Ethical approval
All protocols involving animals were approved by the ONPRC IACUC and conducted in accordance with the NIH Guidelines for the Care and Use of Laboratory Animals and ARRIVE guidelines. General care and animal housing was previously described47,48.
gRNA design
Web-based algorithms were used to design 4 gRNAs targeting exon 3 of the MYO7A gene with optimal on-target efficiency and minimal off-target effects. Hyb-gRNAs were purchased from Integrated DNA Technologies, and sgRNAs were synthesized and obtained from Synthego.
Target efficiency validation
To confirm the targeting efficiency of the 4 gRNAs, MYO7A target site amplicon was digested with Cas9 nuclease (New England Biolabs) and hyb-gRNAs. A total of 100 ng of wild type MYO7A PCR amplicon was digested by 8 ng of Cas9 nuclease and four different concentrations (50, 10, 2, and 0.4 ng/μL) of hyb-gRNA. Reactions were incubated for 30 minutes at 37°C, 4 μg of RNAase was added to eliminate the hyb-gRNA, and samples were run on a 1.5% agarose gel.
The CMMT cell line was used to test hyb-gRNA efficiencies alone or in combination. At 75-80% confluency, 600 ng of Cas9 plasmid (Addgene, #64324) was transfected using Lipofectamine 2000 (Invitrogen), followed 24 hour later with transfection of hyb-gRNAs using RNAiMax (Invitrogen). A total of 6 pmoles of hyb-gRNA was used for individual hyb-gRNA transfections while 3 pmoles per hyb-gRNA was used for multi hyb-gRNA transfections. Genomic DNA was obtained (QuickExtract, Epicentre) 2-3 days after transfection. The MYO7A target region was amplified and subjected to a T7 endonuclease 1 (T7E1, New England Biolabs) assay according to the manufacture’s protocol using 200 ng of target amplicon or 150 ng target amplicon combined with 50 ng of wild type amplicon were used.
Zygote injection of CRISPR/Cas9
Oocyte collection and in vitro fertilization (IVF) were conducted as previously reported49. Zygotes were stripped free of sperm and transferred to warmed TALP-HEPES under oil. Cas9 nuclease and sgRNAs were prepared as described26. Cas9 mRNA injection buffer contained 10 mM Tris-HCl, 10 mM NaCl, and 0.1 mM EDTA, whereas Cas9 protein injection buffer omitted the NaCl. Injection materials included 45 ng/μL of hyb-gRNAs (15 ng/μL of each) with 100 ng/μL of Cas9 mRNA (Trilink) or 45 ng/μL sgRNAs (15ng/ μL of each) with 224 ng/μL of Cas9 nuclease (New England Biolabs). Injections were into the cytoplasm and pronucleus when visible. Injected zygotes were then cultured (BO-IVC, IVF Bioscience) under oil in a 6%, 5%, 89% mixture of CO2, O2, and N2 at 37°C in humidified air, respectively.
TE biopsy and embryo genotyping
An objective-mounted laser (Hamilton Thorne) was used to biopsy TE cells of blastocysts placed in warmed Biopsy Medium (IVF Bioscience) under oil. The blastocyst was frozen using the global® Blastocyst Fast Freeze® Kit (LifeGlobal Group). Arrested embryos with the zona pellucida removed by acid Tyrode’s solution, TE biopsy samples, and in some instances, biopsied blastocysts were subjected to WGA (Repli-g Single Cell, Qiagen). The resultant DNA was used to PCR amplify the MYO7A target region and amplicons were subjected to next-generation sequencing (NGS) using NEXTFLEX® Rapid DNA-Seq Kit (PerkinElmer). The DNA library was sequenced on an Illumina MiSeq platform and sequence reads were assessed for indel mutations (Geneious software).
Off-target editing analysis
Putative off-target sites were identified using CAS-OFFinder50. Exons were selected over introns and intergenic regions, and ranked based on off-target scores. Using these criteria, three potential off-target sites for each gRNA target (Supplementary Table 2) were amplified by PCR. Resultant amplicons were subjected to adapter ligation and sequenced on an Illumina MiSeq platform. CRISPresso2 software was used for sequence analysis51.
Embryo transfer
Embryos selected for transfer were thawed (Global® Blastocyst Thawing Kit, Life Global) and maintained for 2 hours in 100 µL drops of BO-IVC under oil to monitor post-thaw viability. Embryos with blastocoel reformation or expansion were transferred into recipients as previously described52. A transfer catheter was inserted into the oviduct and the embryo was deposited in 10 µl of TALP-HEPES. After embryo transfer, pregnancy was confirmed by ultrasound at gestational day 28.
MYO7A genotyping
Genomic DNA was extracted from blood, buccal swabs, skin, and placental tissue and used to PCR amplify the targeted region of MYO7A (primers detailed in Supplementary Table 3). PCR amplicons were purified and prepared for NGS sequencing (NEXTFLEX® Rapid DNA-Seq Kit) using an Illumina MiSeq platform with roughly 5-10 thousand reads per sample. Geneious software and the CRISPResso pipeline were used for sequence analysis. Single cell genotyping was conducted on leukocytes isolated from whole blood. Using a Becton Dickinson Aria-II FACS, cells were sorted by morphology and then underwent WGA, with the resultant DNA being used for PCR amplification of MYO7A exon 3. Purified amplicons were then Sanger sequenced.
Assessment of auditory function
The animal was anesthetized and placed within an IAC Acoustics MAC-3 RF-shielded Mini-Acoustical Chamber. The patency of the distal ear canal was confirmed by otoscopic evaluation. Middle ear function was tested at 1 kHz or 226 Hz by the ZODIAC Diagnostic Tympanometer (Model 8-04-16045). For 1-month and 3-month testing, stimuli were delivered in the ear canal via a custom acoustic assembly comprised of two dynamic drivers (Etymotic Research ER4B-SG) and a miniature electret condenser microphone (Knowles FG-23329-P07). For testing at 6- and 12-months, stimuli were delivered with an open field speaker (Fostex FF85WK) calibrated and monitored with a ¼-inch condenser microphone (Larson Davis Model 377C10).
Acoustic stimuli were conducted as previously reported53. Five-ms tone pip stimuli with 0.5-ms rise/fall times (cosine squared) in alternating polarity were presented in an interleaved stimulus train from 20-85 dB SPL in incremental steps of 5 dB at each frequency tested. The order of frequencies was 26, 12, 4, 0.5, 16, 8 and 2 kHz presented at a stimulus rate of 60/sec. The response from the electrodes was amplified 10,000x, filtered at 300 Hz–3 kHz band pass, and averaged for 300 samples at each frequency-level combination. Responses were recorded from subcutaneous electrodes located at the vertex, the ipsilateral mastoid, and the shoulder.
DPOAEs were recorded using the ear canal sound delivery apparatus controlled by Eaton Peabody Laboratory Cochlear Function Test Suite software (ver. 2.6.3.831). Primary f2 tones were presented at 2, 4, 12, 16 and 26 kHz with a primary frequency ratio f2/f1 of 1.2 and the f2 primary level 10 dB less than the f1 level. Primaries were incremented in 10 dB steps, with f2 from 20 to 70 dB SPL. Ear-canal sound pressure was amplified, digitally sampled and averaged, Fast Fourier Transforms were computed and averaged, and the DPOAE and noise floor at 2f1-f2 were extracted.
Assessment of retinal structure and function
Retinal structure was evaluated under isoflurane anesthesia with retinal imaging, including color fundus photography and fluorescein angiography (Zeiss FF450), sdOCT (Heidelberg Spectralis) and quantitative fundus autofluorescence54 (Spectralis Blue Peak mode), and ultra-widefield imaging (Optos California)55-58. Retinal function was assessed with full-field electroretinography (Espion Diagnosys), and visual acuity was measured with a behavioral test (Teller Acuity Cards). For each of these measures, results for Mya were compared to age-matched normal rhesus infants and measurements.
OCT imaging parameters included a standard scan (30° x 25°; ART:20; 61 lines with ~120 µm spacing), plus two high density scans (#1:15° x 10°; ART:20; 49 lines with ~60 μm spacing; and #2: 15° x 5°; ART:20; 97 lines with ~30 μm spacing) for characterization of the macula/fovea and any salient features of interest. The directional OCT method was performed as described to obtain an accurate measurement of the foveal outer nuclear layer59. Spectralis segmentation software (EyeExplorer 1.10.4) was used to obtain layer thicknesses, with manual inspection and corrections of each slice. Corneal curvature and axial length were measured with an IOL Master at each retinal imaging session to allow conversion of retinal parameters from degrees of arc to mm. Ultra-wide field imaging in the peripheral retina, including FAF, fluorescein angiography of the retinal vasculature, and indocyanine angiography of the choroidal vasculature, was performed to identify peripheral changes associated with retinitis pigmentosa58. Full-field ERGs were recorded with a Diagnosys Espion system using custom bipolar Burian-Allen contact lens electrodes and included photopic, flicker and scotopic intensity-response series, and a dark adaptation series.
Statistics analysis
To compare the results of embryo development and target gene editing between the two conditions (mRNA and Protein), we conducted a two-tailed Fisher’s Exact Test. A P value < 0.01 was considered statistically significant.