Porcine antiviral activity is increased by CRISPRa-SAM system

Clustered Regularly Interspaced Short Palindromic Repeat activation-synergistic activation mediator system (CRISPRa-SAM) has been efficiently used to up-regulate the targeted genes in human and mouse. But it is not known whether the CRISPRa-SAM system can be used against porcine disease because its two important transcriptional activation domains (P65 and heat shock transcription factor 1 (HSF1)) are from mouse and human, respectively. Pig is one of the most important meat sources, porcine viral infectious diseases cause massive economic losses to the swine industry and threaten the public health. We aimed to investigate whether the CRISPRa-SAM system could increase porcine antiviral activity by mediating two pig-specific target genes (Mx2 and β1,4 N-acetylgalactosaminyltransferase (B4galnt2)). First, we constructed PK-15 and IPEC-J2 cell lines expressing nuclease-deficient Cas9 (dCas9)-vp64 and MS2-P65-HSF1 stably. Next, in these two cell models, we activated Mx2 and B4galnt2 expression through CRISPRa-SAM system. Antiviral activity to PRV or H9N2 was improved in PK-15 cells where Mx2 or B4galnt2 was activated. Altogether, our results demonstrated the potential of CRISPRa-SAM system as a powerful tool for activating pig genes and improving porcine antiviral activity.


Introduction
Pig is one of the most important meat producing animals. Porcine viral infectious diseases cause massive economic losses to the swine industry. Enhancing the antiviral ability of pigs has been a keen interest for researchers. Zhao et al. (2107) reported to reduce the fat deposition, and increase lean meat production in porcine, through reconstitution of UCP1 using CRISPR/Cas9 [1]. Burkard et al. (2017) used CRISPR/Cas9 to produce pigs completely resistant to porcine reproductive and respiratory syndrome virus (PRRSV) by removing the CD163 subdomain 5 [2].
Nonetheless the Clustered Regularly Interspaced Short Palindromic Repeat activation-synergistic activation mediator system (CRISPRa-SAM) system needs to be explored for pig antiviral breeding. This system is an extension of the conventional CRISPR knockout system. Instead of wild-type Cas9, designed variants of a nuclease-deficient Cas9 (dCas9) carrying the D10A and H840A mutations are used in the CRISPRa-SAM system. The dCas9 is fused to herpes simplex virus (HSV) viral protein 16 (VP16) activation domains (known as VP64) and the system includes some synergistic activation mediators. These mediators comprise fusion proteins (MS2-guide RNA (gRNA), dCas9-VP64 and the MS2-P65-HSF1), MS2 bacteriophage coat protein fused to the NF-κB (P65) and heat shock transcription factor 1 (HSF1)  [3][4][5]. It is widely used to up-regulate genes in human cells [6][7][8], rats [9,10] and fruit flies [11]. The CRISPRa-SAM system's activation domains P65 and HSF1 are from mouse and human, so it is required to reveal the role of CRISPRa-SAM as a gene regulator in pig cells for porcine antiviral activity. Myxovirus resistance proteins (Mx) are a family of dynamin-like GFPases, and can inhibit influenza A virus. In mammals, Mx gene has two isoforms, Mx1 and Mx2. Mx1 can cause hindrance in replication of RNA and DNA viruses [12]. Mx2 has a tendency to affect the anti-human immunodeficiency virus type 1 (HIV-1) activity of type I IFN [13,14]. Meanwhile, Mx2 can inhibit PRRSV infection in pig [15] and herpes virus infection [16]. Glycosylation is one of the most important post-translational modifications of proteins in eukaryotic cells. In the Golgi, terminal glycosylation reactions generate a huge panel of glycan that confer a variety of structural and functional roles to the glycoproteins exposed at the cell surface. β1,4 N-acetylgalactosaminyltransferase (B4GALNT2) catalyzes the last step in the biosynthesis of the human Sd(a) antigen through the addition of an N-acetylgalactosamine residue via a β-1,4 linkage to a subterminal galactose residue substituted with an α-2,3-linked sialic acid [17,18]. Recently, a study suggested that B4GALNT2 overexpression can prevent the infection of every avian influenza virus strain [19]. So we chose these two pig genes for the present study.
In the present study, we evaluated the ability of the CRISPRa-SAM activation systems to activate Mx2 and B4galnt2 gene expression in two kinds of porcine cell lines (PK-15, IPEC-J2). We observed PK-15 cells had more antiviral activity to PRV or H9N2 when Mx2 or B4galnt2 expression was activated. The present study highlights that biotechnology has great potential to manipulate pig breeding for improving porcine antiviral activity.

Plasmids construction
The location of the transcription start site (TSS) of sgRNA is very important to cause activation of genes. The design of the sgRNA was based on the principle around the upstream of the target gene TSS followed the GN (19) NGG. Four sgRNAs of each gene were designed and synthesized, named as sgRNA1, sgRNA2, sgRNA3 and sgRNA4, respectively ( Figure 2). The sgRNAs used in the present study were designed by http://crispr.mit.edu/, and are listed in Table 1.

Cell culture
PK-15 cells (pig kidney cell line), IPEC-J2 cells (pig intestinal epithelial cell line) and HEK293T cells were maintained in DMEM supplemented with 10% fetal bovine serum (FBS) and were cultured in an incubator at 37 • C with 5% CO 2 .

Virus infection
PK15-CRISPRa-SAM cells and PK-15 cells with activated (600-fold) Mx2 gene were infected with PRV BAC (pBecker2) at an MOI (100 or 10) for 24 h. pBecker2 was gifted by Professor Hanzhong Wang from Wuhan institute of Virology, Chinese Academy of Sciences. Fluorescence microscopy (Carl Zeiss, LSM 800) was used to observe under a 20× objective.

Statistical analysis
All the experiments were run three times, and results were documented as means + − SEM. The qPCR results were analyzed by two-way ANOVA in GraphPad Prism 6.01, while other results were analyzed by Student's t test in GraphPad Prism 6.01. Statistical significance was defined by a P-value of less than 0.05.

Results
The transcriptional activation domains of pig P65 and HSF1 genes were highly homologous with that of P65, HSF1 of the CRISPRa-SAM system First, we checked whether the transcriptional activation domains of pig (P65, HSF1) are homologous to the activation domains (P65, HSF1) of the CRISPRa-SAM system. Results computed by NCBI blast showed pig P65 (FN999988.1) and HSF1 (XM 005655310.3) were highly homologous to that of CRISPRa-SAM system. The identity and gap for HSF1 in pig were 89.37% (311/348) and 6.45% (24/372), respectively (Figure 1), while those of P65 were 71.03% (385/542) and 2.34% (13/555), respectively (Figure 1). It is worth mentioning that the efficacy of CRISPRa-SAM system has not been validated in pigs before.

The important components of CRISPRa-SAM system efficiently expressed in two pig cell lines
We checked the expression till day 14 post selection, for important components of CRISPRa-SAM system by infecting PK-15 and IPEC-J2 cells with lentiviruses expressing dCas9-VP64 and MS2-P65-HSF1, respectively. The result showed that dCas9-VP64 and MS2-P65-HSF1 were stably and effectively expressed in two pig cell lines.

CRISPRa-SAM system could effectively activate transcription in different pig cell lines with different efficiencies
We designed the gRNAs of Mx2 and B4galnt2 through the website http://crispr.mit.edu/. These gRNAs are listed in Table 1 (Figures 3 and 4). Up-regulation of activated Mx2  was 600-and 10-fold in pig cells PK-15 and IPEC-J2, respectively, while activated B4galnt2 gene was up-regulated by 10-fold in PK-15 cell line and 7-fold in IPEC-J2 cell line (Figures 3 and 4).

Antiviral activity was increased in PK-15 cells
Mx2 gene has broad-spectrum antiviral effect, hence the overexpression of Mx2 could inhibit DNA and RNA viruses. To check this hypothesis, up-regulation of pig genes mediated by the CRISPRa-SAM system in pig cells could improve their normal functions, we tested the Mx2 expression level with relation to its antiviral activity against Pseudorabies virus (PRV) (MOI = 100 or 10) by infecting PK-15 cells activated Mx2 approximately 600-fold higher by CRISPRa-SAM system and PK-15 cells. Results shown in Figure 5A illustrate that PRV replication was inhibited in PK15 cells after overactivation of Mx2.
For further investigations, antiviral activity of B4galnt2 against pan-avian influenza virus infection [19] was checked for any increase by infecting the PK-15 cells activated B4galnt2 approximately ten-fold higher by the CRISPRa-SAM system and PK-15 cells with H9N2 virus. The H9N2 virus is widely circulating in the world, causing rasorial infections and serving as a gene donor for H5N1, H7N9 and H10N8 viruses [21]. The results showed that the virus titer of the H9N2 virus produced by PK-15 cells activated B4galnt2 was lower than the control (without gene activation) ( Figure 5B). It is concluded that CRISPRa-SAM-mediated gene activation efficiently enhanced host antiviral response by increasing the expression of antiviral genes.
These results suggest that the CRISPRa-SAM system is suitable for pig gene induction and can be used in pig genome study. Nonetheless, the HSF1 and P65 used as transcriptional activation domains in CRISPR-SAM system are based on human and murine genome sources, respectively.

Discussion
There are many advantages of the CRISPRa-SAM system over traditional overexpression gene technology. Exogenous expression vectors are used to clone cDNA sequences for traditional gene overexpression that is hard to achieve when the target gene has a long sequence or is rich in GC contents.
The CRISPRa-SAM system is capable of activating the target gene with only an sgRNA regardless the size of target gene. The CRISPRa system has been shown to activate target genes in many species [6,9,22]. Gain-of-function screening using a pool of sgRNA libraries, has been a powerful and effective tool for target gene screening [3,5,19,[23][24][25] while cDNA library overexpression still retains some problems [3,5,19,[23][24][25].
In the present study, we investigated the ability of the CRISPRa-SAM system to activate endogenous pig gene expression for porcine antiviral activity. Our results suggested that the CRISPRa-SAM system can effectively activate the pig endogenous gene transcriptional expression in both pig PK-15 and IPEC-J2 cell lines with increased porcine antiviral activity. We concluded that the CRISPRa-SAM system could effectively activate transcription in different pig cell lines with different efficiencies. The effect of CRISPRa-SAM was higher in PK-15 cells with 600 fold up-regulation of sgRNA1 for Mx2, compared with ten-fold in the IPEC-J2 cell line (Figures 3 and 4), while B4galnt2 gene up-regulation was ten-fold in PK-15 cells vs seven-fold in IPEC-J2 cells. In a pool of four sgRNAs, the effect of sgRNA1 of Mx2 gene had significantly higher efficiency than three others in PK-15 cells and IPEC-J2 cells and the effect of sgRNA2 and sgRNA1 of B4galnt2 were significantly higher than others in both the cells. The porcine antiviral activity to PRV or H9N2 was improved in PK-15 cells where Mx2 or B4galnt2 was activated.
Similarly, in human cell lines, the efficiency of CRISPRa system to activate different target genes varied with different sgRNA. Konermann et al. (2015) [3] reported fold activation of 12 different genes (VEGFA, HBG1, TERT,  IL-1B, IL-1R2, ZFP42, MYC, LIN28A, SOX2, NANOG, KLF4, POU5F1) mediated by the CRISPRa-SAM system and plotted against the sgRNA location in 293FT cells. In another study, a panel of validated sgRNAs was used to target the promoters of both coding (TTN, RHOXF2, ASCL1, HBG1) and non-coding (MIAT, TUNA) genes in HEK293 cell, MOLM14 cell, K562 cell and HIL60 cell, and found sgRNAs gave the highest activation in MOLM14 cell [24]. Conclusively, the efficiency of CRISPRa-SAM system is highly dependent on the cell type, target gene basal expression level and the location of gRNA. Porcine Mx2 and B4galnt2 were activated by CRISPRa-SAM system and porcine antiviral capacity was increased because target genes were activated in pig cells. CRISPRa-SAM system is a powerful tool to activate the expression of pig endogenous genes. The development of these pig cell models and results are not only valuable for the genome-wide screening of antiviral pig genes but also are the foundation for pig antiviral breeding with the CRISPRa-SAM system.