Generation of ETV5 Knockout Pigs with CRISPR/Cas9

Spermatogenesis relies on SSCs self-renewal to maintain an appropriate number of SSCs and being able to differentiate into mature sperm. It requires more than 1 month for SSCs to undergo meiosis and then progress to sperm formation (Zheng et al., 2014; Sato et al., 2011). The process of differentiation of SSCs in mammalian testes for the generation of sperm continues throughout a male’s entire lifespan (Sharma et al., 2015). However, SSCs are the only stem cells in male animals that can transmit genetic information to the next generation (Sato et al., 2011).
       
The proliferation and differentiation of SSCs are the result of germ cell gene expression, which are principally regulated by exogenous factors present in the body’s microenvironment. ETV5 plays an important regulatory role in SSCs self-renewal and maintenance. Loss of ETV5 in mice causes abnormalities in testicular and physical development, where mice exhibit progressive loss of SSCs until eventually depleted (Chen et al., 2005). ETV5-KO male mice start a quantitative diminution in the number of SSCs during the first wave of spermatogenesis, resulting in a Sertoli cell-only phenotype (Tyagi et al., 2009). ETV5-KO female mice exhibited decreased ovulation and no interest in mating, indicating that ETV5 is important for germ cell development in female mice (Eo et al., 2011). Collectively, the data indicate that ETV5-null mice possess severe defects in fertility due to reduced spermatozoa and oocytes in male and female mice, respectively. However, there are no published studies on ETV5 in livestock.
       
In the present study, the CRISPR/Cas 9 system and SCNT were used to produce homozygous ETV5 knockout (KO) Duroc pigs. If the phenotype of ETV5-KO pigs is consistent with that of mice, these mutant pigs can be used as models to study the apoptosis mechanism of SSCs and further study the function of ETV5 gene in pigs.

The construction of gene editing plasmid and cell screening were carried out in the National Engineering Research Center for Breeding Swine Industry of South China Agricultural University and the experiments of somatic cell nuclear transfer and gene editing pigs were carried out in the Pig Breeding Laboratory of Wens Foodstuff Group. The experimental period of this study is from June 2017 to June 2019.
 
Construction of gene editing plasmid
 
Guide RNA (gRNA) was designed to target the regions within exon 3 of porcine ETV5 gene. A target immediately preceding 5¢-NGG protospacer adjacent motif (PAM) was selected: GAAGTTTTTGGACACTGATCTGG (PAM site is underlined). A Cas9 protein-expressing plasmid pX330 was digested with restriction enzyme BbsI and the linearized plasmid was ligated with gRNA to generate the genome-editing plasmid pX330-ETV5.
 
Genome editing of porcine fetal fibroblasts
 
Male Duroc fetal fibroblasts were transfected with pX330-ETV5 plasmid by electroporation (BTX ECM830), diluted to single cells and cultured until single colonies were established. The single colonies were then harvested and transferred to 48-well culture plates. When the cells fully covered the well, they were harvested and lysed to be used as templates for PCR amplification. PCR was performed using the followed primers ETVF (5'-CTCAATGCTGAGAC CTTCCAA-3') and ETVR (5'-TTGCCTTCAGCTAACCAAGC-3') with Tm at 58°C. The PCR product was 460 bp in length and the presence of targeted mutations was detected by DNA sequencing.
 
Somatic cell nuclear transfer (SCNT)
 
The gene editing cell was then used as the donor for SCNT. Donor cells were microinjected into denucleated porcine oocytes-with no fewer than 200 transferred to estrous sows-and cloned pigs were born after 114 days. Live piglets were weighed and numbered within 24 hours of birth and their ear tissue DNA was obtained using an animal tissue genome DNA extraction kit (Omega, USA). The targeted gene modification status was examined on every pig by PCR and sequencing and sperm DNA was tested again for gene modification when the pig was sexually mature.
 
HE staining, immunohistochemistry and western blot
 
To examine for the presence of pathology in major organs, the testis, brain, lung and heart tissues were fixed in 4% paraformaldehyde, paraffin embedded, sectioned, stained with hematoxylin and eosin and analyzed. Testis HE-stained sections were evaluated to ascertain structural normality of seminiferous tubules. The distribution of SSCs was assessed by immunohistochemical staining (primary antibody, rabbit anti-UCHL-1, 1:500, purchased from Proteintech, USA; HRP-labeled secondary antibody; DAB for color development). Total testicular protein was subjected to western blotting analysis (primary antibody, rabbit anti-ETV5, 1:3000, purchased from Proteintech, USA; secondary antibody, goat anti-rabbit, 1:10000).
 
Testosterone concentrations and spermatogenesis in ETV5-KO pigs
 
To test the hormonal levels of pigs, blood samples from ETV5 knock-out pigs (n=1) and wild-type pigs (n=4) were obtained at 8 months of age. The blood samples were collected every 15 minutes 3 times, allowed to sit at room temperature for 30 minutes and centrifuged for 5 minutes at 3000 rpm. The supernatants were stored at -80°C and the testosterone concentrations were examined using a porcine testosterone (T) Elisa kit (Nanjing jiancheng, China). After general anesthesia, the boars were castrated to obtain testes and epididymides. The spermatozoa were extracted from the epididymis, diluted 1:100 with sperm dilution solution and observed under a microscope (E800, Nikon, Japan). The statistical calculations were made using student’s t-test (SPSS 17.0). All the data were expressed as mean ± standard deviation and P<0.05 was considered to be significance.
 
RT-PCR of reproduction-related genes in ETV5-KO pigs
 
To examine the expression of reproduction-related genes in ETV5-KO pigs, total RNA was obtained from testicular tissue at 8 months with an animal tissue RNA extraction kit (Omega, USA). The expressions of Dazl, Thy-1, Gfra-1, Plzf, C-kit, Stra8 and Prm2 genes were examined using RT-PCR (Table 1).
 

Generation of ETV5-KO pigs
 
The target site was the third exon of the pig ETV5 gene (Fig 1A) and the pX330-ETV5 plasmid was electroporated into Duroc fetal fibroblasts, 461 single-cell formed colonies were selected and 68 homozygous knockout cell clones were obtained (Table 2).
 

 


Finally, seven ETV5-modified cell clones (10, 72, 109, 254, 272, 275 and 294) were used as donor cells for SCNT (Table 3). The reconstructed embryos were surgically implanted into 10 surrogate mothers-6 of which became pregnant-and 3 pregnant surrogates delivered 7 piglets by natural birth (Table 4), numbered #925801, #925803, #925805, #925807, #925901, #926011 and #926012. Cloned pigs showed normal activity and breastfeeding (Fig 1B). PCR amplification and sequencing analysis (Fig S1) confirmed the homozygous EVT5-null mutation in all 7 pigs, 5 of which contained indels of D1/D1 and 2 were +1/+1(Table 5). Pigs #925803, #925807 and #926011 died at 1 week of age, pig #925901 died at 2 weeks of age and pig #926012 died at 1 month. Pig #925801 had leg injuries that resulted in the inability to walk at 2 months of age. Because of the inability to develop normally to adulthood, euthanasia was performed on pig #925801 and its testicular tissues were collected for SSCs analysis. Pig #925805 was raised to sexual maturity at 8 months of age and its sperm genotype showed results identical to its ear tissue detected at birth. There was also no ETV5 protein in the testicular tissue of the ETV5-KO pigs, indicating that the ETV5 protein had been successfully knocked-out (Fig 1C).

Pigs are the primary livestock species worldwide, not only provide the major source of meat for humans (Harshini et al., 2018), but also a useful model for the study of mammalian physiology (Yang et al., 2019). The use of genome-editing tools in domestic pigs is expected to generate valuable large animal models to further elucidate the roles of allele-specific variations in growth, development and disease (Park et al., 2017). In the present study, we first obtained allele-modified fetal fibroblast cells using the CRISPR/Cas9 system and then prepared animals genetically modified at the ETV5 gene using SCNT. All 7 cloned pigs obtained had bi-allelic modification of ETV5. In our case, the offspring resulting from injected eggs did not require mating to generate homozygosity. This is an efficient, rapid and low-cost method for preparing genetically edited large animals with identical genetic modification (Zheng et al., 2017; Zhou et al., 2015; Mehta et al., 2017).

Phenotypic assessment of ETV5-KO pigs during the piglet stage

ETV5-KO pigs possessed normal tissue structure with no lesions as observed with HE staining of brain, heart and lung tissues (Fig 2A), indicating that the ETV5 KO was not lethal. There were no significant changes in the seminiferous tubules of the 1-week-old ETV5-KO pig (#925803), the interior of the seminiferous tubules was lined with seminiferous epithelium, Sertoli cells resided on the basement membrane of the seminiferous tubules and tubules were structurally intact. SSCs were present in the middle and at the base of the seminiferous tubules (Fig 2B). As for the 2-month-old pig (#925801), HE staining of testicular tissue showed that SSCs began to migrate to the basement membrane of the seminiferous tubules and that there was no loss of SSCs (Fig 2C).

       
In the current study, we discovered that ETV5 homozygous null pigs manifested no pathologies in brain, heart, or lung tissues, indicating that ETV5 deletion did not cause increased risk of lethality in pigs. In addition, these ETV5-KO piglets were viable and appeared normal during breastfeeding and other activities. Unfortunately, 2 of the 7 piglets were crushed to death by the sow and 3 of them died due to weakness caused by diarrhea, resulting in a lower piglet survival rate in ETV5-KO pigs. A similar testicular structure was identified in 1-week-old and 2-month-old ETV5-KO pigs compared with normal WT pigs. By immunostaining with the porcine SSC marker protein UCHL-1 (Luo et al., 2006), we showed that SSCs were distributed throughout the seminiferous tubules and were not depleted.
 
Phenotypes of ETV5-KO pigs at sexual maturity
 
ETV5-KO pig (#925805) reached sexual maturity at 8 months of age (Fig 3A) and ELISA was performed to detect blood testosterone levels at different time points. Results showed that there was no significant difference in serum testosterone between ETV5-KO pig and wild-type (WT) pigs, indicating that testicular hormone secretion was normal in the ETV5 KOs (Fig 3B). ETV5-KO pig was then anesthetized and the epididymides were removed and semen samples were collected with a pipette. Samples were analyzed under a light microscope and analysis showed that ETV5-KO pigs contained viable sperm, with numbers similar to WT pigs (Fig 3D). The testis of the ETV5-KO pig was removed by surgery and its size was slightly smaller than WT pig (Fig 3C). HE and UCHL-1 immunostaining were performed with the testicular tissue, the results showed that the SSCs were all located at the basement membrane of the seminiferous tubule and a large number of sperm were observed. The structural integrity was complete and SSCs were not depleted (Fig 3D). RT-PCR was performed on total cellular RNA from testes of 8-month-old ETV5-KO pig (Fig 3E). The results showed ETV5-KO pig expressed the Dazl, Thy-1, Gfra-1, Plzf and C-kit mRNA, which are known marker genes for proliferation and differentiation of SSCs. Furthermore, eight-month-old ETV5-KO pig expressed the Stra8 and Prm2, which are sperm markers and indicators of spermatogenic initiation, suggesting normal spermato genesis in 8-month-old ETV5-KO pig.
 

The results of ETV5-KO mice showed that SSCs began to disappear 4-8 days after birth and mice reached sexual maturity by 6 weeks of age followed by the first wave of spermatogenesis and from this time to 8-10 weeks of age, SSCs became depleted (Chen et al., 2005; Jamsai et al., 2013).  These results indicated that ETV5 affects mouse reproductive function in the early postnatal period (Schlesser et al., 2008). The process of SSCs depletion takes approximately one spermatogenic cycle of 5 weeks (35 days) in mice. Assuming that the pattern of SSCs depletion in ETV5-KO pigs is similar to that for ETV5-KO mice, SSCs depletion should begin when ETV5-KO pig reach sexual maturity at 6 months of age. However, our study did not show signs of SSCs depletion in 8-month-old ETV5-KO pig, suggesting that the impact on SSCs of loss of ETV5 in pigs may not parallel that in mice. Large numbers of sperm were also found in the epididymides, showing normal spermatogenesis in the ETV5-KO pig. Although the current results showed that ETV5-KO pigs did not exhibit a depletion of SSCs at 8 months of age, they do provide valuable insights into the exploration of reproductive roles for ETV5 and its application to pig reproduction. The species-specific mechanism of ETV5 on SSCs development still needs further investigation in pig.

This study presents the ETV5 knockout pigs prepared by CRISPR/Cas9 and SCNT were produced for the first time. ETV5 knockout pig did not show apoptosis in SSCs at 8 months old, which may provide valuable information on ETV5 gene function in pig reproduction and the application of SSCs transplantation.

This work was supported by grants from the National Natural Science Foundation of China (31772555) and the National Science and Technology Major Project for Breeding of New Transgenic Organisms (2016ZX08006002).