representative images with the first lane loaded with whole cell lysates as input, the second lane with a sample immunoprecipitated with an IgG antibody, and the third lane with a sample that was immunoprecipitated with an AAVR…

• 

  Figure 4. Blockage of AAVR by its antibody resulted in decreased AAV2 attachment and a lower GFP transduction rate in HEI‐OC1 and sensory hair cells of cochlear explants. A) The schematic diagram illustrates the antibody inhibition assay procedure in HEI‐OC1 cells. AAVR or IgG control antibody is applied to the media of cultured HEI‐OC1 cells for 1 h. Cells are then washed with prewarmed PBS before AAV2 administration. Samples are fixed 20 min after AAV2 exposure for the AAV attachment assay and 72 h after for GFP‐positive rate analysis. B) Representative immunolabeling of AAV2 particles (A20, green) showed fewer green dots on HEI‐OC1 cells in the AAVR antibody pre‐treated group compared to the IgG isotype control group. Scale bar = 10 µm. C) Quantification of AAV2 vector particles confirmed a significantly lower number in the AAVR antibody pre‐treated group compared to the control group 20 min after administration of AAV2. Data are presented as individual samples with means + SD, n = 12 in the IgG control group and n = 14 in the anti‐AAVR group from 3 independent experiments, ***p < 0.001, analyzed by unpaired t‐test. D) Representative fluorescence images showed significantly lower…

• 

  Figure 5. Knockout of AAVR blocks GFP expression via AAV2 transduction in cochlear hair cells in vivo. A) The design concept of AAVR knockout (KO) mice is shown in schematic form. Two guiding RNAs were generated to target deletion of exon 3 of AAVR. Forward primer 1 (F1) and two reverse primers (R1 and R2) were designed for genotyping. B) Representative PCR results from tail lysis showed a 2010‐bp band by F1R1 primer pairs and a 604‐bp band by F1R2 primer pairs in wild‐type (AAVR+/+) mice, a 587‐bp band by F1R1 primer pairs and no band by F1R2 primer pairs in homozygous AAVR KO mice (AAVR−/−). C) Immunoblots using whole inner ear homogenates from AAVR KO mice revealed strong reduction of AAVR band density at 150 kDa compared to wild‐type controls. GAPDH serves as the sample loading control. The experiment was repeated twice with 4 inner ears pooled into each sample. D) AAVR KO mice showed no obvious phenotypical differences compared to wild‐type littermates at neonatal (P3) and adult ages (P30). E) Immunolabeling for AAVR (red) on cochlear surface preparations of P3 AAVR KO mice confirmed knockdown of AAVR immunolabeling in IHCs and OHCs compared to wild‐type littermates. Hair cells were…

• 

  Figure 6. Design of conditional overexpression of AAVR in hair cells through cross with Gfi1‐Cre mice. A) The schematic diagram illustrates the design of AAVR conditional overexpression in sensory hair cells of mice. Briefly, conditional overexpression elements, including full‐length AAVR cDNA connected with an HA‐tag in the 5’‐end followed by a tdTomato sequence as a reporter, are separated from the CMV promoter by a loxP‐stop‐loxP (lsl) element. When crossed with Gfi1‐Cre +/− mice, the stop signal is removed to initiate exogenous AAVR translation and expression. Rosa26‐CMV‐lsl‐tdTomato mice were crossed with Gfi1‐Cre +/− mice to generate matched control mice. B) Representative genotyping PCR results of Gfi1‐Cre mice, CKI‐AAVR mice, and CKI‐Ctrl mice. Since Gfi1‐Cre homozygosity is lethal, only Gfi1‐Cre heterozygous mice were used in this study. The band for the wild‐type allele Gfi1‐Cre mice was detected at 609 bp and the mutant allele (±) at 672 bp by a F3R3 primer pair. The band for the WT allele of CKI‐Ctrl mice was detected at 297 bp by a F5R5 primer pair and the mutant allele at 196 bp by a F4R4 primer pair; the WT allele of CKI‐AAVR mice was detected at 453 bp by a F6R6 primer pair…

• 

  Figure 7. Conditional overexpression of AAVR in hair cells restores their sensitivity to AAV transduction in adult mice. A) AAV2 carrying a GFP reporter gene was injected through the LSC into CKI‐AAVR+/+; Gfi1‐Cre +/− and CKI‐Ctrl+/+; Gfi1‐Cre +/− mice at the age of P30. Twenty days post vector injection, GFP (green) expression was clearly seen in OHCs in the apical, middle, and basal turns of the OC and vestibular HCs of the crista and utricle in CKI‐AAVR+/+; Gfi1‐Cre +/− mice, but only weakly expressed in CKI‐Ctrl+/+; Gfi1‐Cre +/− mice. Red: tdTomato, blue: myosin‐7a‐labeled structure of HCs. Scale bar = 10 µm. B) Quantification of GFP‐positive hair cells confirmed a significant increase in transduction into OHCs in the OC and in vestibular HCs in the crista and the utricle. There was no difference in GFP expression in IHCs between the CKI‐AAVR+/+; Gfi1‐Cre +/− group and CKI‐Ctrl+/+; Gfi1‐Cre +/− mice. Data are presented as means + SD, n = 4 in each group. ****p < 0.0001, ns: non‐significant, analyzed by unpaired t‐tests.