Background

Since the initial discovery of a highly conserved 29 base pairs (bp) sequence tandemly repeated with a spacing of 32 bp downstream of the iap gene in Escherichia coli genome (Ishino et al., 1987; Nakata et al., 1989), a family of short regularly spaced repeats, varying in size from 25 to 50 bp, were found in about 50% of bacteria and 90% of archaea (Makarova et al., 2015). According to their characteristic structures, the name clustered regularly interspaced short palindromic repeats (CRISPRs) were introduced by Mojica (Mojica et al., 2009) and Jansen (Jansen et al., 2002) and are currently in general use. A set of CRISPR associated genes, cas1 to cas4, was first identified flanking the CRISPR loci by nucleotide sequence alignments (Jansen et al., 2002). New members of cas genes were identified in different bacterial species. According to the cas members within each host, the CRISPR-Cas systems can be classified into three major types based on the hallmark genes (Makarova et al., 2011; Burmistrz and Pyrc, 2015). The function of the CRISPR-Cas system was demonstrated by a seminar experiment. After being challenged by virulent bacteriophages: phage 858 and phage 2,972, new repeat-spacer units were observed on the leading end of the CRISPR array in the surviving Streptococcus thermophlis host cells. The DNA sequences of newly acquired spacers were matched to corresponding fragments, named proto-spacers, in the phage genomes. Streptococcus thermophlis strains with phage spacer(s) were resistant to the phage infection, while strains without phage spacer(s) were sensitive to the phage infection (Barrangou et al., 2007). To distinguish between the spacer in the host bacterial genome and the proto-spacer, which has the same sequence as the spacer in the invader genome, a proto-spacer adjacent motif (PAM) was evolved (Mojica et al., 2009; Shah et al., 2013). The CRISPR-Cas immunity was revealed, it can be divided into three stages (Rath et al., 2015; Wright et al., 2016). The first adaptation or acquisition stage is responsible for the acquisition of spacers into CRISPR array following the exposure to foreign mobile genetic elements, such as phages or plasmids. A Cas2 homodimer was sandwiched by two Cas1 homodimers forming a heterohexameric complex for all three types of CRISPR-Cas systems (Sternberg et al., 2016). In the second stage, the promoter embedded within the AT-rich leader sequence upstream of the CRISPR array transcribed the precursor CRISPR RNAs (pre-crRNAs), these were further processed into short CRISPR RNA (crRNA) guides by Cas proteins. Cas6 was involved in the RNA processing step in both of the type I and type III CRISPR-Cas systems (Charpentier et al., 2015; Hochstrasser and Doudna, 2015). Accompanied by the Cas9 protein, a trans-activating crRNA (tracrRNA) which contains an anti-repeat segment for duplex formation with the repeat compartment of crRNA was involved in the maturation of the crRNAs in the type II system (Deltcheva et al., 2011). In the last interference stage, in cooperation with a mature crRNA and a cascade of Cas proteins, the signature proteins Cas3 and Cas10 were integrated into the RNA guided endonuclease complex in the type I and type III CRISPR-Cas systems, respectively. The type II effector is simply composed of a Cas9 protein, a pair of processed mature crRNA and tracrRNA (Gasiunas et al., 2012). The crRNA and tracrRNA in the ternary complex can be substituted by a fused crRNA-tracrRNA single-guide RNA (sgRNA) (Jinek et al., 2012). Because of extreme simplicity, the Cas9-sgRNA, now commonly termed as CRISPR/Cas9, binary complexes were immediately applied in the field of gene editing (Cong et al., 2013; Mali et al., 2013).

Swine share a number of anatomic and physiologic characteristics with humans. Systems that are mostly cited as suitable models include cardiovascular, urinary, integumentary, and digestive system (Swindle et al., 2012). Therefore, pigs are considered as a good source of organs for xenotransplantation. Two strategies are currently utilized to overcome the interspecies rejection hurdles. The first one is to block donor organs expressing the antigens causing hyper-acute rejection, such as galactose-α1,3-galactose, N-glycolylneuroaminic acid and β1,4-N-acetylgalactosamine, by targeting the GGTA1, CMAH and β4GalNT2 genes, respectively (Estrada et al., 2015; Cooper et al., 2016). The second strategy is to prepare organs which are composed of acceptor’s cells in a surrogate animal by a blastocyst complementation technique (Kobayashi et al., 2010; Usui et al., 2012; Matsunari et al., 2013). To produce gene-knockout pigs, the traditional method uses gene editing in somatic cells, such as fetal fibroblasts, and somatic cell nuclear transfer (SCNT) techniques to create knockout zygotes. The examples of combining CRISPR/Cas9 and SCNT are reported to generate IgM J_H (Chen et al., 2015), RUNX3 (Kang et al., 2016b), IL2RG (Kang et al., 2016a), and GGTA1/CMAH/β4GalNT2 triple knockout pigs (Estrada et al., 2015). Direct microinjection of DNA or RNA is another choice to prepare gene knockout pigs. Since the burst of de novo transcription in porcine zygotes was reported at the 4-cell stage (Anderson et al., 1999), RNA is preferred for micro-injection into embryos at the 1-cell stage. Previous studies have demonstrated that cytoplasmic microinjection of Cas9 mRNA with sgRNAs can produce Mitf (Wang et al., 2015) and DJ-1/Parkin/PINK1 triple-gene knockout pigs (Wang et al., 2016). Because de novo zygotic transcription had only been reported in the 1-cell stage mouse embryo (Ram and Schultz, 1993; Bouniol et al., 1995; Aoki et al., 1997), a trial was reported to produce GGTA1 knockout pigs by cytoplasmic microinjection of a CRISPR/Cas9 plasmid. With this strategy, CRISPR/Cas9 was expressed at, or later, than the 2-cell stage, and mosaic mutations on GGTA1 gene were found (Petersen et al., 2016). Pronucleus microinjection is needed to interpret whether zygotic transcription occurs at the 1-cell stage of pig embryo (Chuang et al., 2016).

A series of Cas9 and sgRNA expression vectors was constructed as shown in Figure 1. pCX-Flag₂-NLS₁-Cas9-NLS₂, pCX-HA-NLS₁-Cas9-NLS₂, and pCX-Myc-NLS₁-Cas9-NLS₂ can be used to express Cas9 in mammalian cells. Three common tags are available to monitor Cas9 expressions. (Figure 1A) Porcine U6 promoter which could effectively derive short hairpin RNA [Chuang et al., 2009] was used to construct the ppU6-(BsaI)₂-gRNA vector (Figure 1B) (Su et al., 2015). A pair of primers containing spacer and part of CRISPR repeat sequences, as shown in Figure 1B, is needed for each target site. Because guanine (G) is favored for U6 promoter as the first transcribed nucleotide, only proto-spacers initiated with G were chosen in our recent works.


Figure 1. Scheme of the Cas9 and single-guide RNA expression vectors. A. pCX-Flag₂-NLS₁-Cas9-NLS₂, pCX-HA-NLS₁-Cas9-NLS₂, and pCX-Myc-NLS₁-Cas9-NLS₂; B. ppU6-(BsaI)₂-gRNA and primer pair designation.