Generation of gene-modified goats targeting MSTN and FGF5 via zygote injection of CRISPR/Cas9 system

Recent advances in the study of the CRISPR/Cas9 system have provided a precise and versatile approach for genome editing in various species. However, the applicability and efficiency of this method in large animal models, such as the goat, have not been extensively studied. Here, by co-injection of one-cell stage embryos with Cas9 mRNA and sgRNAs targeting two functional genes (MSTN and FGF5), we successfully produced gene-modified goats with either one or both genes disrupted. The targeting efficiency of MSTN and FGF5 in cultured primary fibroblasts was as high as 60%, while the efficiency of disrupting MSTN and FGF5 in 98 tested animals was 15% and 21% respectively, and 10% for double gene modifications. The on- and off-target mutations of the target genes in fibroblasts, as well as in somatic tissues and testis of founder and dead animals, were carefully analyzed. The results showed that simultaneous editing of several sites was achieved in large animals, demonstrating that the CRISPR/Cas9 system has the potential to become a robust and efficient gene engineering tool in farm animals, and therefore will be critically important and applicable for breeding.

in a variety of species, including model organisms, as well as crops and animals that are crucial to agriculture.
As one of the first domestic farm animals, the goat is one of the most important livestock species and provides a variety of products, including fiber, milk, meat, and hides. Furthermore, goats have also been used as a model in biomedical studies [3][4][5] . Although specific gene knockout strategies based on homologous recombination (HR) and somatic cell nuclear transfer (SCNT) have been established in goats 6 , precise gene modification of the goat genome is still challenging. In a previous study on disruption of four genes simultaneously in goat primary fibroblasts by the CRISPR/Cas9-mediated approach, only the myostatin (MSTN) knockout fibroblasts were achieved and resulted in live-born goats by SCNT 7 . However, an anti-biotic selection cassette is normally essential for isolating single-cell colonies from seeded donor cells, and reconstructed embryos have a low developmental potential, leading to a relatively low SCNT targeting efficiency (for example, 1-5% in pigs and >10% in cattle) 8 .
Co-injection of Cas9 mRNA and sgRNA into one-cell stage embryos has been demonstrated to be an efficient approach for the generation of genetically modified mice, rats, monkeys and pigs [9][10][11][12] , which encourages us to extend the application of this strategy to gene targeting in goats. In the present study, through co-injection of one-cell stage embryos of cashmere goats with sgRNAs of two functional genes (MSTN and FGF5) and Cas9 mRNA, targeted modifications of one or two genes were achieved at an efficiency around 26.5%. We also carefully analyzed the on-and off-target mutations of the targeted genes in the somatic tissues and gonads, providing comprehensive evidence for the efficiency and reliability of injection of zygotes with Cas9 mRNA and sgRNAs for the generation of gene-modified farm animals.

Results and Discussion
Design of sgRNAs. The Shannbei cashmere goat is a cultivated dual-purpose breed that provides both meat and fibers (fine cashmere). In an attempt to improve the performance of cashmere goats considering both meat and cashmere production purposes, two genes that are associated with muscle development (MSTN) and hair length (FGF5), were selected as target genes. MSTN is a secreted growth differentiation factor that inhibits muscle differentiation and growth. Function loss of MSTN is known to cause an increased muscle mass phenotype in several mammals, including mice, dogs, cattle and humans [13][14][15][16] . Fibroblast growth factor 5 (FGF5), a secreted signaling protein during the hair growth cycle, inhibits hair growth by blocking dermal papilla cell activation 17 , and is regarded as the causative gene underlying the angora phenotype (long hair coat) in mice 18 . Mutations in FGF5 underlie trichomegaly (excessively long eyelashes) in humans 19 , as well as being associated with hair length in other mammal species such as cats 20 , dogs 21 and donkeys 22 . MSTN and FGF5 were therefore selected to generate gene-modified cashmere goats with both genes disrupted by the CRISPR/Cas9 system. To ensure the successful targeting, two sgRNAs independently targeting exons 2 and 3 of MSTN, and two sgRNAs targeting the first exon of FGF5, were selected as described 23 (Fig. 1a).

Efficiency of the CRISPR/Cas9 System in Fibroblasts. Although microinjection of zygotes with
Cas9 mRNA and sgRNA had not previously been tested in goats, the successful application of this system in embryos of other species 12,24 , suggested that this system could work in goat embryos. To verify this possibility, we overexpressed the Cas9 nuclease and sgRNAs targeting MSTN and FGF5 in goat fibroblasts isolated from goat foetus. Subsequently, genomic DNA was extracted from fibroblasts 72 h after transfection, and screened for the presence of site-specific gene modification by PCR amplification of the region around the targeted site (Fig. 1b, c and e) using a T7 endonuclease I (T7E1) cleavage assay. The T7E1 cleavage bands were visible in the targeted genes (Fig. 1b, c and f). The cleavage was characterized further by Sanger sequencing, which displayed overlapped peaks in the sequencing chromatographs ( Fig. 1d,g), suggesting different genotypes from target modifications. Distinguishable indels were observed at the target sites of MSTN and FGF5 with a range of mutation sizes (Fig. 1d,g, Supplementary  Table S1). These data demonstrated that the designed sgRNAs work efficiently with Cas9 on targeted genes in the cultured goat fibroblasts.

Generation of Gene-Modified Goats.
To establish target modification in goats, a total of 926 early zygotes (one-cell stage) were surgically collected from 79 naturally mated ewes; ~11.72 zygotes were obtained from each donor. The Cas9 mRNA and sgRNA mixture targeting MSTN and FGF5 was injected into the cytoplasm of the embryos at the one-cell stage. After the injected embryos developed to the two-cell stage (~24 h, 37 °C), only 416 (48.25%) of the 862 injected embryos were transferred into 137 pseudopregnant mothers. On average, 3.04 injected embryos were transferred to each recipient (Table 1). Of the 137 recipient ewes, 64 pregnancies (46.7%) were established according to observation of the estrus cycles. After full-term gestation, which lasted around 150 days, 93 lambs from surrogates were successfully delivered, 14 of which died immediately after birth ( Fig. 2a and Table 1).
Genomic DNA was isolated from aborted, dead and live lambs. In total, 98 individuals (including 93 delivered and 5 aborted) were used for genotyping. The sgRNA:Cas9-mediated genome modifications were first screened using the genomic DNA as described above. Genotyping was performed by PCR amplification, T7E1 assay, and TA-cloning sequencing. The additional bands were observed by PCR amplification of the target region in some of the infant goats (Fig. 2b), indicating that genomic modification occurred in these goats. Then, the PCR products of all the animals were subjected to the T7E1 cleavage assay (Fig. 2c). Impressively, the cleavage products for MSTN were observed in 15 infants (15.3%) and FGF5 in 21 infants (21.4%), with 10 lambs showing disruption of two genes simultaneously ( Table 2). In total, 26 lambs (26/98, 26.5%) demonstrated the disruption of one or two genes, indicating efficient genomic modifications in the infant goats. As expected, different types of indels were observed, and further confirmed by Sanger sequencing (Fig. 2d).
Currently, there are two descriptions about generation of genetically modified goat and sheep via the CRISPR/Cas9 system. Ni et al. reported the application of the CRISPR/Cas9 system to disrupt four genes (MSTN, NUP, PrP, and BLG) in goat primary fibroblasts, and generated live-born goats via SCNT approach for the single MSTN gene. Only the MSTN biallelic mutations were used for SCNT; seven pregnancies were yielded from 21 transfers, and two live-born goats were eventually obtained 7 . Through co-injection of Cas9 mRNA and sgRNA targeting a single gene-MSTN, Han et al. described the successful generation of MSTN knockout sheep at a targeting efficiency of 5.7% (2/35) 25 . Taken together, our results support that direct microinjection of Cas9 mRNA and sgRNAs into one-cell stage embryos is efficient in goats, and could be applied to produce gene-modified animals.

Cas9-Mediated Genome Targeting Extensively Integrates into Different Tissues of Goats.
To further evaluate the integration of the Cas9-mediated modifications into the derivatives of three germ layers, we performed extensive analysis of target mutagenesis in seven different somatic tissues (heart, liver, spleen, lung, kidney, skin, and muscle) and testis from the aborted and dead lambs. PCR amplification and T7E1 cleavage assay showed similar cleavage patterns among different tissue (Fig. 3a,b), suggesting that the Cas9-mediated mutations did occur in various goat tissues.
Off-Target Detection. Previous studies have described that off-target mutagenesis occurred in Cas9 medicated human cell lines 25 , mice 26 and zebrafish 27 , suggesting that mismatches between the sgRNA and target DNA exist due to Cas9-mediated DNA cleavage 25 . To verify whether off-target mutations occurred in these Cas9 mediated goats, we predicted putative off-target sites using SeqMap 28 and assessed the off-target mutations in both fibroblasts and founder animals. A total of 13 (6 for MSTN and 7 for FGF5) most potential off-target sites were predicted across the goat genome (Supplementary Table S2). Out of the 13 off-target sites, two sites (OT5 and OT7) showed off-target modifications in goat fibroblast cells (Supplementary Table S3, Fig S1). The remaining 11 off-target sites were therefore excluded for further examination in the founder animals. Subsequently, the PCR products at OT5 and OT7 were subjected to the T7E1 cleavage assay using the DNA from founder animals, we found that mutations at these two sites indeed occurred in some of the founder animals (Fig. 4a,b). To further verify the off-target cleavage events, PCR products from MSTN-disrupted animals and FGF5-disrupted animals were subjected to Sanger sequencing, showing #43, #70 of MSTN mutated founders and #9, #43, #70, #82, #84, #93 of FGF5 mutated founders had off-target modification (Fig. 4c). Except #41 of FGF5, which had a TGG-------GGA/AGGAGGGGTGGGG SNP 113 bp before OT7, no mutations were found in #53 and #81 of FGF5 mutated founders, indicating the slight off-target efficiency. In addition, #46 and #47 of FGF5 showed false positive cut bands due to confirmed poly-T structure. Considering the off-target effects are site-dependent, we believe the off-target mutagenesis can be avoided by choosing a single or several sites.
Taken together, we present here the successful generation of gene-modified goats targeting multiple genomic loci via the CRISPR/Cas9 approach. We further carefully analyzed the sgRNA:Cas9-mediated on-and off-target mutation in various somatic tissues, as well as testis, providing detailed information about generating gene-modified goats by injection of zygotes with Cas9 mRNA and sgRNAs.

Materials and Methods
Ethical statement. Eleven rams (2-3 years old, body weight: 30-50 kg) and 216 ewes (79 donors and 137 recipients, 2-3 years old, body weight: 24-40 kg) were used in this study. The animals were regularly maintained in the Shaanbei Cashmere Goat Farm of Yulin University. All the protocols involving the use of animals were in accordance with approved guidelines of the Animal Care and Use Committee of the Northwest A&F University (Approval ID: 2011ZX08008-002).   Supplementary Table S4 were annealed to be double-stranded, and then sub-cloned into the pUC57-T7-gRNA vector as described 9 . The constructed recombinant vector was completely linearized by DraI endonuclease and used as the template. sgRNAs were produced by in vitro transcription using the MEGAshortscript kit (Ambion) and purified using the MEGAClear kit (Ambion) according to the manufacturer's instructions. Using the Cas9 mRNA in vitro transcription vector (Addgene No. 44758) as templates, Cas9 mRNA was produced and purified according to description 9 . The sequence of both the Cas9 and sgRNAs targeting MSTN and FGF5 are listed in Supplementary Table S5.
Cas9/sgRNA efficiency test in goat fetal fibroblasts. We established a fibroblast cell line from 40-day-old goat fetus trunk tissue. The fibroblasts went through five passages cultured in DMEM medium (Gibco) supplemented with 10% FBS (Gibco) and 1% penicillin streptomycin (Gibco) to achieve 80-90% confluency on the day of transfection. The transfection procedure was carried out using Lipofectamine 2000 Reagent (Invitrogen) according to the manufacturer's instructions. Briefly, goat fibroblasts were transfected with MSTN sgRNA1 (0.8 ug), MSTN sgRNA2 (0.8 ug), FGF5 sgRNA1 (0.8 ug), FGF5 sgRNA2 (0.8 ug) and FGF5 sgRNA1 & 2 (0.4 ug/each), along with 0.8 μ g of Cas9 plasmid by Lipofectamine 2000 in a 24-well culture plate. 24 h after transfection, 10 mg /mL of blasticidine S hydrochloride was added to the medium at a 1:1000 dilution for 24 h. Genomic DNA was extracted from fibroblasts 72 h after transfection using a saturated solution of phenol and chloroform, then the DNA was precipitated with alcohol and sodium acetate. Subsequently, a T7E1 cleavage assay was performed as described by Shen et al. 9 .
Briefly, the targeted fragments were amplified by PrimerSTAR HS DNA polymerase (TaKaRa, DR010A) from the genomic DNA, then purified with a PCR cleanup kit (Axygen, AP-PCR-50). The primers for amplifying MSTN and FGF5 targeted fragments are listed in Supplementary Table S6. The purified PCR products were denatured and re-annealed in NEBuffer 2 (NEB) using a thermocycler (BioRad). The PCR products were digested with T7E1 (NEB, M0302L) for 30 min at 37 °C and then separated on a 2.5% agarose gel. The PCR products with mutations detected by the T7E1 cleavage assay were then sub-cloned into T vector (Takara). For each sample, the colonies were picked randomly and sequenced with the M13F (-47) primer: 5′ -CGC CAG GGT TTT CCC AGT CAC GAC-3′ ).   Production of gene-modified goats via zygote injection with Cas9/sgRNA. Healthy ewes with regular estrus cycles were selected as donors for zygote collection. Zygotes were collected through surgical oviduct flushing from the donors by estrus synchronization and superovulation treatment as we previously described 30,31 . In brief, donors were treated with EAZI-BREED CIDR Sheep and Goat Device (CIDR, contain progesterone 300 mg) by inserting in vagina for 14 days, and the superovulation was performed 60 hours prior to CIDR Device removal. A total of 260 mg follicle stimulating hormone (FSH) (Folltropin ® -V) was administered by intramuscular injection in 7 dosages, at 12 h intervals (the first dose was 70 mg and other doses were decrease progressively to 25 mg). And subsequently injected with 0.1 mg cloprostenol after 60 hours from FSH was administered. Estrous detection was carried out 12 h after CIDR withdrawal and mating were repeated at 8 h intervals.
Goat zygotes at the one-cell stage (around 14 h post-fertilization) were surgically collected and were immediately transferred into TCM199 medium (Gibco, NY, USA). Cas9 mRNA (2 ng/μ L) and sgRNAs (5 ng/μ L for each sgRNA) targeting MSTN and FGF5 were mixed and injected into the cytoplasm of fertilized oocytes using the FemtoJect system (Eppendorf, Hamburg, Germany). The injection pressure, injection time and compensatory pressure were 45 kpa, 0.1 s and 7 kpa, respectively. Microinjection was conducted in manipulation medium TCM199 on the heated platform of the Olympus micromanipulation system ON3. After injection, the zygotes were cultured in Quinn's Advantage Cleavage Medium (Sage Biopharma, NJ, USA) for 24 h at 37 °C, and then transferred to Quinn's Advantage Blastocyst Medium (Sage Biopharma, NJ, USA) at 37 °C, 5% concentration of carbon dioxide and saturated humidity conditions. The surrogate animals for transfer were determined according to their oestrus cycles. About three divisive embryos were transferred into the ampullary-isthmic junction of the oviduct of the T7E1 cleavage assay and sequencing. Samples including blood, heart, liver, spleen, lung, kidney, skin, testis, and muscle, were collected and digested in lysis buffer (0.4 M NaCl, 2 μ M EDTA, 1% SDS, 10 μ M Tris-HCl, and 100 μ g/ml Proteinase K). The genomic DNA of the sample was extracted from the lysate by the phenol-chloroform protocol, and recovered by alcohol precipitation. A T7E1 cleavage assay was performed as described 9 .
Off-target assay. To determine the site-specific cleavage of the CRISPR-Cas9 system in vivo, the potential off-target loci were searched using an open tool, SeqMap 29 . The mismatch parameter for the target sequence was set as 5. 'NGG' and 'NAG' were chosen as the protospacer adjacent motifs (PAM). The sites with 7 bp conserved proximal to PAM and total mismatches < 5, and sites with total mismatches <4, were chosen as potential off-target sites for subsequent testing. The selected potential off-target sites were initially PCR-amplified using genomic DNA from cultured goat fibroblasts. The PCR products were then subjected to a T7E1 cleavage assay. The potential off-target sites yielding the typical cleavage bands were considered as candidates and further evaluated in the founder animals by PCR amplification, T7E1 cleavage assay and TA sequencing. The information on the off-target loci and primer pairs used are listed in Supplementary Tables S7.