Gene editing in a Myo6 semi-dominant mouse model rescues auditory function

Abstract

Myosin VI（MYO6） is an unconventional myosin that is vital for auditory and vestibular function. Pathogenic variants in the human MYO6 gene cause autosomal-dominant or -recessive forms of hearing loss. Effective treatments for Myo6 mutation causing hearing loss are limited. We studied whether adeno-associated virus (AAV)-PHP.eB vector-mediated in vivo delivery of Staphylococcus aureus Cas9 (SaCas9-KKH)-single-guide RNA (sgRNA) complexes could ameliorate hearing loss in a Myo6^WT/C442Y mouse model that recapitulated the phenotypes of human patients. The in vivo editing efficiency of the AAV-SaCas9-KKH-Myo6-g2 system on Myo6^C442Y is 4.05% on average in Myo6^WT/C442Y mice, which was ∼17-fold greater than editing efficiency of Myo6^WT alleles. Rescue of auditory function was observed up to 5 months post AAV-SaCas9-KKH-Myo6-g2 injection in Myo6^WT/C442Y mice. Meanwhile, shorter latencies of auditory brainstem response (ABR) wave I, lower distortion product otoacoustic emission (DPOAE) thresholds, increased cell survival rates, more regular hair bundle morphology, and recovery of inward calcium levels were also observed in the AAV-SaCas9-KKH-Myo6-g2-treated ears compared to untreated ears. These findings provide further reference for in vivo genome editing as a therapeutic treatment for various semi-dominant forms of hearing loss and other semi-dominant diseases.

Keywords: CRISPR-Cas9; gene therapy; hearing loss; myosin VI; semi-dominant hearing loss.

Graphical abstract

Figure 1

The genome-editing strategy to disrupt the Myo6^C442Y mutant allele (A) SaCas9-KKH sgRNA design. The Myo6 p.C442Y mutation site is highlighted in red (a G to A transition at position 1325 of the Myo6 cDNA). Nucleotides of the PAM sites are marked by green. (B) Schematic for the in vitro studies in Myo6^C442Y/C442Y ESCs. Lipofectamine 3000-mediated delivery of the SaCas9-KKH sgRNA therapy vector into ESCs. Cells were sorted by GFP expression. (C and D) Indel percentages of g1 and g2 in Myo6^WT/WT, Myo6^{C442 /C442Y}, and Myo6^WT/C442Y mESCs were determined by targeted NGS and analysis using Seqkit. Individual values (n = 3) are shown, and horizontal lines and error bars represent the mean values ± SD of three independent biological replicates. Each dot represents the DNA from one sample. (E) The indels causing in-frame versus frameshift mutations were calculated after SaCas9-KKH-Myo6-g2 transfection in Myo6^{C442Y/ C442Y} ESCs. The data were obtained from the average of three independent repeated experiments. (F) Indel profiles from SaCas9-KKH-Myo6-g2-transfected Myo6^C442Y/C442Y ESCs. Negative numbers represent deletions, and positive numbers represent insertions. Sequences without indels (value = 0) are not included in the chart. (G) Indel frequency at the Myo6^C442Y mutation locus (On) and at each of the off-target (OT) loci in SaCas9-KKH-Myo6-g2-treated Myo6^C442Y/C442Y ESCs. Blue, samples treated with DNA plasmids encoding SaCas9-KKH-Myo6-g2; orange, control samples with SaCas9-KKH-Myo6-NT-sg.

Figure 2

In vivo genome editing with the AAV.PHP.eB-SaCas9-KKH-Myo6-g2 system (A) Schematic of the AAV vectors (control and g2) and the experimental overview of the in vivo studies. (B and C) The indel percentages at P14 in Myo6^C442Y/C442Y, Myo6^WT/C442Y, and Myo6^WT/WT in the sensory epithelia were determined by targeted NGS. The samples were from ears treated with or without AAV.PHP.eB-SaCas9-KKH-Myo6-g2 at P0–2 via the scala media (n = 3–7). Data are presented as the mean ± SD. (D) The indels causing in-frame versus frameshift mutations were calculated after AAV.PHP.eB-SaCas9-KKH-Myo6-g2 infection in Myo6^C442Y/C442Y and Myo6^WT/C442Y mice. (E) Indel profiles from SaCas9-KKH-Myo6-g2-transfected Myo6^C442Y/C442Y and Myo6^WT/C442Y mice. Negative numbers represent deletions, and positive numbers represent insertions. Myo6^C442Y and Myo6^WT reads are plotted separately. Sequences without indels (value = 0) are omitted from the chart. (F) The most abundant reads in the AAV.PHP.eB-SaCas9-KKH-Myo6-g2-treated Myo6^C442Y/C442Y and Myo6^WT/C442Y mice are shown.

Figure 3

Effects of the AAV-PHP.eB-SaCas9-KKH-g2 virus on the inner ear ABR thresholds in Myo6^WT/C442Y mice (A) ABR thresholds in Myo6^WT/C442Y ears treated with AAV.PHP.eB-SaCas9-KKH-Myo6-g2 (blue), untreated Myo6^WT/C442Y ears (orange), and WT ears (black) after 10 weeks, 15 weeks, and 5 months. At 10 months, the dashed line of the same orange color denotes the mean data of untreated Myo6^WT/C442Y ears, and the solid lines (purple, black, and blue) denote the three individual AAV-SaCas9-KKH-Myo6-g2-treated ears showing mild hearing improvement. Statistical analysis between untreated and AAV.PHP.eB-SaCas9-KKH-Myo6-g2-treated Myo6^WT/C442Y ears was by one-way ANOVA: ∗∗∗p < 0.001; ∗∗p < 0.01; ∗p < 0.05. The statistical analysis between WT and AAV.PHP.eB-SaCas9-KKH-Myo6-g2-treated Myo6^WT/C442Y ears was by one-way ANOVA: ###p < 0.001; ##p < 0.01; #p < 0.05. Values and error bars reflect the mean ± SD. (B) Latency of ABR wave 1 at 8 kHz and 16 kHz at 10 weeks. (C) Amplitude of ABR wave 1 at 8 kHz and 16 kHz at 10 weeks. AAV.PHP.eB-SaCas9-KKH-Myo6-g2-treated Myo6^WT/C442Y ears (blue) were compared with untreated Myo6^WT/C442Y ears (orange). Values and error bars reflect the mean ± SEM. Statistical analysis was by unpaired two-tailed Student’s t tests. ∗∗∗p < 0.001.

Figure 4

Effects of the AAV-PHP.eB-SaCas9-KKH-g2 virus on cell survival rates in Myo6^WT/C442Y mice (A and B) The number of IHCs (A) and OHCs (B) per 100-μm section for four untreated and four AAV-PHP.eB-SaCas9-KKH-g2-treated Myo6^WT/C442Y cochleae from four mice at 5 months. (C and D) The numbers of IHCs (C) and OHCs (D) per 100-μm section for three untreated and three AAV-PHP.eB-SaCas9-KKH-g2-treated Myo6^WT/C442Y cochleae from three mice at 10 months. Values and error bars reflect the mean ± SEM. Statistical analysis was by two-tailed Student’s t tests. ∗∗∗p < 0.001. NS, not significant. (E) Representative confocal images of cochleae harvested at 10 months from the AAV-PHP.eB-SaCas9-KKH-g2-treated and untreated ears of Myo6^WT/C442Y mice. The apical, middle, and basal turns were dissected and stained with MYO7A (red) for hair cells and with DAPI (blue) for nuclei. Scale bar, 50 μm.

Figure 5

Effects of the AAV-PHP.eB-SaCas9-KKH-g2 virus on hair bundle morphology of Myo6^WT/C442Y mice Scanning electron microscope images of the cochlea sensory epithelium showing the hair bundle morphology for untreated Myo6^WT/WT (left), untreated Myo6^WT/C442Y (middle), and AAV.PHP.eB-SaCas9-KKH-Myo6-g2-treated Myo6^WT/C442Y mice (right) at 10 weeks. Representative bundles (arrows) of OHCs and IHCs were enlarged in the corresponding lower panel. Scale bar in low-magnification figures, 10 μm. Scale bar in high-magnification figures, 4 μm. High, high magnification; Low, low magnification.

Figure 6

Effects of the AAV-PHP.eB-SaCas9-KKH-g2 virus on the electrophysiology of Myo6^WT/C442Y mice (A) Representative I_Ca recordings in apical IHCs at P15–16 from untreated Myo6^WT/C442Y mice (orange), WT mice (black), and Myo6^WT/C442Y mice injected with AAV.PHP.eB-SaCas9-KKH-Myo6-g2 at P0–2 (blue). Untreated Myo6^WT/C442Y = −150.84 pA. Treated Myo6^WT/C442Y = −191.71 pA. WT = −291.77 pA. (B) I_Ca values were recorded from a holding potential of −90 mV to +70 mV in 300 ms at P15–16 in IHCs from untreated ears (orange) and AAV.PHP.eB-SaCas9-KKH-Myo6-g2-treated ears (blue). Values and error bars reflect the mean ± SEM. Statistical analysis was by one-way ANOVA. ∗∗p < 0.01; ∗p < 0.05. (C) Representative C_m measurements in IHCs, one from each group. Untreated Myo6^WT/C442Y = 73.91 fF. AAV.PHP.eB-SaCas9-KKH-Myo6-g2-treated Myo6^WT/C442Y = 96.46 fF. WT = 162.16 fF. (D) The net increase in C_m before and after stimulation (ΔC_m) was used to quantify exocytosis from the same IHCs as I_Ca. Values and error bars reflect the mean ± SEM. Statistical analysis was by one-way ANOVA. ns, not significant.