Duck enteritis virus (DEV) is an alpha-herpesvirus that infects ducks, geese and swans [1
]. It causes acute contagious diseases in susceptible birds with mortality rates that can reach 100% in ducks, and is therefore a significant economic concern [1
]. The viral genome comprises a double-stranded DNA of about 160 kilobase pairs, and comprises a number of unique long (UL) and unique short (US) sequence formats. Live attenuated DEV vaccines are widely used for reduction of diseases impact in ducks [1
]. Due to its large genome and narrow host range, DEV has been exploited as a vector for development of recombinant multivalent vaccines [2
]. Virus vector vaccines have advantages over inactivated vaccines through induction of both cellular and humoral responses, and enabling differentiation between infected and vaccinated animals (DIVA) approaches [4
Avian influenza viruses (AIV) are enveloped and carry a segmented and negative-strand RNA genome characterized by very high evolutionary rates that facilitate the process of antigenic drift and immune escape. Aquatic birds form the enzootic reservoir for the majority of subtypes of influenza A viruses [5
]. Domestic duck populations in southeast Asia are considered to play a key role in maintenance of highly pathogenic avian influenza (HPAI) subtype H5N1 viruses [6
]. Ducks can frequently sustain HPAI infection without any overt signs of disease, thus enabling “silent” transmission cycles and continued opportunities for “spillover” transmission to in-contact chickens, causing outbreaks that are often extremely lethal and difficult to contain [7
]. Thus, vaccination to protect ducks against H5 HPAI is warranted to reduce production losses to duck farmers, to safeguard against transmission to other poultry species, and to mitigate the risks for zoonotic emergence.
Several genome modification methods have been adopted in the past to produce duck enteritis vector vaccines, such as homologous recombination, bacterial artificial chromosome (BAC), and fosmid system construction [2
]. However, these methods are generally time-consuming and labor-intensive. Clustered regularly interspaced palindromic repeats (CRISPR)/associated (Cas9) is a gene-editing technology that has gained popularity in recent years for its versatility and specificity. In this system, a single guide RNA (sgRNA) recognizes a 20 nucleotide target sequence adjacent to a 5′ NGG 3′ protospacer adjacent motif (PAM), and Cas9 introduces a double strand break (DSB) in this target sequence. The DSBs can then be repaired by either the error-prone non-homologous end-joining (NHEJ) or the high-fidelity homology-directed repair (HDR) pathway [10
]. Extensive research has demonstrated the value of the HDR-CRISPR/Cas9 system for vaccine development [3
], whereas the alternative approach utilizing the NHEJ-CRISPR/Cas9 system is less well established, despite potential higher insertion efficiency in comparison to the HDR method.
The Cre-Lox system is a site-specific recombination system that has been used to excise BAC [13
]. The Cre recombinase enzyme, originally derived from the P1 bacteriophage, can recognize specific 34 base-pair DNA sequences called Lox sites and the DNA between Lox sites can be excised [14
In this study, we employed the NHEJ-CRISPR/Cas9 system for DEV-AIV bivalent vaccine development. We introduced a GFP expression cassette into the DEV genome as an indicator to confirm the foreign gene insertion and expression, which was later removed using the Cre-Lox system. We show that NHEJ-CRISPR/Cas9 together with Cre-Lox is an efficient method for rapid generation of a recombinant DEV-AIV vaccine.
2. Materials and Methods
2.1. Viruses, Cells and Transfection
We obtained a prototype strain of duck enteritis virus from LGC Standards (England, UK) (ATCC® VR-684™). The virus was propagated in primary chick embryo fibroblasts (CEF) prepared from 10-day-old specific-pathogen-free embryonated chicken eggs, and virus stocks were kept at −80 °C. Cells were maintained with Dulbecco’s Modified Eagle’s medium (DMEM) (Gibco, Life Technologies Ltd., Paisley, UK), supplemented with 10% fetal calf serum (FCS) (Gibco), 100 U/mL penicillin, and 100 μg/mL streptomycin (Gibco) at 37 °C under a 5% CO2 atmosphere. CEF cells were transfected with plasmids using TransIT-X2® (Mirus, Cambridge Bioscience, Cambridge, UK) according to the manufacture’s protocol.
2.2. Virus Infection and Titration
CEF cells were washed once with phosphate-buffered saline (PBS) before being infection with the DEV virus. The inoculum was removed at 2 h post-infection and replenished with either fresh medium or 2% Minimum Essential Medium (MEM)-agarose overlay. The MEM-agarose overlay medium contains MEM (Sigma, St Louis, MO, USA), 2% agarose (Thermo Fisher Scientific, Waltham, MA, USA), 100 units/mL penicillin, 100 µg/mL streptomycin, 2 mM l-glutamine (Sigma), 0.3% bovine serum albumin (BSA) (Sigma), 15 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) (Sigma), 0.22% sodium bicarbonate (Sigma) and 0.01% Diethylaminoethyl (DEAE)-Dextran (Sigma). For the virus plaque titration, infected cells were incubated at 37 °C under a 5% CO2 atmosphere for 5 days before being fixed with 1% crystal violet (Sigma) in 20% ethanol for plaque counting and plaque size measurement.
2.3. Multi-Step Growth Curve
CEF cells were infected with DEV at multiplicity of infection (MOI) 0.01. The supernatant and cells were harvested at 6 h, 12 h, 24 h, 48 h and 72 h post-infection. The harvested virus was kept at −80 °C until further analysis.
CEF cells were infected with DEV at MOI 0.01 for 48 h and then fixed in acetone:methanol (1/1) for 10 min, followed by incubation in blocking buffer (5% FCS in PBS) for 10 min. The expression of H5 hemagglutinin (HA) antigen in DEV-AIV vaccine infected cells was visualized by incubating cells with AIV H5 HA-specific antibody (mouse monoclonal) diluted in blocking buffer (1 in 1000 dilution) for 1 h at room temperature. Cells were subsequently rinsed with PBS and probed with horseradish peroxidase-labeled rabbit anti-mouse immunoglobulins (DAKO, Agilent Technologies, Santa Clara, CA, USA) for 40 min. After gentle rinsing with PBS, cells were stained with 3,3′-diaminobenzidine (DAB) substrate-chromogen solution (DAKO) for 7 min. The stained cell images were taken using Leica TCS SP5 confocal laser scanning microscope (Leica, Wetzlar, Germany).
2.5. Western Blot
DEV infected-CEF cells were lysed using Radioimmunoprecipitation assay (RIPA) Lysis and Extraction Buffer (Life Technologies Ltd, Paisley, UK). HA and alpha-tubulin proteins were detected via Western blot using mouse monoclonal antibody for influenza virus H5 HA (1 in 2000 dilution) and rabbit polyclonal anti-alpha tubulin (Abcam, Cambridge, UK) antibody (1 in 6000 dilution) with corresponding secondary anti-mouse or anti-rabbit IgG antibodies (both were 1 in 10,000 dilution) labeled with florescent dyes IRDye 800CW or IRDye 680RD (Li-COR, Lincoln, NE, USA) respectively and visualized using the Odyssey CLx (Li-COR).
2.6. DEV Genome Extraction and High-Resolution Melting (HRM)
CEF cells were transfected with 1 µg of sgRNA per well of a 12-well plate before infection with DEV at MOI 1.0. The DEV infected cells were harvested at 48 h post infection and lysed in 1× squishing buffer (10 mM Tris-HCl, pH 8, 1 mM EDTA, 25 mM NaCl, and 200 µg/mL Proteinase K) at 65 °C for 30 min. Primers for HRM analysis were designed using Primer Express 3 (Thermo Fisher Scientific). The PCR was performed in a 7500 fast real-time PCR machine (Applied Biosystems, Foster City, CA, USA) and analysed using Applied Biosystems™ HRM Software v3.0 according to the manufacturer’s instructions.
2.7. Construction of sgRNAs and Donor Plasmids
The single nucleotide guide RNA (sgRNA) was designed using CRISPR design tool (http://crispr.mit.edu/
, Feng Zhang’s Lab). The universal bait sequence from copGFP was adapted from published data [15
]. The DNA oligo of sgRNA was synthesized (Sigma) and cloned into plasmid pX459-v2 (Addgene, Cambridge, Massachusetts, USA) using BbsI cloning sites. DNA containing two copies of bait sequence, two copies of Lox site, PacI enzyme site and BsmBI enzyme site was synthesized (IDT, Leuven, Belgium) and further cloned into plasmid pExpreS2-v1 (ExpreS2ion Biotechnologies, Hørsholm, Demark). The resultant plasmid was designated pExpreS2-v1-SgU-Lox. The GFP expression cassette was amplified from plasmid pEGFP-N1 (Addgene) and subsequently cloned into PacI site of pExpreS2-v1-SgU-Lox to construct pExpreS2-v1-SgU-Lox-GFP. The HA expression cassette was amplified from pGEMT-H5N8-HA (A/duck/England/36254/14) plasmid and further cloned into BsmBI site of pExpreS2-v1-SgU-Lox-GFP, the resulting plasmid was termed pExpreS2-v1-SgU-Lox-GFP-HA.
2.8. NHEJ-CRISPR/Cas9-Mediated Gene Insertion
CEF cells were transfected with 0.3 µg sg2, 0.3 µg sgU and 0.6 µg donor plasmids per well of a 12-well plate before infection with DEV at different MOI. Virus was harvested at 48 h post infection and subjected to the plaque purification.
2.9. Cre Enzyme Treatment
The GFP expression cassette was excised by Cre recombinase. Cells were transfected with Cre recombinase plasmid and then infected with DEV at MOI 0.01 or 0.0025 at 24 h post-transfection. The supernatants were harvest at 48 h post-infection and kept at −80 °C until further analysis.
2.10. Statistical Analysis
Statistical analysis was performed using GraphPad Prism 6 (GraphPad Software, La Jolla, CA, USA). Paired student t-test and one-way ANOVA were used to test differences between different groups. p values < 0.05 were considered significant.
To our knowledge, this is first study to explore the NHEJ-CRISPR/Cas9 system for DEV vaccine development. Here, we have demonstrated that the NHEJ-CRISPR/Cas9 system is an efficient method for gene knock-in with minimal indel generation. Furthermore, a universal donor system was established by introducing a bait sequence from the copGFP. It can be shared between different vector systems and will be beneficial in resource sharing and developing vaccines against a broad range of pathogens. In addition, the Cre-Lox system was demonstrated to be efficient for excision of the reporter gene from the virus genome in a dose-dependent manner.
CRISPR/Cas9 is an efficient tool for gene modification, and two distinct mechanisms, NHEJ and HDR, were used for recombinant vaccine development. NHEJ is error-prone and unpredictable, as the indel may be introduced during gene modification. So far, the majority of research favors the HDR mechanism due to its precision in modification [3
]. However, NHEJ is generally more efficient and occurs throughout the cell cycle whereas HDR is less frequent, occurring only during S and G2 phases of the cell’s life-cycle [17
]. Additionally, NHEJ is also free from the restriction of homology arms. Xu and colleagues used the NHEJ mechanism for RFP knock-in pseudorabies and reached high knock-in efficiency [18
]. However, only a single bait sequence from the virus genome was introduced to the donor plasmid, which may lead to large undesirable plasmid vector segment insertion into the recombinant viruses. In our system, two duplicate bait sequences from the copGFP were introduced to the donor plasmids. A guide RNA was introduced to cut and release the insertion segment. Because the bait sequence is independent of virus and mammalian cell genomes, the donor plasmids constructed can be shared widely between different insertion sites of one virus vector or between different virus vectors.
Consistent with previous applications of reporter gene knock-in in human cells via NHEJ-CRISPR/Cas9 [15
], we found minimal evidence for indels at the junction site of insertion. No vector DNA was found in the recombinant viruses. Because the indels occurred within the bait sequence, the open reading frame of the inserted expression cassette was not affected. As expected, gene expression from the knock-in cassette was not impacted by orientation of insertion.
In agreement with the study by Bi et al., we found the highest knock-in efficiency happened at 24 h rather than 6 h post-transfection [12
]. This is probably linked to the peak time for the expression of heterologous genes post-plasmid transfection. As the virus dose increased the knock-in efficiency dropped, likely due to excessive quantities of viral genomes entering the cells.
GFP expression cassette knock-in between the UL26 and UL27 intergenic region did not alter plaque size and the growth kinetics of DEV. This is in line with previous studies showing that H5HA insertion did not affect the growth property of DEV [3
]. However, plaques appeared relatively smaller and growth was markedly delayed when H5HA expression cassette was inserted. We speculate that this might be due to an influenza strain-specific effect or over expression of HA antigen. Alternatively, smaller plaque size may reflect the fact that different promoter and terminator elements were used in this study as compared to previous publications [3
]. Potential implications of slower replication of recombinant DEV-HA on the vaccine efficiency warrant further research.
To conclude, our study demonstrated that NHEJ-CRISPR/Cas9 system together with Cre-Lox is a powerful and speedy technology in recombinant vaccine development.