Vero CCL-81 cells (ATCC) were cultured in Dulbecco’s modified Eagle medium (DMEM) (HyClone) containing 10% fetal bovine serum (FBS) (Gemini Bio) and 1% penicillin–streptomycin (Gibco™). PEDV strain HM was isolated from the intestine sample of a neonatal piglet with acute diarrhea in China in 2016 using Vero CCL-81 cells supplemented with 10 μg/ml trypsin (Gibco™), and the passage 12 (P12) on Vero CCL-81 cells was used as seed stock in this study. A monoclonal antibody (mAb) against PEDV N protein was purchased from YouLong Biotech, and a mAb against β-tubulin was purchased from Bioworld.

The consensus sequence of the complete PEDV genomes was generated by multiple sequence alignment analysis using the CLC Genomics Workbench (QIAGEN) with all PEDV complete genomes in the GenBank sequence database. Seven pairs of primers targeting the highly conserved regions in the consensus sequence were designed to amplify the complete PEDV genome by reverse transcription PCR. The sequences of these primers are available upon request. Briefly, viral RNA of PEDV HM P12 was extracted with a viral DNA/RNA extraction kit (Vazyme Biotech) according to the manufacturer’s instructions. cDNA was generated with the Superscript™ IV first-strand synthesis system (Thermo Fisher Scientific) followed by viral RNA removal with RNase H (NEB). Seven overlapping fragments covering the PEDV HM P12 genome were PCR amplified using Q5 high-fidelity DNA polymerase (NEB). The PCR products were purified and sent for DNA sequencing by the GENEWIZ facility in Suzhou, China. Finally, the complete genome of PEDV HM P12 assembled with SnapGene was submitted to GenBank under accession No. MZ342899.

The pYES1L vector (Thermo Fisher Scientific) containing a yeast artificial chromosome (YAC) and a BAC was utilized to assemble an infectious cDNA clone of PEDV HM (P12) in yeast. With the YAC/BAC, this infectious cDNA clone will be able to replicate in both yeast and E. coli. We inserted the CMV early promoter, hepatitis delta virus (HDV) ribozyme sequence, and bovine growth hormone (BGH) termination signal into linearized pYES1L vector, generating the plasmid pYES1L-CMV-HDVrbz-bGH. The linearized pYES1L-CMV-HDVrbz-BGH vector was prepared by PCR amplification with Q5 high-fidelity DNA polymerase (NEB) using primers pYES1L-F and pYES1L-R. The cDNA of PEDV HM P12 was produced with the Superscript™ IV first-strand synthesis system (Thermo Fisher Scientific) followed by viral RNA removal with RNase H (NEB). With the viral cDNA, seven overlapping DNA fragments (F1: nt 1–4,020, primers YES1L-PEDV-F1/YES1L-PEDV-R1; F2: nt 3,986–8,107, primers YES1L-PEDV-F2/YES1L-PEDV-R2; F3: nt 8,062–12,071, primers YES1L-PEDV-F3/YES1L-PEDV-R3; F4: nt 12,034–16,836, primers YES1L-PEDV-F4/YES1L-PEDV-R4; F5: nt 16,807–20,223, primers YES1L-PEDV-F5/YES1L-PEDV-R5; F6: nt 20,191–24,495, primers YES1L-PEDV-F6/YES1L-PEDV-R6; F7: nt 24,455–28,064, primers YES1L-PEDV-F7/YES1L-PEDV-R7) covering the complete PEDV genome were amplified using Q5 high-fidelity DNA polymerase (NEB) according to the manufacturer’s instructions and then assemble with the linearized pYES1L-CMV-HDVrbz-BGH vector through transformation-associated recombination (TAR) in yeast. The overlapping regions (>30 nt) between neighboring DNA fragments enabled the homologous recombination in yeast. Briefly, the transformation of the MaV203 competent yeast cells (Thermo Fisher Scientific) with the mixture of 100 ng linearized vector and 200 ng of each F1–F7 DNA fragment was conducted with the PEG/LiAc solution according to the manufacturer’s instructions of the GeneArt® High-Order Genetic Assembly System (Thermo Fisher Scientific). Colony PCR was performed to screen yeast colonies containing the full-length cDNA clone pYES1L-PEDV using primer pairs, YES1L-PEDV-F1/PEDV-seq-R12 and PEDV-seq-F25/YES1L-PEDV-R7 (Table 1). The lysate of a positive yeast colony was further electroplated into DH10B competent E. coli cell to prepare pYES1L-PEDV plasmid using NucleoBond® Xtra Midi kit (MACHEREY-NAGEL). In addition, a plasmid pCAGGS-PEDV-N for ectopically expressing PEDV N was constructed by cloning N protein-coding sequence into pCAGGS vector using SacI and XhoI restriction enzyme sites.

A cDNA clone containing an EGFP gene was constructed by swapping a partial ORF3 coding sequence which is nt 24,856–25,395 of the PEDV HM genome. In brief, two guide RNAs, PEDV-sgRNA1 and PEDV-sgRNA2, synthesized by GenScript were used for CRISPR/Cas9 cleavage of the pYES1L-PEDV plasmid. The cleavage reaction containing 3 μg plasmid was conducted with Streptococcus pyogenes Cas9 nuclease (Vazyme Biotech) according to the manufacturer’s instructions. The linearized pYES1L-PEDV was verified by electrophoresis and purified with the Zymoclean Large Fragment DNA Recovery kit (ZYMO RESEARCH) per the manufacturer’s instructions. The EGFP gene amplified with primers EGFP-F and EGFP-R was inserted into the linearized pYES1L-PEDV through homologous recombination using HiFi DNA Assembly Master Mix (NEB) according to the manufacturer’s instructions. Two homologous arms between the EGFP gene and the linearized pYES1L-PEDV are around 20 bp in length. The recombinant plasmid was designated as pYES1L-PEDV-EGFP.

BHK-21 cells at ~80% confluence into a six-well culture plate were co-transfected with 0.5 μg pCAGGS-PEDV-N and 2 μg full-length cDNA clone, pYES1L-PEDV or pYES1L-PEDV-EGFP, using Lipofectamine® 3000 transfection reagent (Thermo Fisher Scientific) according to the manufacturer’s instructions. At 2 days post-transfection (dpt), culture supernatant was harvested to infect Vero CCL-81 cells in a six-well plate. After 2-h incubation, the cell monolayers were rinsed twice with 1xPBS and further cultured with 3 ml of DMEM supplemented with 10 μg/ml of trypsin (Thermo Fisher Scientific) in a 37°C, 5% CO₂ incubator. Cells were observed daily for the appearance of cytopathic effect (CPE) or green fluorescence under the IX73 epifluorescence microscope (Olympus). Around 5 dpt, the culture supernatant was harvested as P0 virus and stored at −80°C and then further passaged on Vero CCL-81 cells for at least 10 times.

Vero CCL-81 cells in 12-well culture plates were infected with PEDV HM, rPEDV, or rPEDV-EGFP for 24 h. The cell monolayers were fixed with 4% paraformaldehyde for 10 min and then permeabilized with 1xPBS containing 0.1% Triton X-100 for 10 min at room temperature. After 30 min blocking with 1xPBS containing 1% bovine serum albumin, the cell monolayers were washed thrice with 1xPBS and then incubated with mAb against PEDV N protein at 37°C for 1 h. After five washes, the cell monolayers were stained with TRITC-conjugated goat anti-mouse IgG(H + L) (Jackson ImmunoResearch Inc.). Cell nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) solution (Solarbio life sciences) for 5 min at room temperature. After extensive washes with 1xPBS, fluorescent images were visualized with the IX73 epifluorescence microscope (Olympus).

Vero CCL-81 cells seeded in 12-well plates were infected with PEDV HM P12 or rPEDV, and mock infection was used as control. At 24 h post-infection (hpi), cells were harvested with IP lysis buffer (Beyotime Biotechnology) supplemented with protease inhibitor cocktail (Roche). The cell lysates mixed with 5x loading buffer were boiled at 95°C for 5 min and separated by electrophoresis in 12% SDS-PAGE gel. The proteins in SDS-PAGE gel were transferred onto nitrocellulose membranes (Sangon Biotech). After 1 h blocking with 10% skimmed milk, the membranes were incubated with primary antibodies against PEDV N protein and β-tubulin, followed by HRP-conjugated secondary antibody, and visualized with ECL substrate (Vazyme Biotech) using the Tanon 5200 Multi imaging system.

We confirmed that EGFP protein was expressed through an additional subgenomic RNA in Vero CCL-81 cells infected with the rPEDV-EGFP virus. Total cellular RNA was extracted from Vero CCL-81 cells infected with rPEDV-EGFP P5 virus using FastPure Cell/Tissue Total RNA Isolation Kit (Vazyme Biotech). The subgenomic RNA expressing EGFP was amplified with primers anchoring 5’UTR and EGFP using HiScript II One-Step RT-PCR Kit (Vazyme Biotech). The PCR product was purified and sent for DNA sequencing by GENEWIZ.

The confluent Vero CCL-81 cells in 24-well culture plates were infected with PEDV HM, rPEDV, or rPEDV-EGFP at a multiplicity of infection (MOI) of 0.1. Cells were incubated with virus supernatant diluted with DMEM for 1 h, then washed twice with 1xPBS, and supplemented with 1 ml DMEM containing 10 μg/ml of trypsin (Thermo Fisher Scientific). Virus supernatants of two wells for each virus were harvested at 0, 24, 48, and 72 hpi) for virus titration by TCID₅₀. The viral growth curves were created with GraphPad Prism 8.

The confluent Vero CCL-81 cells in 12-well culture plates were infected with 10-fold diluted PEDV HM, rPEDV, or rPEDV-EGFP, respectively. Cells were incubated with virus supernatant diluted with DMEM for 2 h, then washed twice with 1xPBS, and then overlaid with 5 ml of DMEM containing 1% UltraPure™ Low Melting Point Agarose (Thermo Fisher Scientific) and 10 μg/ml trypsin (Thermo Fisher Scientific). At 3 days post-infection (dpi), the cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet to visualize plaques.

We isolated the PEDV HM strain through inoculation of Vero CCL-81 cells with an intestine sample from a piglet with acute diarrhea. This virus was further passaged 12 times on Vero CCL-81 cells. At 2 dpi, the typical CPE formation of PEDV infection was observed in Vero CCL-81 cells infected with the P12 virus (Figure 1A). To determine the complete genome of this virus, a panel of primers targeting the highly conserved regions of the PEDV genome was designed to amplify seven overlapping fragments covering the entire genome. These seven fragments were sequenced by the Sanger sequencing method, and the complete genome was assembled with SnapGene software and saved under the GenBank accession No. MZ342899. We further conducted a phylogenetic analysis with this genome and 25 representative PEDV genomes downloaded from the GenBank database using MEGAX (Kumar et al., 2018). Based on the phylogenetic tree generated with the neighbor-joining method, PEDV strain HM was clustered with GII PEDV strains (Figure 1B) which are new PEDV variants and shared 98.2% sequence identity with AJ1102 strain which was isolated from a neonatal piglet with acute diarrhea in China in 2011 (Bi et al., 2012).

To assemble the full-length cDNA clone of PEDV through the TAR technology (Figure 2A), we chose the pYES1L vector containing the YAC/BAC because of its replication ability in yeast and high stability in E. coli. To facilitate TAR in yeast, the pYES1L vector was initially modified to include the CMV early promoter, HDV ribozyme, and BGH termination signal, and designated as pYES1L-CMV-HDVrbz-BGH (Figure 2B). Through transformation with linearized pYES1L-CMV-HDVrbz-BGH into the MaV203 competent yeast cells, seven overlapping DNA fragments were amplified and assembled into pYES1L-PEDV (Figure 2C). In addition, to distinguish from wild-type virus, a genetic marker which is a SalI site knockout in ORF1b was created by incorporating two nucleotide substitutions in primers pYES1L-PEDV-F5 and pYES1L-PEDV-R4 (Figure 2B). The yeast colony containing pYES1L-PEDV was screened with colony PCR with primer pairs, YES1L-PEDV-F1/PEDV-seq-R12 and PEDV-seq-F25/YES1L-PEDV-R7 (Figure 2D). All five yeast colonies screened are positive, suggesting the superior efficiency of the TAR method in the assembly of a large viral genome. Then, one positive yeast colony was lysis to transform E. coli, and the infectious clone plasmid was prepared for verification and virus recovery. Restriction fragment length polymorphism by MluI digestion and DNA sequencing was conducted to verify the sequence of pYES1L-PEDV (Figure 2E). The results suggested that the full-length cDNA clone of PEDV HM strain P12 was successfully constructed, and no deletion or insertion compared with the wild-type sequence was identified.

BHK-21 cells were co-transfected with pYES1L-PEDV and pCAGGS-PEDV-N to rescue the recombinant PEDV, and virus supernatant was further transferred to the confluent Vero CCL-81 cell monolayer (Figure 3A). At 3 dpi, the typical CPE characterized by cell fusion and syncytium formation was observed. The successful recovery of rPEDV was confirmed by indirect immunofluorescence assay (IFA) and western blot analysis (WB) with mouse mAb against PEDV N protein (Figures 3B,C). To rule to the possibility of contamination, the designed genetic marker which is the inactivation of the SalI site (nt 16,818–16,823) in ORF1b of rPEDV was verified by RT-PCR and DNA sequencing (Figure 3D). We further in vitro characterized the rescued rPEDV (P5) and the parental PEDV HM P12 by multiple-step growth curve and plaque assay using Vero CCL-81 cells. As shown in Figure 4A, rPEDV displayed similar growth behaviors to the parental virus, and both reached peak titers at 48 hpi. For the size and morphology, no significant difference was observed between the plaques formed in cells infected with the recombinant virus and the parental virus (Figure 4B).

The ORF3 gene is non-essential for PEDV replication and can be replaced with reporter genes (Wang et al., 2012; Beall et al., 2016; Wongthida et al., 2017; Peng et al., 2020). To prove that our YAC/BAC-based PEDV infectious cDNA is convenient for genome manipulation, the CRISPR/Cas9 technology and homologous recombination technology were employed to construct the pYES1L-PEDV-EGFP variant clone in which the PEDV ORF3 gene was swapped with an EGFP gene. As shown in Figure 5A, two guide RNAs, PEDV-sgRNA1 and PEDV-sgRNA2, compatible with S. pyogenes Cas9 protein were used to cleave at the sites of 24,855–24,856 nt and 25,395–25,396 nt within the ORF3 gene, and the linearized pYES1L-PEDV was purified for homologous recombination in vitro. The EGFP gene containing at least 20 nt homologous arm at both terminals with linearized pYES1L-PEDV was amplified with primers, EGFP-F and EGFP-R (Table 1). The homologous recombination was conducted with the purified DNA fragments using HiFi DNA Assembly Master Mix (NEB; Figure 5A). The recombination reaction was electroporated into DH10B competent cells using Gene Pulser Xcell™ (Bio-Rad) according to the manufacturer’s instructions, and colony PCR amplifying the EGFP gene was performed to screen colonies containing pYES1L-PEDV-EGFP, and the positive clones were verified by DNA sequencing (Figure 5B). To our surprise, all five colonies tested by colony PCR are positive, which suggested the high efficiency of our strategy in the manipulation of the PEDV genome.

To rescue the recombinant virus, the culture supernatant of BHK-21 cells co-transfected with pYES1L-PEDV-EGFP and pCAGGS-PEDV-N was used to infect Vero CCL-81 cells. At 24 h post-transfection (hpt), a green fluorescence signal was observed in BHK-21 cells with DNA transfection, but not in the mock-transfected cells. Also, the typical CPE with EGFP expression was observed in Vero CCL-81 infected with the rPEDV-EGFP virus at 3 dpi, but no green fluorescence was observed for CPE caused by rPEDV infection. To further confirm the recovery of rPEDV-EGFP, N protein expression in red color was only detected by immunofluorescence assay in cells infected rPEDV-EGFP (Figure 6A). We also confirmed that the EGFP gene was expressed as an additional subgenomic RNA. As shown in Figure 6B, sequence analysis of subgenomic RNA suggested that the body transcriptional regulating sequence (TRS) of ORF3 mediated the jump of the subgenomic RNA expressing EGFP to leader TRS within 5’UTR. rPEDV-EGFP was serially passaged in Vero CCL-81 cells to check the stability of EGFP insertion. As expected, the EGFP gene was amplified and verified by DNA sequencing for P5 and P9 virus, suggesting the genetic stability of this reporter virus in vitro (Figure 4C).

The in vitro growth kinetics of rPEDV-EGFP were evaluated by the growth curves and plaque assay in Vero CCL-81 cells. As shown in Figure 6A, the growth kinetics of rPEDV-EGFP is similar to those of rPEDV, although its titers are about 0.5 logs lower than those of rPEDV at 48 and 72 hpi. Taken together, those results supported that ORF3 is not essential for PEDV replication.