CRISPR/Cas9-mediated PINK1 deletion leads to neurodegeneration in rhesus monkeys

Dear Editor, PINK1 mutations cause autosomal recessive and early-onset Parkinson’s disease (PD) with selective neurodegeneration. Unfortunately, current PINK1 knockout (KO) mouse and pig models are unable to recapitulate the selective and overt neurodegeneration seen in PD patients. Furthermore, endogenous Pink1 in the mouse brain is expressed at very low levels and can only be detected via immunoprecipitation, meaning that PINK1’s function in the mammalian brain needs to be assessed using larger animals that are closer to humans. We previously used CRISPR/Cas9 to target the monkey gene in one-cell stage embryos. Using the same approach, we designed two gRNAs to target exon 2 (T1) and exon 4 (T2), which encode a kinase domain in the PINK1 gene of rhesus monkeys (Fig. 1a). CRISPR/Cas9 and PINK1 gRNAs were injected into one-cell stage rhesus monkey embryos. A T7E1 assay and sequencing of PCR products from the injected embryos showed high efficiency (61.5%) in targeting PINK1 (Fig. 1b and Supplementary information, Fig. S1). Transfer of 87 embryos to 28 surrogate rhesus monkeys resulted in 11 pregnancies (39.2%) (Fig. 1b). Eleven fetuses developed to term and were born naturally. Of these live monkeys, eight carried PINK1 mutations (M), and three were wild type (WT). However, three mutant monkeys (M1, M3 and M4) were newborn triplets that struggled to survive and died 3-4 days after birth. One WT newborn monkey also died after a difficult labor. Another mutant monkey (M2) died 7 days after birth without noticeable warning signs or symptoms. The other three mutant monkeys (M6, M7 and M8) have lived for three years; M5, however, reduced its food intake and showed weakness at the age of 1.5 years, and died 30 days after anesthesia for MRI examination. The mosaicism of CRISPR/Cas9-mediated mutations can induce different extents of PINK1 loss and phenotypes, allowing us to examine the relation of PINK1 mutations with PINK1 expression and pathological changes. Indeed, a T7E1 assay and sequencing analysis of the targeted DNA regions revealed various types of DNA mutations (Supplementary information, Fig. S1). Importantly, we identified a large deletion (7,237 bp) between exon 2 and exon 4 in dead monkey tissues via PCR and sequencing of PCR products, as well as whole-genome sequencing (Fig. 1c and Supplementary information, Fig. S2). T7E1 analysis of several potential off-target genes in brain cortical tissues from PINK1 mutant monkey (M1, M2, M3, and M4) and blood tissues from live PINK1 mutant monkeys (M5, M6 and M7) revealed no mutations (Supplementary information, Fig. S3a). Whole-genome sequencing of M1, M2, and M3 monkeys showed no significant mutation rates in the top 20 potential off-target genes (Supplementary information, Fig. S3b) and analysis of 2,189 possible off-target sites with up to five mismatches of the gRNA sequences in the genome also revealed no off-targeting events (Supplementary information, Fig. S3c). The large deletion in PINK1 is different from point mutations found in humans and should completely eliminate PINK1 expression. To provide more evidence for the specific targeting of the PINK1 gene and the resulting phenotype, we used western blotting to assess PINK1 expression and PCR to evaluate the relative degree of the large deletion by detecting the ratio of truncated DNA resulting from this deletion to the remaining intact exon 3 DNA (Fig. 1c–e). We found that ~65%–70% of PINK1 alleles in M1 cortex and striatum and M2 cortex carry the ~7.2 kb deletion (Fig. 1f). Western blotting analysis of PINK1 mutant monkey brains also confirmed differing extents of deficiency in PINK1, neuronal proteins (NeuN, PSD95, CRMP2, and SNAP25), and doublecortin (DCX) (Fig. 1e). All our results clearly showed that M1 and M2 monkey cortical tissues had the highest degree of the large PINK1 deletion and the lowest level of PINK1 and neuronal proteins (Fig. 1g). Counting NeuN-positive cells also verified that M1 and M2 cortical and M1 striatal tissues had significantly fewer neuronal cells than WT controls (Fig. 1h, i). Moreover, we verified that PINK1 is also abundantly expressed in the human brain (Fig. 1j). For the live monkeys, MRI and video monitoring studies revealed that 1.5-year-old adult monkeys with PINK1 mutations showed significantly decreased gray matter density in the cortex (Supplementary information, Fig. S4a-c). M5 and M6 monkeys also displayed decreased movement despite no alteration in sleep behavior (Supplementary information, Fig. S4d, e, Movie S1). The M5 monkey lived up to 1.5 years and died 30 days after MRI examination. T7E1 assay revealed PINK1 mutations in its brain and peripheral tissues (Supplementary information, Fig. S5a). Immunohistochemical studies showed reduced density of NeuNpositive neuronal cells and increased GFAP staining in the cortex and striatum of the M5 monkey compared with a 1.5-year-old WT monkey (Supplementary information, Fig. S5b). Because M5 monkey brain tissues were not isolated immediately after death for electron microscopic (EM) examination, we euthanized the symptomatic M6 monkey at the age of three years for EM to provide ultrastructural evidence for neurodegeneration. Analysis of M6 monkey brain genomic DNA also revealed the large deletion between PINK1 exon 2 and exon 4 in various tissues (Supplementary information, Fig. S2b), and western blot analysis showed significantly decreased PINK1 expression in the cortex and substantia nigra compared with an age-matched WT monkey (3year-old) (Fig. 1k). EM revealed degenerated neurons in the cortex, substantia nigra and striatum, as characterized by their electrondense cytoplasm with no clear organelles and no identifiable nuclear membrane (Fig. 1l). Interestingly, in those degenerated neurons, the mitochondrial morphology is indistinguishable from WT monkey neurons. The remarkable neuronal loss seen in PINK1 mutant monkeys was not reported in PINK1 KO mice or pigs, and may be

Male macaques were electro-ejaculated with a current isolation stimulator (JL-C4 V2a, JIALONG, China) equipped with electrocardiographic pad electrodes for direct penile stimulation (30-50 V, 20-ms duration, 18 pulses/s). Semen samples were collected into 15-mL tubes. Ejaculated sperm were diluted to 2 × 10 5 in 10% polyvinylpyrrolidone (PVP) to reduce motility and placed in a separate drop on the manipulation dish. A single sperm aspirated from the sperm drop into the injection needle, was transferred to the oocytes in the TALP-Hepes drop. MII oocytes immobilized with a holding pipet on the polar body at the 6 o'clock or 12 o'clock position, and then injected with a sperm through a needle through the zona into the cytoplasm (ICSI). After ICSI, oocyte was washed twice in Hamster Embryo Culture Medium 9 (HECM-9) before being transferred into a pre-equilibrated 50 μL drop of HECM-9, covered with mineral oil and incubated at 37.5 °C with 6% CO2 for 8-10 h.
Oocytes with a second polar body and two pronuclei arising after ICSI were confirmed successful fertilization. Zygotes were injected with Cas9 mRNA (200 ng/L) and gRNAs (50 ng/L), and the injected zygotes were cultured for embryo development. Embryos at 4-8 cell stages were used for transfer or divided into single blastomere for PCR. The pronuclear formation was recorded 16-20 h post-ICSI, and the progression of embryo growth was recorded daily.
Microinjections were performed in the cytoplasm of zygotes using a Narishige (Narishige Inc. Japan) microinjection system under standard conditions. The zygotes were cultured in embryo culture medium-9 (HECM-9) containing 15% fetal calf serum (Hyclone Laboratories, SH30088.02) at 37.5 °C in 6% CO2. Cleaved embryos of high quality at the 4-cell stage were transferred into the oviduct of the matched recipient female monkeys. Typically, three embryos were transferred into each female. Ultrasonography detected pregnancies 30-35 days after the embryo transfer. Both clinical pregnancy and number of fetuses were confirmed by fetal cardiac activity and presence of a yolk sac as detected by ultrasonography.

Western blot analysis, immunohistochemistry and electron microscopy
Human brain hippocampus tissue was obtained with informed consent under a protocol approved by The First Affiliated Hospital of Jinan University Institutional Review Board.
Adult human brain hippocampal specimens involved in epileptic foci were obtained from surgeries for treating epilepsy. Tissue was immediately frozen in liquid nitrogen after resection. Human or monkey brain tissues were lysed in ice-cold RIPA buffer (50 mM Tris, pH 8.0, 150 mM NaCl, 1 mM EDTA, pH 8.0, 1 mM EGTA, pH 8.0, 2% SDS, 0.5% DOC, 50 mM NaF and 1% Triton X-100) containing Halt protease inhibitor cocktail (Thermo Scientific) and PMSF. The lysates were incubated on ice for 30 min, sonicated, and centrifuged at the maximum speed for 10 min. Equal amounts of proteins from the supernatants determined by For immunohistochemistry, monkey brain tissues were fixed overnight (12-16 h) in 4% paraformaldehyde in 0.01 M PBS, and then transferred into 30% sucrose at 4 °C to let the brain completely sink to the bottom of the tube. Brain tissue was sectioned at 20 μm using a cryostat at −19 °C. Monkey tissue slides were fixed for 10 min in 4% paraformaldehyde in 0.01 M PBS at room temperature, blocked with 0.1% Triton X-100/2% NGS/3% BSA/1× PBS for 30 min, and incubated with primary antibodies to relative proteins in 3% BSA/2% NGS/1× PBS overnight at 4 °C. The slices were washed three times with 1× PBS and rinsed with secondary antibodies.
For electron microscopy (EM), M6 and age-matched (3 year old) monkeys were deeply anesthetized by intraperitoneal injection of 0.3-0.5 mL of atropine, followed by 10-12 mg of ketamine and 15-20 mg of pelltobarbitalum natricum per kg body weight. The freshly isolated brain tissues from sacrificed M6 and age-matched (3 year old) monkeys were fixed with 2.5% glutaraldehyde/0.1 M PB overnight at 4 °C. Brains were sectioned into 50 μm using a vibratome (Leica, VT1000s) and the sections were processed for electron microscopic examination. In brief, all sections were osmicated in 1% OsO4 in 0.1 M PB and embedded in Eponate12 (Ted Pella). The dried brain sections were cut into ultrathin sections (60 nm) with a Leica Ultracut S ultramicrotome under a Hitachi H-7500 transmission electron microscope equipped with a Gatan Bio-Scan CCD camera at Emory University.

Off-target analysis and whole genome sequencing analysis
Potential off-target sites (OTs) of sgRNA guided Cas9 endonucleases for the PINK1 gene were predicted by a bioinformation-based search tool (http://www.rgenome.net/cas-offinder/). Wholegenome sequencing were performed to analyze the off-target loci. Top 20 off-target loci in the gRNA targeted sequences were selected by blasting the PINK1 gRNAs sequences in the monkey genome for analyzing and comparing with PINK1 targeting rates, which were determined by CRISPR/Cas9 targeted and non-targeted monkeys.
Novogene prepared the DNA library preparation and conducted whole genome sequencing on the monkey genomic DNA. The edited monkey genome (Macaca mulatta) was used to produce a custom-made index file, and PEMapper was then used to map sequencing data to the custom-made index file. For each on-target and off-target locus, 1 kb of flanking region to each side of the locus was added. We next used the pileup file generated by PEMapper to retrieve the base pair-level sequencing read coverage and reported the average.
For large deletion test, short reads were aligned using the Burrows-Wheeler Aligner (BWA) with default parameters. SAMtools was used to convert between SAM and BAM file format, and Picard tools to sort alignments. Sorted BAM files were uploaded to UCSC genome browser (http://genome.ucsc.edu) for visualization and analysis.
Genomic sites with sequence similarity to the T1 and T2 gRNAs were identified by a base-bybase scan of the entire monkey genome, allowing for ungapped alignments with up to 5 mismatches. Off-target sites juxtaposed to an 'NGG' PAM site were identified by comparing with the genomic DNA sequences of wild-type rhesus monkeys from University of Nebraska Nonhuman Primate Genome Center (http://www.unmc.edu/rhesusgenechip/index.htm.) and Ensemble (Macaca mulatta *8.0.1 edition).

Monkey behavioral studies
Monkey behavioral studies were conducted at the Institute of Laboratory Animal Science, Chinese Academy of Medical Sciences. Four PINK1 mutant monkeys (1.5 years of age) and four agematched control monkeys were examined. Monkeys were individually observed in an observation cage (1.4 × 0.9 × 1.15 m) that was similar to their home cage. All movement activities were videorecord without interruption for 30 min each day for 6 consecutive days. For sleep examination, monkeys were video recorded from 8 pm to 6 am. The wake-up times and duration of each wakeup were measured by a Vigie Primates image analyzer system (View point, Lyon, France, Version: 4.7.0.520). Three experiments were performed on each monkey, and each experiment monitored the sleep behavior for five consecutive days.

Statistical analysis
Two-tailed Student's t-test (unpaired) was used to compare differences between age-matched control and mutant monkeys. Statistical analyses were performed with Excel. Values are represented in the text as mean ± SEM.     Fig S4   Supplementary information, Fig. S4. MRI imaging data and movement activity of live PINK1 mutant monkeys. a MRI shows reduced gray matter density in live PINK1 mutant monkeys. In each row, 5 consecutive axial slices are shown with slice thickness of 0.3 mm, and the rightmost sagittal image indicates the position of those slices. Five brain regions indicated by numbers in yellow boxes represent different areas in the gray matter. b Based on 2D imaging data in a, 3D images consisting of clusters in a glass brain was constituted to show the alteration in the density of gray matter. Cluster 1: the right parietal cortex, Cluster 2: the left parietal cortex, Cluster 3: the right pons and medulla, Cluster 4: the right parietal cortex, Cluster 5: the left putamen. c Quantitative summary of the reduced gray matter density in PINK1 mutant monkey brains. The gray matter density is decreased in distinct brain regions in PINK1 mutant monkeys at the age of 1.5 year (n = 4 for PINK1 mutant monkeys (M5, M6, M7, M8) and n = 4 for age-matched WT controls). Two of them are in the right parietal cortex, 1 in the left parietal cortex, 1 in the left putamen, and 1 in the right pons and medulla. No significant difference was identified in the intracranial volume, tissuespecific volumes, or regional volumes between mutant (M) and WT. The values (mean ± SE) from distinct brain regions (clusters 1-5) were shown. **P < 0.01. Statistical differences between groups were determined by Student's t-test. d, e Movement activities (d) and sleep time (e) of live PINK1 mutant monkeys and age-matched wild-type control animals at 1.5 years of age. Each monkey was examined 6 times for consecutive 6 days, and the average values (mean ± SE) were presented. When compared with the WT controls, M5 and M6 showed reduced distance and duration in movement. The WT group (n = 4) was used to compare with each individual PINK1 mutant monkey using Student's t-test. *P = 0.0318. NS: not significant.  H e a rt K id n e y L u n g L iv e r In te s ti n e * * * * * * * * * * * * * * * * * * * * b