Testing multiplexed anti-ASFV CRISPR-Cas9 in reducing African swine fever virus

ABSTRACT African swine fever (ASF) is a highly fatal viral disease that poses a significant threat to domestic pigs and wild boars globally. In our study, we aimed to explore the potential of a multiplexed CRISPR-Cas system in suppressing ASFV replication and infection. By engineering CRISPR-Cas systems to target nine specific loci within the ASFV genome, we observed a substantial reduction in viral replication in vitro. This reduction was achieved through the concerted action of both Type II and Type III RNA polymerase-guided gRNA expression. To further evaluate its anti-viral function in vivo, we developed a pig strain expressing the multiplexable CRISPR-Cas-gRNA via germline genome editing. These transgenic pigs exhibited normal health with continuous expression of the CRISPR-Cas-gRNA system, and a subset displayed latent viral replication and delayed infection. However, the CRISPR-Cas9-engineered pigs did not exhibit a survival advantage upon exposure to ASFV. To our knowledge, this study represents the first instance of a living organism engineered via germline editing to assess resistance to ASFV infection using a CRISPR-Cas system. Our findings contribute valuable insights to guide the future design of enhanced viral immunity strategies. IMPORTANCE ASFV is currently a devastating disease with no effective vaccine or treatment available. Our study introduces a multiplexed CRISPR-Cas system targeting nine specific loci in the ASFV genome. This innovative approach successfully inhibits ASFV replication in vitro, and we have successfully engineered pig strains to express this anti-ASFV CRISPR-Cas system constitutively. Despite not observing survival advantages in these transgenic pigs upon ASFV challenges, we did note a delay in infection in some cases. To the best of our knowledge, this study constitutes the first example of a germline-edited animal with an anti-virus CRISPR-Cas system. These findings contribute to the advancement of future anti-viral strategies and the optimization of viral immunity technologies.

Recent studies have highlighted the immense potential of utilizing CRISPR-Cas9 systems to target viral genomes, such as those of HBV and HIV, to manage viral titers in vitro by disrupting the vital genes or inactivating genome in a non-cleavage mecha nism (7,8).Targeting the ASFV p30 gene has shown significant potential in effectively suppressing viral replication.P30 protein plays a crucial role in the virus's life cycle, as it is involved in host cell attachment and entry of ASFV (9).In our study, we explore the possibility of engineering a pig strain with acquired immunity against ASFV by incorporating an anti-ASFV CRISPR-Cas system into its genome.

Multiplex CRISPR strategy targeting the ASFV genome to protect pig cells from ASFV infection
ASFV is a double-stranded DNA virus with a genome size of 170,000 to 190,000 base pairs, encoding 150-200 proteins depending on the strain (10).We first analyzed all the published ASFV genomes and identified six guide RNAs (gRNAs), targeting nine places in the viral genome (Fig. 1a; Table S1).We hypothesized that by targeting multiple loci in the ASFV genome, we could minimize ASFV mutant escape.In addition, the cohort of targeting sites was chosen to cover the consensus sequences of all known ASFV strains and to avoid off-target cutting into the pig genome (Table S1).Next, we designed single transcription constructs to express these six gRNAs linked by ribozyme sites, driven by Type II and Type III RNA polymerase promoters, respectively (Fig. 1a and c, EF-1a and U6 promoters).It has been reported that the ASFV genome is exposed in both the cytoplasmic "virus factory" and nucleus, where it is under replication (11,12) (Fig. 1b).As such, we designed Cas9-expressing constructs with or without nuclear localization signal (NLS) to test which one can achieve optimal targeting efficiency.

Multiplex CRISPR strategy can efficiently cut ASFV DNA in vitro and different construct design has different effects on restricting ASFV replication in COS-7 cells
We first investigated the cutting efficacy of the CRISPR-Cas on the ASFV genome.To this end, we generated single gRNA transcripts via in vitro transcription (IVT) and observed robust cutting activities against ASFV genomic sequences (Fig. 2a; Fig. S2; Table S2).Later, we aimed to examine whether such a biochemical property of CRISPR-Cas can inhibit viral infection and replication.We chose COS-7 as the model system because it is permissive to ASFV infection and has been generally used as an in vitro model to quantify ASFV infectivity (13).We integrated the designed CRISPR-Cas constructs into the COS-7 genome via PiggyBac-transposition.Single-cell clones with CRISPR-Cas-gRNA integration were isolated, and corresponding Cas9 expressions were measured by RT-qPCR (Fig. 2b).Subsequently, we inoculated ASFV into the COS7 clones and monitored ASFV titers over 5 days.Interestingly, ASFV replication was significantly suppressed in the clones with high Cas9 expression (Fig. 2b, GC49, GE22, GE64, GF16, GF29 clones) but not in those with low Cas9 expression (FZ2, FY0, FY2, GC7 clones).This suggested that CRISPR-Cas inhibited ASFV replication in a dose-dependent manner (Fig. 2c).In addition, we found that the regulatory elements in the designed constructs were not determining factors for inhibiting ASFV replication.(i) Both Type II and Type III RNA polymerases were sufficient to drive the multiplexable gRNA expression, as evidenced by the significant inhibition of viral replication in Clone GE and GF.(ii) Cas9 with and without NLS could both significantly inhibit ASFV replication (Fig. 2, GE22, GF64 with NLS, GC49 without NLS).This observation is consistent with previous reports that ASFV utilizes both the host nucleus and cytoplasmic "virus factory" to replicate its DNA so that its genome is exposed in both sites (14,15).Taken together, the data demonstrate that the multiplex able CRISPR-Cas9 can cut ASFV by cutting its genome at multiple loci and effectively prevent ASFV replication in vitro.
Having validated that CRISPR-Cas9 could target ASFV and inhibit its replication in vitro, we attempted to incorporate the construct into a pig strain and test if the resulting pigs gained immunity against ASFV infection.We chose the NLS-Cas9-EF1a-gRNA construct (identical to the construct of the GC49 COS7 clone) as it demonstrated significant ASFV suppression in vitro (Fig. 2c).

Production of transgenic pigs constitutively expressing Cas9 and 6 sgRNA targeting ASFV genome and virus challenge of the transgenic pigs
We first integrated the construct via the PiggyBac transposon system into the fibroblasts of a Large White pig strain.After isolating the single-cell clones with the validated genetic modification, we performed pig cloning via somatic cell nuclear transfer technology (SCNT) and successfully obtained pigs with the intended modifications (Fig. 3b and Materials and Methods).We validated genomic integration and the expression of Cas9 using the ear pouch sample isolated from the engineered pigs (Fig. 3a, GI58 P10, GI58 P12).
Next, we aimed to determine whether the newly established pig strain, genetically modified with multiplexed anti-ASFV CRISPR-Cas, could develop immunity against ASFV.We conducted this experiment using two separate challenge methods, direct and indirect and closely tracked the viral titers over time.In the direct challenge group, we inoculated an identical amount of ASFV intramuscularly into three transgenic pigs (G1, G2, and G3) and three wild-type pigs (W4, W5, and W6).The two groups were kept in separate housing to prevent cross-contamination.To assess immunity against ASFV through horizontal transmission, we also set up sentinel pigs groups.In this arrange ment, one transgenic pig and three wild-type pigs, none of which had prior contact with  ASFV, were placed in each of the two housing units.In the first, transgenic pig G4 and wild-type pigs W1, W2, and W3 shared quarters with the directly challenged transgenic pigs G1, G2, and G3.In the second unit, transgenic pig G5 and wild-type pigs W7, W8, and W9 cohabitated with the directly challenged wild-type pigs W4, W5, and W6.Throughout the course of the study, we maintained rigorous monitoring of each pig's serum viral titers and survival time.
Consistent with previous reports (16,17), intramuscular injection of ASFV was lethal and two wild-type pigs died within 7 days, although surprisingly, one outlier (W4) survived over 2 weeks with elevated ASFV titers in the blood (Fig. 3e).As expected, the pigs challenged indirectly survived longer (13.5 days vs 9.7 days) than directly challenged groups.
Interestingly, we observed no significant survival advantage of the transgenic pigs compared to the wild-type pigs in direct exposure groups (Fig. 3d, G1, G2, and G3 vs W4, W5, W6).This indicated that CRISPR-Cas was not sufficient to prevent ASFV replication in vivo.However, pigs cohabitating with infected transgenic pigs had a consistent extension in survival time (Fig. 3d and G4 vs G5; W1, W2, W3, vs W4, W5, W6), suggest ing a potential delay in virus spread due to the CRISPR-Cas system.Notably, G4, the only transgenic pig that cohabitated with the infected transgenic pigs, demonstrated a significant delay in viral titer elevation (Fig. 3e).We reasoned that this could be attributed to the combination of the anti-ASFV CRISPR-Cas genome modification and the slower virus spread due to the effect of the CRISPR-Cas system in its cohabitants.However, by day 15, both wild-type and transgenic pigs succumbed to viral infection, highlighting the need for further optimization of the current strategy to combat ASFV in vivo (Fig. 3d).

DISCUSSION
In summary, we developed a CRISPR-Cas9 system targeting nine genomic loci within the ASFV genome.We demonstrated that these constructs effectively disrupt the ASFV genome and manage viral titers in vitro.Interestingly, we discovered that both Type II (EF-1a) and III (U6) RNA polymerases, in addition to Cas9 with or without NLS, can achieve substantial ASF viral control in vitro.Moreover, we successfully bred a pig strain expressing this multiplexed anti-ASFV CRISPR-Cas system through germline genome editing.We observed a delayed onset of ASF symptoms in the transgenic herd, suggesting that pigs carrying the multiplexed CRISPR-Cas9 modification might exhibit some degree of resistance to ASF.However, transgenic pigs did not significantly outlive their wild-type counterparts upon direct ASFV challenges.Further investigations to refine the gene editing system and conduct comprehensive cohort studies are needed to clarify the effects of multiplexed anti-ASFV CRISPR-Cas9 and determine its potential for application in agricultural and clinical contexts.To the best of our knowledge, our study presents the first example of an animal with an anti-virus CRISPR element in its genome generated via germline editing.Prior studies showed successful suppression of ASFV replication in vitro by targeting a single ASFV genome site (the CP204L gene) with CRISPR-Cas-gRNA in cell lines (9); however, it remains unknown whether such a mechanism works in vivo.In addition, targeting a single genomic locus is vulnerable to escape mutants; hence, a multiplexable gene editing approach is promising to eliminate the virus before accumulating mutants circumvent the acquired defense.
Encouragingly, we observed that pigs cohabitating with infected transgenic pigs exhibited a delay in ASFV-related death.Also, one transgenic pig that lived with other infected transgenic pigs had significantly lower serum viral titers, suggesting CRISPR-Cas could slow the virus's spread within the herd.Nevertheless, we did not observe a significant survival advantage for transgenic pigs facing ASFV challenges.This might be due to the high viral titers in our experiment, far exceeding those in typical contact-based infections.In addition, the expression level of Cas9-gRNA in the pigs might not have been sufficient.As suggested by our in vitro assays, the Cas9-gRNA expression level is inversely proportional to the extent of viral replication (Fig. 2b), so a higher expression level could potentially enhance immunity against ASFV.In the future, advanced genome-engineering work could be done to improve herd immunity in pigs, including increasing Cas9 and gRNA expression levels, delivering both Cas9 constructs with and without NLS to maximize the odds of cutting the viral genome throughout its life cycle (Fig. 1b and 2c), and using synthetic approaches to drive Cas9 expression in a tissue-specific or pathogen-induced manner.
Preventing the spread of ASFV is not just crucial for agriculture but also car ries significant clinical implications.The recent advances in xenotransplantation have encouraged more clinical trials involving pig organ transplants in human patients (18,19).However, this also necessitates stringent monitoring and control of zoonotic diseases transmitted by endogenous pig viruses like PERVs (20,21) or exogenous porcine viruses such as ASFV and pig CMV.As we approach the end of the COVID-19 pandemic, it is imperative to preempt any potential cross-species viral transmission via pig organs to ensure the safety of clinical applications (22).

CRISPR-Cas9 gRNA design
R library DECIPHER was used to design specific gRNAs (Table S2).Guide RNAs are designed to target the ASFV genome based on the BA71V strain (RefSeq: GCF_000858485.1).The potential gRNA off-targets in a large white pig genome are predicted bioinformatically.Select six sgRNAs with a lower probability of off-target effects.

ASF virus strain
The gz201801 strain was isolated by Zhang et al. from South China Agricultural University.The virus strain was sequenced, and the data were subsequently uploaded to the NCBI database (https://www.ncbi.nlm.nih.gov/nuccore/MT496893.1).The strain exhibited a high degree of purity, as demonstrated through fragment analysis of the PCR data, specifically for the target genes of interest.

IVT and Cas9 cleavage assay
PCR products were from a pBv1-U6-BB plasmid with primers ASFV-sg01-T7IVT-F (ctaatacgactcactataggggcttgcacaggtgtctacat)and pBv1-U6-Filler-XhoI-R, (ctcgagttagcggcatccctgcaagg) to append the T7 promoter at the 5′ end of the multiplex guide RNA cassette.The PCR products were TOPO cloned using pClone007 Versatile Simple Vector Kit (Tsingke) and the TOPO plasmid was isolated.The TOPO plasmid containing the multiplex guide RNA cassette was digested with Xho I and purified to be used as an IVT template.IVT was carried out using MEGAscript T7 Kit (Thermo Fisher Scientific) and purified by MEGAclear Kit (Thermo Fisher Scientific) following the manufacturer's instructions.The purified IVT product was analyzed using 2% agarose gel electrophoresis (Fig. 2a).DNA substrates containing the guide RNA target sites were PCR amplified from purified ASFV genomic DNA using primers detailed in Table S2.
In vitro, the Cas9 cleavage assay was carried out using purified IVT product and Cas9 protein (Cas9 Nuclease, S. pyogenes, New England BioLabs) following the manufacturer's instructions.The digested products were analyzed using 2% agarose gel electrophoresis (Fig. 2a).
Gene editing of pig fibroblast cells was performed by electroporation of wildtype large white pig fibroblast cells with a plasmid encoding Piggybac transposase and pBv1-EF-6X plasmid using Neon Transfection System (Invitrogen).The edited cells were single-cell sorted using SONY SH800S into 96-well plates, single-cell clones were grown and clones positive for Cas9 gene cassette were chosen for Cas9 expression analysis.Clone number GI58 with an expression of Cas9 was chosen for pig cloning (Fig. 3a).
EDTA anticoagulated blood and whole blood were collected every 2 days from day 1 until the end of the experiment or the death of animals.During the experiment, dying animals were euthanized and dissected for sampling.The experiment ended on day 15.All remaining animals were euthanized then dissected for sampling.
ASFV antigens were tested in EDTA anticoagulated blood samples and swab samples through real-time PCR.During the necropsy, various tissues and organs were harvested, and the presence of ASFV antigens in these fresh samples was assessed using real-time PCR (Table S3).

FIG 1
FIG 1 Multiplex CRISPR strategy targeting the ASFV genome to protect pig cells from ASFV infection.(a) DNA construct encoding six sgRNAs, each flanked by HH ribozyme and HDV ribozyme.Genes can be transcribed under one single promoter into a single RNA transcript that can be automatically processed into six mature sgRNA.The mature sgRNA assembles with Cas9 protein to form the catalytic active CRISPR complex.(b) sgRNA and Cas9 are constitutively expressed in the pig cells, cutting invading the ASFV genome either in the nucleus (route I) or cytoplasm (route II).The viral genome is subsequently degraded, and viral replication is stopped or attenuated.(c) Design illustration of the constructs for expressing both Cas9 protein and six sgRNAs.The transgenes are flanked by ITR sequences and inserted into the pig genome via Piggybac transposase.We tested Cas9 with and without the NLS signal.We also tested hEF1a and U6 promoter for 6-sgRNA expression.

FIG 2
FIG 2 Multiplex CRISPR strategy can efficiently cut ASFV DNA in vitro and different construct design has different effects on restricting ASFV replication in COS-7 cells.(a) Left panel, the single RNA transcript generated by IVT can be efficiently processed into monomeric gRNA with few dimeric (blue triangle) and trimeric gRNA.Right panel, the IVT single RNA transcript can mediate efficient cleavage of PCR amplicons amplified from the ASFV genome in vitro Cas9 cleavage assay."DNA target #" indicates the PCR amplicon corresponding to the respective gRNA.The starting PCR amplicons (lane 01a, 03a, 04a, 05a, 06a, PCa) are digested into two or more fragments (red triangles, lane 01b, 03b, 04b, 05b, 06b, PCb.PC: positive control) after co-incubation with IVT single RNA transcript and Cas9 protein.(b) Engineered COS-7 single-cell clones with different gRNA promoters and Cas9 versions (left table) and relative Cas9 expression levels in those clones (right panel) determined by RT-qPCR.() ASFV replication in different COS-7 single-cell clones was measured by ASFV copy number in cell lysis using qPCR over 5 days.Strong inhibition of ASFV replication was only observed in COS-7 clones with high levels of Cas9 expression (GC49, GE22, GE64).

FIG 3
FIG 3 Production of transgenic pigs constitutively expressing Cas9 and 6 sgRNA targeting ASFV genome and virus challenge of the transgenic pigs.(a) Top table, the large white pig fibroblast single-cell clone GI58 harbors the pBv1-EF-6X construct with hEF1a promoter driving sgRNA and mEF1a promoter driving Cas9 with NLS.Bottom left panel, the presence of the transgene (Continued on next page)

FIG 3 (
FIG 3 (Continued) cassette was confirmed by PCR for the Cas9 gene in the genomic DNA of GI58.Bottom right panel, Cas9 expression was confirmed by RT-qPCR in the fibroblasts of the cloned pigs (GI58P10 and GI58P12) generated from GI58 fibroblasts.(b) photo of the cloned pigs generated from GI58 fibroblasts.(c) ASFV virus challenge design of the gene-edited pigs and wild-type pigs.Direct virus challenge was performed by muscular injection, while indirect virus challenge was performed by the cohabitation of the pigs in the same room with pigs under direct virus challenge.Each rectangle with four black squares indicates a separate room.In the light-red room, three gene-edited pigs were subjected to direct virus challenge, while in the light-gray room, three wild-type pigs were subjected to direct virus challenge.The additional one gene-edited pig and three wild-type pigs in each room are subjected to an indirect virus challenge.(d) Survival of the pigs in direct and indirect virus challenge.No statistical difference in survival time was observed in the two virus challenge modes comparing gene-edited pigs with wild-type pigs under the same condition.The arrow indicates indirect virus challenge, for example, W→G means wild-type pigs were under direct virus challenge and gene-edited pigs were under indirect virus challenge released by infected wildtype pigs.(e) ASFV titer in blood samples of the pigs in direct and indirect virus challenge measured by qPCR.Gene-edited pigs showed no detectable viral titer, much lower than wild-type pigs, in an indirect virus challenge released by infected gene-edited pigs (middle panel).No difference was observed in other conditions.