Generation of heritable double muscle buttocks rabbits via myostatin mutation with CRISPR/Cas9 system

Background Myostatin (MSTN) is a member of the transforming growth factor (TGF-β) superfamily and is considered to be a negative regulator that inhibits muscle development and regeneration, inactivity of MSTN gene may affect the development and regeneration of muscle in a few of animals. Results In the present study, the muscle mass negative regulator gene myostatin (MSTN) was knocked out at two novel sits in exon3, and the function of these mutation was determined in the offspring rabbits. The typical double muscle phenotype with hyperplasia and hypertrophy of muscle ber was observed in the MSTN KO rabbits, and similar phenotype was conrmed in the F1 generation rabbits. The body weight of MSTN-KO +/− rabbits 3708 ± 43.06 g were signicant heavier at 180 days than that of control WT rabbit 3224 ± 48.64 g (P < 0.001). Fourteen litters of F1 generation rabbits were produced, and the mutation of MSTN could be stably inherited in the MSTN KO rabbits. Heritable double muscle buttocks rabbits via myostatin mutation with CRISPR/Cas9 system were generated, which will be valuable in meat rabbits breeding and a useful animal model for the study of human muscle development related diseases.

for Chemistry. "It has not only revolutionized basic science, but also resulted in innovative crops and will lead to groundbreaking new medical treatments."Gustafsson said that, as a result, any genome can now be edited "to x genetic damage." And now, As the most e ciency genome edit tool, CRISPR/Cas9 system has been widely used to study the role of functional genes. Knock out of MSTN with CRISPR/Cas9 system has been processed in rabbits [15,16], sheep [17,18], goats [19,20], and pigs [21,22] in the last few years. Qian, L et al. knock out the MSTN gene of Meishan pig fetal broblasts by Engineered zinc-nger proteins, andprepared the MSTN-KO pig by somatic cell nuclear transfer technology rstly. Compared with wild-type Meishan pigs, the lean meat rate of Meishan pigs with MSTN knockout increased by 11.62% [23]. In addition, the MSTN-KO goats obtained by the Transcription activator-like effectors successfully, and MSTN-KO goats meat production is 32% higher than wild-type goats.
E cient TALEN-mediated myostatin gene editing in goats. BMC Dev Biol Zhang et al. Inserted the fat-1 gene to the goat MSTN gene by CRISPR/Cas9. While overexpressing fat-1, the MSTN gene was knocked out. The results of PCR and gas chromatography analysis showed that the knockout e ciency of MSTN was as high as 25.56%, and a signi cant increase in n-3PUFA level was detected in goat tissues. Zhang et al. [24] used CRISPR/Cas9 to insert the fat-1 gene into the goat MSTN gene, While overexpressing fat-1, the MSTN gene was knocked out, so this strategy edited both genes simultaneously. PCR and gas chromatography analysis results showed that the MSTN knockout e ciency was as high as 25.56%, and a signi cant increase in the expression level of n-3PUFA was detected in goat tissues. Moreover, further studies found that loss activity of MSTN is also relate to the obesity, muscle atrophy and type 2 diabetes [25,26].
It has a long tradition of rabbit meat production and consumption in France, Italy and Spain, and according to the database of FAOSTAT [27] the world per capita rabbit meat consumption in 0.242 kg, while it is very high in many European countries, expecially 4.39 kg per capita in Italy. It presents the stable growth of rabbit meat consumption state from 2000 to 2013 in many European countries [28]. rabbit meat offers excellent nutritive and dietetic properties [29] and contais high-quality protein, low fat, low cholesterol, which is ideal meat of the obese and cardiovascular patients. In the present study, we designed a highly e cient new site, which is different from previous studies, and further veri ed that the exon 3 region can regulate the MSTN gene. At the same time, the double-muscle breech rabbits obtained in this experiment can be normally inherited to the offspring, and have stable double-muscle breech characterization.

Animals
In present study, all of the experiments on rabbit were performed by the Principle Guidance for the Use and Care of Laboratory Animals. Rabbits were fed regularly at Laboratory Animal Center of Guangxi University. The rabbits used for ovulation were approximately 6-8 months old and weighed between 3.5 kg to 4.5 kg.

SgRNA Design and Plasmid Construction
A total of 8 sgRNAs targeting the MSTN sites were designed using the website (http://www.genomeengineering.org/) and named g1, g2, g3, g4, g5, g6, g7 and g8 (Fig. 1A). The two complementary DNA oligos were annealed to become double-stranded, and cloned into the pMD18-hU6-gRNA vector. Primer T76 and T78 were used as target gene fragments ampli ed by the two sgRNA targeting vectors. Linearized DNA plasmids and PCR products were extracted and puri ed using the MEGA shortscrip TM T7 kit (Ambion, USA) and sgRNAs were generated according to the manufacturer's recommendations.

Culture and electroporation of broblasts
Primary rabbit broblasts were cultured in 10% (v/v) FBS, 35 mm glutamine and 1 × DMEM (Gibco). The plasmid was transfected into broblasts by electroporation at a pulse of 225 volts for 10 milliseconds. Brie y, 4 µg of pCMV-T7-NLS-hSpCas9-NLS (hSpCas9) DNA plasmid and 2 µg of gRNA plasmid were used to transfect 5-10 × 10 5 broblasts. the culture medium was changed to DMEM containing 10% FBS 24 hours later. After cultured for 4 days, the DNA was extracted and the MSTN was ampli ed by PCR. Sent to the company for sequencing to analyze the editing effect of the gene fragment. Optimal sgRNA was selected for the following experiment.

Zygotes collection and Blastocyst genotyping testing
Rabbits of estrus were mated two or three times 16-20 hours before mating, and zygotes were collected from the fallopian tubes with cell culture medium. The cell culture medium was consisted of M199 (Gibco, USA), 3% bovine serum (Gibco, USA), 5 mM Hepes (Sigma, USA) and 5 mM NaHCO 3 (Sigma, USA). The zygotes were microinjected with Cas9 protein and sgRNA and in vitro cultured to blastocyst, and the mutation was analyzed by PCR and T7EI digestion of the target region.

Generation of transgenic rabbit
The zygotes was injected with 10 pL CRISPR/Cas9 mixture, containing 200 ng/µL Cas9 protein (A36497, Hot Fisher, USA), 20 ng/µL sgRNA6 and 20 ng/µL sgRNA8, in the cytoplasm and the zygotes were transferred back to the oviduct of recipient rabbit. Pregnancy phenomenon was checked on 10 days after the transplantation and the cubs were born about 30 days later.

T7 Endonuclease I (T7EI) assay
Genome DNA of the cubs were extracted and the mutation induced by CRISPR/Cas9 was analyzed by PCR and T7E1 assay. PCR was performed using primers anking the target site and the product was sequenced (Shanghai Sangon Company, Shanghai, China) and digested with T7EI (NEB, USA) to characterize the mutation.

Off-target analysis
Potential off-target sites of the sgRNAs were evaluated using the CRISPR design tool (http://crispor.tefor.net/). The top 5 potential off-target sites of each sgRNA were selected and ampli ed by PCR with speci c primers (Table.1), and the PCR products were analyzed by sequencing and T7E1 digestion.

Western Blot analysis
The gluteus Maximus tissue samples of MSTN +/− and WT rabbits (euthanized at 6 months of age) was collected, tissues were ground in liquid nitrogen, and added 2.5 µL / mL protease inhibitor with ice for 30 minutes. Protein concentration was determined using the Bradford method (Bio-Rad). 35 µg of protein sample was subjected to a 5%~12% separation of SDS-polyacrylamide gel. Antibodies used in this study included anti-MSTN polyclonal antibody (Abcam) and goat anti-rabbit IgG coupled to horseradish peroxidase (HRP) (Santa Cruz, USA). The β-actin antibody (Santa Cruz) was used as an internal control.

Weight, and sample collection and Histological analysis
In this study, 5 F0 MSTN +/− and 5 WT rabbits were feed at the same conditions. All rabbits were weaned at 30 days and were kept in separate cages in the same litter. Body weight was recorded every 10 days up to 200 days. At 2 months of age, rabbits were anesthetized with anesthesia with pentobarbital sodium (1 mL/ kg). The corpses were dissected and the hearts, tongue, gluteus maximus and lateral femoral muscles of the MSTN KO and WT groups were weighed.
The tongue and gluteus Maximus tissues of F0 generation MSTN KO and WT rabbits (6 months old sacri ce) were xed with 4% paraformaldehyde at 4℃, The increasing concentrations of dehydrated sucrose (30% for 7 h, 40% for 7 h, 45% for 5 h), freezed section at -25℃ for histological examination. The tissues which were sectioned into 5 µm thick slices were stained with hematoxylin and eosin (H&E) and analyzed by Fluorescence Inverted Microscope (Nikon, Japan). The integral optical density (IOD) analysis of histological sections was performed by Image-Pro Plus 6.0 software.

Statistical analysis
All of the data we obtained in this study were analyzed using Graphpad prism software (T test) and one-way ANOVA followed by Duncan's Multiple Comparison Tests in the SPSS 17 statistical software package.

Design and construction of sgRNA
Eight sgRNAs targeting to the MSTN exon 1 and exon 3 were designed and cloned into the pMD18T vector, respectively (Fig. 1A). The recombinant vector was named as pMD18-hU6-gRNA and con rmed that the sgRNA has been successfully inserted into the vector by sequence analysis (Fig. 1C).

Modi cation e ciency of the sgRNAs in rabbit broblasts
The mutation e ciency of these designed sgRNAs was com rmed in rabbit broblasts, and no mutation were generated in g1 to g4, the mutation rates of g5-g8 from 20-45% (Fig. 2). The sgRNAs target to exon 3 is signi cantly better than that of exon 1. The gene modi cation rate of sgRNA6 (45%, 9/20) and sgRNA8 (40%, 8/20) were higher than that of the other gRNAs. Thus, sgRNA6 and sgRNA8 were selected in the following experiments.

CRISPR/Cas9-mediated MSTN site-speci c KNOCK-OUT in rabbit zygotes
The CRISPR/Cas9 system containing sgRNA6 and sgRNA8 were injected into the zygotes to verify the sitespeci c deletion of MSTN gene. Atotalof16 zygotes was injected and 10 of them developed to blastocyst stage in vitro (Fig. 3A). The genome editing e ciency in the blastocysts were analyzed by PCR and Sanger sequence, and results revealed that 7 of them showed geneticmutations (70 %, 7/10) ( g. 3B).

Generation of MSTN knock-out rabbits
A total of 99 cytoplasmic injected zygotes were transplanted to oviduct of 10 receptor female rabbits, and 23 pup rabbits were born after full-term gestation ( Table 2). The presence of genome editing was analyzed by T7E1 digestion and Sanger sequencing analysis, and the results showed that 5 of them are MSTN knock-out rabbits (Fig. 4). Intrestingly, three MSTN +/− rabbits were all males, and two MSTN −/− rabbits were all females. The further T7E1 and sequencing analysis showed that 5 MSTN knock-out rabbit are all chimeras.

Western Blot analysis of MSTN KO rabbits
The expressionof MSTNin muscle tissue of MSTN KO rabbit was signi cantly lower than that of WT rabbit based on QRT-PCR analysis (p<0.05). And the further western blot analysis results showed that the expression level of MSTN protein of the mutant rabbits was lower in the muscle than wild rabbit (Fig. 6).
To determine whether the MSTN KO could be stably transmitted to the offspring, the two female founders MSTN -/were mated with male rabbit, but now offspring pub rabbits were produced. The threeMSTN +/-F0 male rabbits were mated with wild rabbits, and live F1 offspring pub rabbits generated from all those three rabbits. T-cloning sequence analysis and T7E1 cleavage assay demonstrated that 28 out of 53 newborn F1 rabbits carried MSTN mutations. In these pups, the 23 were monoallelic, while 5 were biallelic MSTN-KO rabbits (Fig. 8).

Discussion
In this experiment, the Cas9 / sgRNA mixture was injected into the single-cell stage embryos by cytoplasmic injection to delete long fragments and ensure the inactivation of target gene function. The MSTN KO rabbit was successfully produced, and the results showed that the system can effectively produce knockout rabbits. The loss of the MSTN fragment could leads to muscle growth.
Although the increase in muscle mass makes MSTN-KO livestock production more attractive, However, in these MSTN mutant animals, calving di culties are often caused by various reasons. MSTN is an important regulator of muscle cell proliferation and differentiation during muscle development. MSTN signals can be detected from the early myogenic stage of embryonic sarcomere formation to the adult skeletal muscle development stage [7,30,31]. The expression of MSTN was observed for the rst time during the embryonic period [7,30,31]. The birth weight of MSTN mutant animals has been observed in sheep, cattle and goats [32,33].Therefore, the knockout of the MSTN gene may affect muscle development in the embryonic stage. The researchers believe that during fetal development, MSTN is only expressed in skeletal muscle, thereby controlling the differentiation and proliferation of myoblasts. In our study, MSTN-KO rabbits exhibited a typical double-muscle phenotype, but there was no signi cant difference in body weight at rst 60 days and Compared with the wild-type control group, there is almost no difference in body size at birth (P > 0.05). It's might beacause IGF-1 is an important positive regulator of muscle cell proliferation and differentiation in skeletal muscle [34]. Despite the inhibition of MSTN signal expression, IGF signaling will still upregulate the expression of myostatin in skeletal muscle tissue models, which also indicates that there is an inhibitory autoregulatory loop in the muscular system [35,36].
Studies believe that in most terrestrial vertebrates, the increase in mammalian muscle production is mainly caused by muscle ber hypertrophy, hyperplasia, or a combination thereof caused by the lack of MSTN [21,[37][38][39]. The number of muscle bers mainly determined before birth, and the diameter of muscle bers may expand after birth [40,41]. The muscle mechanisms of hypertrophy and hyperplasia may not occur simultaneously [42,43]. But it seems that the mechanism by which MSTN mutations enhance muscle mass is different between species. In MSTN mutant mice, goats and rabbits, the increase in muscle mass is caused by muscle ber hyperplasia and diameter hypertrophy [37,44,45]. However, only muscle ber hyperplasia was observed in MSTN mutant cattle and pigs [8,46]. The muscle mass of MSTN-de cient mice is 2-3 times than of wild-type mice. The increase in muscle mass is the result of the combined action of muscle ber hyperplasia and hyperplasia [37]. In contrast, double-muscle cattle show muscle mass increased by 20% -25%, and the increased muscle weight seems to be the result of muscle ber diameter proliferation, rather than an increase in muscle number [8,39]. The phenotypic trait of muscle hypertrophy is a quantitative trait. The quantitative trait is the result of the joint action of many independent small-effect multi-genes. The effect of each gene is equal, but in some cases there is a major effect gene Gene is the main effect in this trait, and the other is micro-effect. The muscle growth inhibition is controlled by MSTN gene, and other genes are controlled by the micro-effect gene.
In addition, the MSTN-KO rabbit exhibits the muscular feature of "double mushy buttocks", and the fetus of the rabbit presents with "big tongue syndrome" [47]. In some live rabbits born, some of the MSTN-KO rabbits also have "big tongue syndrome", which causes milk sucking. Success leads to death. Most of the MSTN-KO rabbits show a rapid increase in muscle mass after 60 days of age, and MSTN-KO rabbits are healthy and can inherit the knocked-out gene fragments, which shows that the MSTN-KO rabbits produced by the CRISPR / Cas9 system are suitable for studying muscle development and Related diseases.

Conclusion
In summary, we obtained MSTN-KO rabbits with double muscle typically. These MSTN-KO rabbits could be a promising tool for studying the development of livestock muscles and improving their important economic traits. Thereby improving the meat production e ciency of rabbits and promoting the development of rabbit industry. We established a CRISPR/Cas9 gene editing system suitable for the laboratory, which is more effectively applied to various model animals and large livestock animals.   The results of sequencing for mutations at the speci c sites in rabbit REFs. (A) Rabbit Primary broblast cell in culture from 2days to 5days. (B) Sanger sequence analysis of REFs. g1-g4 target to exon1, g5-g8 target to exon 3. gRNA edited were named on the left. g1-g8 represent different gRNA, PAM sequence is marked in green. The sgRNAs sequence is marked in red. The numbers on the right indicated the type of mutation, with "-" represents deletion of the given number of nucleotides, "+" represents insertion of the given number of nucleotides.      sequence is marked in green. The sgRNAs sequence is marked in red. The numbers on the right indicated the type of mutation, with "-" represents deletion of the given number of nucleotides, "+" represents insertion ofthe given number of nucleotides.