Porcine epidemic diarrhea viruses from Vietnam: isolation, characterization, and neutralizing activity
Article information
Abstract
Porcine epidemic diarrhea (PED) is characterized by acute enteritis, watery diarrhea, weight loss, dehydration, and death, with high mortality in neonatal piglets. In this study, 3 virus isolates collected in Vietnam between 2016 and 2017 were propagated successfully in Vero cells at high virus titers. Sequence analysis of the full-length spike (S) gene showed that all 3 isolates belong to genogroup 2b, which is closely related to other prevalent Asian strains. A comparison of the amino acid sequence revealed a 98.19% to 99.13% homology with the Vietnam isolates circulating during 2013–2015, suggesting that field PED viruses (PEDVs) are evolving continuously. Experiments in animals showed that the antisera from guinea pigs immunized with the vaccine strain resulted in higher levels (5 log2) of neutralizing antibodies against the homologous strain and a relatively moderate level of neutralizing antibodies against the field isolates. This finding would be helpful in selecting a PEDV strain for vaccine development.
Introduction
The porcine epidemic diarrhea virus (PEDV) causing porcine epidemic diarrhea (PED) is a large-enveloped RNA virus belonging to the genus Alphacoronavirus, family Coronaviridae, order Nidovirales. The PEDV genome is approximately 28 kb long containing at least 7 open reading frames (ORF1a, ORF1b, and ORF2–6) and a 3’ untranslated region [1,2]. A PEDV infection in pigs causes watery diarrhea, anorexia, and depression leading to high mortality rates in suckling piglets [1,3].
Over the last 30 years, the disease has been recorded in European and Asian swine farms, with the virus initially isolating in England [4] and Belgium [5] in the late 1970s. PEDV has been reported since 2008 in Vietnam, and the outbreak was a significant event that affected the country’s swine industry. The first reported case of PED occurred in the southern provinces of Vietnam and quickly spread throughout the nation, with significant economic losses [6]. According to a recent descriptive survey, the northern region of Vietnam, where many farms tested positive for PED (30.89%), is an endemic PEDV area [7]. Although some pig farms have used vaccination or the feedback approach, PEDV remains and continually recurs [7].
PEDV isolates need to propagate efficiently in cell culture to examine pathogenesis and vaccine development because of the difficulty growing PEDV in cell culture. The PEDV can be isolated from clinical samples, but the virus may gradually lose infectivity upon further passages in cell culture [8].
The spike (S) protein of PEDV interacts with the cellular receptor in the entry process of the virus and produces neutralizing antibodies in pigs [1]. The S protein is considered a primary target antigen for the induction of neutralizing antibodies and the development of an effective vaccine against PEDV [9,10]. Recent reports have demonstrated the diversity of the S gene in many isolates. The pathogenicity and antigenicity of the PEDV can be changed according to mutations, deletions, and insertions of amino acids in the S protein. As a result, viral antigenicity and viral neutralizing activity may be changed [11,12]. Zhang et al. [12] reported that a field CHN/FL2013 strain reduced virulence in cell culture with a seven-amino acid deletion at N- termination of the S protein.
In this study, PEDV isolates from Vietnam were propagated in Vero cells, and full-length S genes were sequenced to examine the diversity of PEDVs. In addition, the cross-neutralizing activity of the serum of guinea pigs immunized with a PEDV vaccine strain against the isolates was evaluated.
Materials and Methods
Isolation of PEDVs in Vero cells
Stool specimens and the small intestine were taken from suckling piglets and post-weaning pigs exhibiting acute watery diarrhea. The PEDV HID9047 and HID9048 isolates used in this study were collected from infected samples in Ha Nam and Thai Binh provinces, Vietnam in 2016, and the PEDV HID9049 isolate was collected from infected samples in Hung Yen province, Vietnam in 2017.
The presence of the PEDV was confirmed by a reverse transcription polymerase chain reaction (RT-PCR) amplifying approximately 900 bp of the N gene with primers PEDV-N–F (5′-CGG TTC TCA CAG ATA GTG A-3′) and PED-N-R (5′- CTC CTC CAC TCT GGG ATG T -3′).
The isolation and propagation of PEDV in Vero cells were conducted using a slight modification of the methodology reported by Park et al. [13]. Briefly, 80% confluent Vero cells were inoculated with the virus at 37°C for one hour. The infection medium (Dulbecco's modified Eagle's medium [DMEM] supplemented with 0.02% yeast extract, 0.3% tryptose phosphate broth, 5 μg/mL trypsin, and 1% antibiotic–antimycotic) was added to the infected cell flask without removing the inoculum. The infected cells were incubated at 37°C for 7 days until extensive cytopathic effects (CPEs) were observed; viruses were harvested by 3 freeze-thaw cycles.
Virus titration
Vero cells (1 × 105) were seeded in each well of a 96-well plate (Corning, USA) and grown at 37°C one day before the assay. The virus was diluted in the infection medium, and 0.2 mL of 10-fold serial virus dilutions were added to each well after removing the medium. Each virus dilution was used to infect 10 wells of monolayer Vero cells. The CPE was examined under a microscope every day for 7 days postinfection, and the virus titer was calculated as the 50% tissue culture infective dose (TCID50) using the Reed-Muench method [11].
Immunofluorescent assay
Vero cells were infected with the PEDV 24 hours before the assay. The control and infected cells were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100. The cells were then blocked with 5% bovine serum albumin, followed by incubation with 1/1,000 monoclonal mouse anti-PEDV (Median Diagnostics, Korea) and 1/1,000 fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse immunoglobulin G (IgG) (MP Biomedicals, USA). The signals were observed using fluorescence microscopy [13].
Nucleotide sequence analysis
The full-length S genes of PEDVs were amplified by RT-PCR using the primers listed in Table 1. The individual fragment was cloned into the T&A cloning vector (RBC; Real Biotech Corp., Taiwan) and sequenced in both directions. The sequences of the full-length S genes of the PEDV isolates in this study and reference PEDVs were used in sequence alignments and phylogenetic analyses with the Serial Cloner 2.6 (SerialBasic, USA), the Clustal X 2.0 program (European Bioinformatics Institute, UK), and the Mega 6.0 program (megasoftware, USA) [14].
Serum neutralization
The neutralizing activity of the antisera collected from guinea pigs immunized with a PEDV vaccine strain SM98 against the PEDV isolates HID9047, HID9048, and HID9049 was evaluated. Three to four-month-old guinea pigs (n = 3) were immunized twice intramuscularly with 105 TCID50 inactivated vaccine strain SM98 at a two-week interval. Two other guinea pigs were given phosphate-buffered saline (PBS) as a control. Serum samples were collected 2 weeks after the last immunization for the neutralizing assay.
The Vero cells were grown at 2 × 105 cells per well in 96-well plates. The viruses were diluted to 200 TCID50 in 50 µL DMEM, mixed with 50 µL of a twofold diluted serum, and incubated at 37°C for one hour. The cells were rinsed 3 times with PBS, followed by inoculating 100 µL of the mixture at 37°C for one hour. After removing the mixture, the cells were rinsed 3 times with PBS, and the infection medium was added to the cells and incubated at 37°C in an atmosphere containing 5% CO2. The neutralization titer was calculated as the highest serum dilution inhibiting virus-specific CPE [14].
Statistical analysis
One-way analysis of the variance in GraphPad Prism 6 (GraphPad Software Inc., USA) was used, and p-values < 0.05 were considered significant.
Ethics statement
Animal experiments were approved by the Institutional Animal Care and Use Committee (approval number: KW-181031-1) and performed at the Center for Animal Experiments, Kangwon National University.
Results
Virus isolation and characterization
PEDV-PCR positive samples (Fig. 1A) were processed and incubated in Vero cells. Three PEDV isolates, designated HID9047, HID9048, and HID9049, were propagated successfully in Vero cells with clear CPEs, such as cell fusion, large syncytia, and cell detachment (Fig. 1B). The viral titers of HID9047, HID9048, and HID9049 were 107.5, 108.2, and 108.5 TCID50/mL at passage 10, respectively. Immunofluorescence assay using the monoclonal mouse anti-PEDV and FITC-conjugated goat anti-mouse IgG antibodies confirmed virus growth in Vero cells. A specific fluorescence signal was detected in the cytoplasm of Vero cells infected with viruses but not in the non-infected cells (Fig. 1C).

(A) Reverse transcription polymerase chain reaction amplifying a 900 bp fragment of the N gene. M, DNA standard 3 kb plus; -, Vero cell control; 1, SM98; 2, HID9047; 3, HID9048; 4, HID9049. (B) Isolation of Vietnam porcine epidemic diarrhea virus (PEDV) isolates in Vero cells at 5 days postinfection with obvious cytopathic effects, such as cell fusion and large syncytia. (C) Immunofluorescence assay in Vero cells using monoclonal mouse anti-PEDV and fluorescein isothiocyanate-conjugated goat anti-mouse immunoglobulin G antibodies.
Phylogenetic analysis
Fifty-four PEDVs were used to construct a phylogenetic tree (Table 2). In this study, phylogenetic analysis based on the nucleotide sequences of the S gene showed that PEDV strains could be divided into 2 genogroups, classical genogroup 1 represented by prototype CV777 and variant genogroup 2 of pandemic strains. Each group can be divided into subgroups 1a, 1b, 1 S-INDEL, 2a, and 2b. Three isolates (HID9047, HID9048, and HID9049) belonged to the subgroup 2b, which is most closely related to the Asian strains (Fig. 2).
Spike gene sequence comparisons
The entire DNA sequences of the S genes and the deduced amino acid sequences of 3 isolates from Vietnam were compared with the sequences of the reference strains (CV777, SM98, and DR13) and other Vietnam isolates. As shown in Table 3, the nucleotide identities of the S gene between 3 Vietnam isolates and reference strains ranged from 91.80% to 92.71%, and the deduced amino acid sequences were 96.74% to 97.61% homologous. Furthermore, the 3 isolates shared 96.78% to 97.52% nucleotide homology and 98.19% to 99.13% amino acid homology compared with the Vietnam isolates (HUA-PED45, HUA-PED111, VN97, and VN-K28). The nucleotide and amino acid homology of the 3 isolates showed 98.15% to 99.76% and 99.06% to 99.64%, respectively (Table 3).

Pairwise comparisons of the nucleotide and deduced amino acid sequences of the full-length S gene of PEDV isolates compared with the reference strains.
The sequence alignment of amino acids in the S protein illustrated that mutations in amino acid sequence, including substitutions, insertions, and deletions, were observed in 3 Vietnam isolates. Among them, 4 insertions (58NQGV61) and 2 deletions (163DI164) were found in the N-terminal of the S protein (Fig. 3). Alignment analysis of 4 neutralizing epitope regions core neutralizing epitope (COE), SS2, SS6, and 2C10 showed that all 3 Vietnam isolates in this study have 4 substitutions in COE (502I → 502T, 509L → 509S, 529S → 529G, and 607A → 607E), one substitution in SS6 (766Y → 766S), and no mutations in SS2 and 2C10 compared with CV777, DR13 and SM98 strains (Fig. 3). A comparison with other Vietnam strains (VN288, VN-Jafa, VN01, VN97 isolated during 2012–2015) indicated the substitution in COE (502I → 502T), and the conservation in SS2, SS6, and 2C10 region. PEDV VN288, VN-Jafa, VN01, and VN97 showed addition substitutions in COE, while HUA-PED45 showed more homology with 3 isolates, suggesting that 3 isolates (HID9047, HID9048, and HID9049) are closely related to PEDV HUA-PED45.

Analysis of amino acid mutations in the S protein of isolates from Vietnam with reference strains. The insertions and deletions are highlighted in blue and red, respectively. The core neutralizing epitope (COE) and 3 other neutralizing epitopes, SS2 (yellow), SS6 (green), and 2C10 (purple), were aligned.
Neutralizing activity
Guinea pig antisera collected 2 weeks after the final immunization were used for neutralizing tests to determine if vaccine strain SM98 could neutralize the 3 PEDV isolates (HID9047, HID9048, and HID9049). The antisera effectively inhibited infection with the vaccine strain SM98 with mean neutralizing antibody (NA) titers of 5 log2. The antisera showed lower dilutions with mean of NA titers of 3.67 log2 (p < 0.01), 4 log2 (p < 0.05), and 3.27 log2 (p < 0.001) for inhibiting infection with isolates HID9047, HID9048, and HID9049, respectively (Fig. 4). A significant difference in the mean NA titers against the SM98 strain and 3 isolates was observed, suggesting that the antisera reacted strongly with the homologous virus, but with the heterologous field isolates regarding to antigenic variations between PEDVs.

Neutralizing antibodies in the serum samples of guinea pigs immunized against porcine epidemic diarrhea virus isolates. Serum samples were collected 2 weeks after the second immunization and subjected to the virus neutralization assay using PEDV HID9047, HID9048, and HID9049 isolates. Data represent the mean standard deviation. Statistical significance was assessed by one-way ANOVA. *p < 0.05, **p < 0.01, and ***p < 0.001.
Discussion
The S protein is an essential factor mediating PEDV entry into host cells and is considered a primary target antigen for the induction of neutralizing antibodies and the development of an effective vaccine against PEDV [9–11]. Several studies suggested that the S gene is prone to mutations, including replacement, insertion, and deletion of nucleotide sequences [11,12,15,16].
The PEDV has been reported in Vietnam since 2010 and has spread rapidly [17]. The disease has been circulating and becoming a significant issue in the pork industry. This study was conducted to isolate PEDVs from infected pigs in Vietnam, investigate the genetic variation of PEDV isolates based on the full-length S gene, and assess the neutralizing activity of the traditional vaccine strain with the circulating isolates.
This study isolated 3 PEDV isolates from infected samples and propagated them in Vero cells with high virus titers. Nucleotide sequencing and a phylogenetic tree based on the S gene revealed that the 3 isolates from Vietnam (HID9047, HID9048, and HID9049) belong to subgroup 2b. The 3 PEDVs were isolated in 2016 and 2017, suggesting that they are novel PEDVs that share unique genetic features with the subgroup 2b PEDV strains circulating in Vietnam from 2013–2015. In addition, the 3 isolates showed fewer mutations than PEDV HUA-PED111, HUA-V2, and HUA-M17, which were isolated from the northern provinces of Vietnam in 2015, but showed higher homology in amino acid sequences with strain HUA-PED45. Hence, the 3 isolates and PEDV HUA-PED45 are genetically closely related. Amino acid sequence alignment indicated mutations, such as substitutions, mutations, insertions, and deletions in S protein, particularly in the N-terminal region. The N-terminal domain interacts with 5-N-acetylneuraminic acid, and mutations in the N-terminal may impact the PEDV infection in the host [18].
The neutralizing activity of antisera against the 3 Vietnam isolates was determined. Immunized sera showed higher neutralizing activity against the homogenous strain SM98 but moderate against the isolates HID9047, HID9048, and HID9049. This finding is consistent with a previous study showing that the antisera strongly recognized the homologous strain regarding antigenic variations [14]. Lee et al. [14] reported that the antisera of guinea pigs given 2 immunization with strain KNU-141112 effectively inhibited the homogenous strain with mean NA titers of 1:112, but relatively low NA titers of 1:37 against strain SM98-1. In this study, the 3 isolates have one and 4 amino acid substitutions in the SS6 and COE regions, respectively, which may support the hypothesis of variations in neutralizing the epitope and speculate that the changes may influence the antigenicity of PEDV. Further studies on the pathogenicity and neutralizing activity of these 3 isolates need to be carried out in pigs.
Notes
The authors declare no conflict of interest.
Acknowledgements
The authors wish to acknowledge the veterinarians who collected the infected samples and the contribution of all authors for this study.