Mucosal immune response in pigs following the primary and secondary exposure to porcine epidemic diarrhea virus genogroups G1 and G2

Abstract

This study was conducted to evaluate cellular and humoral mucosal immune responses of pigs infected with porcine epidemic diarrhea virus (PEDV) genogroups 1 (G1) or 2 (G2). Anamnestic response following homologous or heterologous reinfection were also investigated. Forty-five PEDV-negative, 3-week-old pigs were allocated into 3 groups of 15 pigs each. Pigs were orally inoculated with EAS1 (G1) or CBR1 (G2) or cell culture supernatant and serially monitored for PEDV-specific IFN-γ producing cells (IFN-γ PC) and IgA antibody secreting cells (ASC). Three pigs served as baseline at day 0. The CD4⁺IFN-γ⁺ PC in G1 group was detected at 3 days post infection (dpi) in both EAS1 and CBR1 recall antigen. In contrast, CD4⁺IFN-γ⁺ PC in G2 group detected only when EAS1 was used. Similar findings were found with CD8⁺IFN-γ⁺ PC. After reinfection, only G2 exhibited a booster effect of CD4⁺IFN-γ⁺ and CD8⁺IFN-γ⁺ cells by heterologous antigen recall. Regarding CD4⁺CD8⁺IFN-γ⁺cells, G1 showed significantly higher levels at 3 dpi and demonstrated a secondary boost at 14 dpi following heterologous recall stimulation. In contrast, G2 showed a slight increase in response to both homologous and heterologous recall at 3 dpi. PEDV-specific IgA ASC were detected at 14 dpi. However, anamnestic response after reinfection was variant-dependent. Our results indicate that the G1 variant specific cellular response is triggered earlier or simultaneously than G2 variant. Meanwhile, the timing of humoral response was similar, and anamnestic response was driven by the variant involved during reinfection.

Introduction

Porcine epidemic diarrhea (PED) is a devastating enteric disease characterized by vomiting, acute watery diarrhea and dehydration in pigs. Clinical disease is significantly more severe in piglets under 7 days of age in which mortality can reach 100%¹. Since the first case of PED reported in the United Kingdom and Belgium in 1976–1978 ^1,2, the disease has continued to cause severe economic losses in many swine producing countries worldwide^{3, 4, 5, 6, 7, 8–9}. The etiological agent is PED virus (PEDV), an enveloped single-stranded positive-sense RNA virus in the genus Alphacoronavirus, family Coronaviridae and order Nidovirales. Based on features of the spike (S) gene, two genetically distinct PEDV variants, including genogroup 1 (G1) and genogroup 2 (G2), are currently recognized. Both PEDV variants have been reported as an emerging disease worldwide^{10, 11, 12, 13, 14, 15, 16–17}. The difference between those two variants is that G2 variants contain two insertions of 4 (⁵⁶GENQ⁵⁹) and 1 (¹⁴⁰N) amino acids, and one deletion of 2 amino acids (¹⁶⁰DG¹⁶¹)¹⁸.

PEDV first emerged in Thailand in 2007. Since then, the disease has reached an endemic state with over 80% of swine herds affected by PEDV. The dominant PEDV variant responsible for outbreaks belongs to G2 and is closely related to PEDV isolates from China¹⁹. Although the severity and frequency of clinical outbreaks have decreased, repeated PED outbreaks have still been reported including farms mainly with re-breaks associated with the presence of PEDV G2. However, new PED outbreaks with less severe clinical manifestation were reported in 2015. Further phylogenetical analysis demonstrated that these outbreaks were due to emergence of a classical PEDV G1 variant genetically distinct from the circulating endemic G2 variant²⁰.

Although the clinical differences, including severity, duration, and mortality rate between PEDV G1 and G2 variants has been well documented²¹the immunogenicity of different variants and the ability of exposure to one variant inducing cross protection to another variant has yet to be investigated.The objectives of the study, therefore, were to characterize the mucosal immune response, especially the production of IFN-γ producing cells and IgA antibody secreting cells (ASC) in mesenteric lymph node (MLN) of pigs experimentally exposed to either PEDV G1 or G2 variant, and to evaluate the mucosal immune response after experimental reinfection with homologous and heterologous PEDV variants.

Results

Detection of stimulated IFN-γ producing cells. To investigate the effect of in vitro homologous and heterologous recall antigen activation, lymphocyte subpopulation of PEDV-specific stimulated IFN-γ producing cells between G1I, G2I and negative control groups were evaluated. Red dash lines represent baseline of the control pigs at 0 dpi in each test. The cellular phenotype response showed a biological trend between 3 groups evaluated. Subpopulations of T lymphocytes including CD4⁺, CD8⁺, and CD4⁺CD8⁺ were further characterized for the IFN-γ production.

Heterologous recall antigen induced the highest PEDV-specific CD4⁺IFN-γ⁺stimulation effect in both the G1I and G2I groups, particularly at 3 dpi. Recall antigen activation of CD4⁺ cells collected from all groups were shown in Fig. 1. At 3 dpi, recall antigen activation of CD4⁺ cells was highly significant in G1I group when stimulated with CBR1 PEDV variant, whereas G2I group showed a highly significant response to EAS1 PEDV variant. Conversely, at 7 dpi, recall antigen activation of CD4⁺ cells was highly significant in G1I group when stimulated with EAS1 PEDV variant. No activation of CD4⁺ cells was observed in G2I group when CBR1 was used as the recall antigen. However, PEDV-specific CD4⁺IFN-γ⁺ stimulation effect showed a short duration in both G1I and G2I groups, and the average of CD4⁺IFN-γ⁺ cells beginning to decline by 7 dpi. The delayed CD4⁺IFN-γ⁺ response was observed in G2I group when heterologous recall antigen was used. The response was observed at 7 dpi. However, following both homologous and heterologous reinfection, only the pigs that were initially exposed to the CBR1 variant (G2HomoRI and G2HeteRI) and subsequently recalled with heterologous variant showed an increment in the average of PEDV-specific CD4⁺IFN-γ⁺ PC stimulated by 7 dpri compared with negative control animals, and G1HomoRI and G1HeteRI groups (Fig. 1A). Meanwhile, recall antigen activation by CBR1, resulted in an increment in the average of PEDV-specific IFN-γ PC at 3 dpi in G1I group but not in G2I compared with negative controls. However, after homologous and heterologous reinfection, no changes on PEDV-specific CD4⁺IFN-γ⁺ cell population were observed compared with negative control animals (Fig. 1B).

G1I group exhibited a high CD8⁺IFN-γ⁺response when recalled with both homologous and heterologous PEDV variants. Recall antigen activation of CD8⁺ cells by EAS1 and CBR1 is shown in Fig. 2. At 3 dpi, G1I group exhibited a higher response than G2I group when EAS1 was used as recall antigen; however, the difference was not statistically significant. In contrast, when CBR1 was used as recall antigen, G1I group showed the highest and most significant difference at 3 dpi. Additionally, at 7 dpi, recall antigen activation by EAS1 on CD8⁺ cells demonstrated the highest significant difference in the G1I group compared with G2I and negative control groups. The response of CD8⁺IFN-γ⁺ PC following the reinfection was similar to that of the CD4⁺IFN-γ⁺ response. Only the group that was primarily exposed to CBR1variant (G2HomoRI and G2HeteRI) demonstrated an increment in the average of PEDV-specific CD8⁺IFN-γ⁺ cells compared with negative control animals, and G1HomoRI and G1HeteRI groups (Fig. 2A). Meanwhile, recall antigen stimulated by CBR1 variant resulted in an increment in PEDV-specific CD8⁺IFN-γ⁺ stimulatory effect at 3 dpi in G1I and 14 dpi in G2I compared with negative control group. After both homologous and heterologous reinfection, only G1I group reinfected with heterologous challenge (G1HeteRI), showed an increment in the percentage of PEDV-specific CD8⁺IFN-γ⁺ cell population at 7 dpri, although the difference was not statistically significant compared with negative control group (Fig. 2B).

[See PDF for image]

Fig. 1

Number of CD4⁺IFN-γ⁺ cells in mesenteric lymph nodes (MLN) of pigs infected with different PEDV variants. PEDV-specific CD4⁺IFN-γ⁺ stimulation effect of MNC isolated at 3, 7, 14 dpi from MLN of PEDV infected pigs, and homologous and heterologous PEDV reinfection at 21 dpi (7dpri) with EAS1 (A), classical variant PEDV recall antigen activation, and CBR1 (B), pandemic variant PEDV recall antigen activation. The red dash line demonstrated the baseline of PEDV in unexposed pigs (negative control at 0 dpi). G1I = EAS1-infected group; G2I = CBR1-infected group; G1HomoRI = EAS1-infected group following by EAS1 reinfection at 14 dpi; G1HeteRI = EAS1-infected group followed by CBR1 reinfection at 14 dpi; G2HomoRI = CBR1-infected group following by CBR1 reinfection at 14 dpi; G2HeteRI = CBR1-infected group followed by EAS1 reinfection at 14 dpi.

PEDV-specific CD4⁺CD8⁺IFN-γ⁺stimulatory effect was slightly lower in all groups when recalled with EAS1. In contrast, recall with CBR1 presented the highest response in G1I group, followed by G2I group at 3 dpi. Recall antigen activation of CD4⁺CD8⁺IFN-γ⁺ producing cells by 2 PEDV variants is shown in Fig. 3. Recall antigen activation by EAS1, G2I group exhibited a high number of CD4⁺CD8⁺IFN-γ⁺ producing cells at 3 dpi; however, no significant modulatory effect was observed at any time point (Fig. 3A). In contrast, recall activation by CBR1 showed a slight stimulatory effect on PEDV-specific CD4⁺CD8⁺IFN-γ⁺ cells at 3 and 14 dpi in the G1I group and at 3 dpi in the G2I group. In addition, pigs reinfected with the homologous EAS1 variant (G1HomoRI) and recalled with the CBR1 PEDV variant exhibited a slight increment in PEDV-specific CD4⁺CD8⁺IFN-γ⁺ stimulatory effect compared with other groups. (Fig. 3B).

G1HeteRI and G2HomoRI subgroups exhibited elevated number of PEDV-specific IgA antibody-secreting cells (ASC) when the CBR1 variant was used as antigen coating. The numbers of PEDV-specific IgA ASC are summarized in Table 1. Pigs in negative control group were not observed PEDV specific IgA ASC in any day points. Following the primary exposure, the number of PEDV-specific IgA ASC against either EAS1 or CBR1 antigen were undetectable at 3 and 7 dpi and turning positive at 14 dpi for both PEDV variants (G1I and G2I). There was not an obvious influence associated with the PEDV variant exposed; however, response seem to be relatively stronger in PEDV G1 variant. Based on the response against both antigens, the average of PEDV-specific IgA ASC observed on PEDV G1I was relatively higher compared with PEDV G2I. However, the responses in both PEDV G1I and G2I groups were moderately higher in CBR1 coated plates compared with EAS1 coated plates.

Following the homologous and heterologous reinfection either with EAS1 or CBR1 variant, the PEDV-specific IgA ASC responses were detected as early as at 7 dpri, suggesting an anamnestic response. Interestingly, the response followed the reinfection challenge was variant-dependent.

Discussion

To investigate the immune response induced by different PEDV variants, cell-mediated immunity (CMI) and humoral-mediated immunity (HMI) against different G1 and G2PEDV variants was evaluated. Although the protection in swine enteric viral infections is mostly due to a mucosal HMI response, CMI continues to play an important role on protection and recovery from enteric infections³².

In this study, the CMI and HMI were evaluated by the proliferation of PEDV-specific IFN-γ producing cells (IFN-γ PC) and PEDV-specific IgA antibody-secreting cells (ASC), following primary infection with either PEDV G1 or G2 variant. We found that both CD4⁺ and CD8⁺ PEDV-specific IFN-γ PC populations, representing T helper and T cytotoxic cells, were observed as early as 3 dpi in animals infected with PEDV G1. Despite early PEDV-specific IFN-γ activation, we found that the heterologous recall activation had a stronger response compared with homologous recall activation. This finding aligns with a previous study that reported elevated levels of double-positive T cells (CD4⁺CD8⁺) following re-infection with a virulent field PEDV strain²². The authors of this study evaluated if pigs previously exposed to PEDV could develop a sufficient immune response to protect against reinfection. In the experiment, pigs were initially exposed to the PEDV USA/Colorado/2013 strain at a concentration of 3.6 × 10⁴ TCID₅₀/mL. At 30 dpi, pigs were challenged with a virulent PEDV field strain. Peripheral blood mononuclear cells (PBMC) and MLN were collected at 4, 10, and 21 dpi, as well as 5 days after challenge, and subsequently analyzed by flow cytometry. The results showed a low frequency of CD4⁺, CD8⁺, and double-positive T cells in MLN, but a high frequency of T cells in blood at 4 dpi in the PEDV-exposed group compared to the control group. At 5 days post-challenge, the PEDV-exposed group showed a significantly higher frequency of CD4⁺CD8⁺ T cells in MLNs, with no significant difference observed in blood, suggesting that heterologous recall activation can stimulate a robust early cellular response.

Separately in our study, pigs exposed to PEDV G2 displayed a PEDV-specific IFN-γ PC stimulation effect in CD4⁺ and CD8⁺ populations only during the heterologous in vitro activation. Previous studies demonstrated that in vitro infection of PEDV CV777 (G1 prototype strain), modulates pro-inflammatory cytokines and type I IFN production of porcine small intestinal epithelial cells through a positive regulatory effect of toll-like receptor 2 (TLR-2), TLR3 and TLR9 ²³. In addition, interference in interferon-beta (IFN-β) modulation has also been observed in vitro between CV777 strain (G1 variant) and AJ1102 strain (G2 variant)^24,25. However, the activation pathway of pro-inflammatory cytokines and type I IFN production induced by PEDV G2 variants is still unknown. We speculate that the PEDV G2 variant might have a differential mechanism to evade host immune response, due to the differential timing in CMI regulation. Thus, PEDV G2 might have different strategies to evade or regulate the cellular immunity at mucosal level. However, the mechanism and signaling pathway responsible for this differential cellular response need further elucidation.

Previous studies demonstrate that PEDV attenuation induce a delayed and insufficient immune response compared with PEDV wild-type^26,27. Virus attenuation by serial passage in adapted cell lines may potentially decrease virus infectivity resulting in a low viral particle number produced upon infection, causing a lowered immune response. The PEDV strains used in this study were isolated from the clinical cases and belonged to PEDV G1 and G2 with no prior serial passage in cell lines. There is hence no evidence of attenuation of these strains from the field history. We therefore postulate that the observed differences in host immune response between PEDV G1 and G2 might be the result of a differential strategy in immune evasion.

In contrast to the early CMI (3 dpi) observed following PEDV G1 and G2 infection, PEDV-specific IgA ASC showed a delayed response in which was detected approximately at 14 dpi. Interestingly, anamnestic response was detected on both genogroups within 7 days after reinfection. PEDV-specific IgA ASC produced during homologous reinfection with PEDV G1 displayed in vitro reactivity with EASI1 (G1) antigen, and cross reactivity with CBR1 (G2) antigen. Conversely, homologous reinfection with PEDV G2 variant only showed reactivity with CBR1 (G2) antigen, not with EAS1 (G1). PEDV-specific IgA ASC induced after heterologous re-challenge showed reactivity against EAS1 (G1) antigen, and cross reactivity with CBR1 (G2) antigen. However, the total number of cells detected was slightly higher in CBR1 antigen plates. Our results demonstrated that the PEDV variant involved during reinfection drives the anamnestic response. Interestingly, previous studies suggest that the anamnestic response occurred only after prime exposure by attenuated strains and boosted by virulent strains. Other studies also demonstrated that this process is not affected by genetic differences in strains used to prime and boost²⁸. Viral attenuation of the priming strain might thus play a role in anamnestic response. We postulate that one reason for the anamnestic response not being detected during homologous reinfection could be the existence of pre-existing neutralizing antibody from previous priming. The total number of ASC after re-challenge with a virulent strain was significantly higher in animals that were primed with either a high or a low dose of PEDV-attenuated strains compared with those primed with PEDV-virulent strains. Thus, the HMI response obtained by PEDV-attenuated strains is not sufficient to protect a challenge against PEDV-virulent strain. However, reinfection with PEDV-virulent strains allows contact with immunocompetent cells resulting in a strong anamnestic response due to high viral loads produced in the intestine in-vivo. In contrast in animals primed with PEDV-virulent strains, preexisting antibodies might completely block reinfection, resulting in a lower anamnestic response. In our study, the anamnestic HMI responses were detected in both PEDV genogroups, suggesting that the immune response induced by each strain cannot completely neutralize reinfection either with homologous or heterologous strains. Additional factors that might affect this incomplete protection could be the amplitude of PEDV-specific humoral response. Antibody levels induced during this experiment could be insufficient and not being able to induce full protection. Partial or incomplete cross protection activity between these strains can also be a factor that needs to be further elucidated.

Our results suggested that even though PEDV G1 infection causes milder clinical disease compared with PEDV G2, the CMI response can occur earlier compared to PEDV G2 infection. We also observed that the timing of the HMI response between these two variants was similar, and the anamnestic response was driven by the variant involved during the reinfection. The results of this study present further evidences that support the hypothesis that PEDV G1 and G2 might have different evasion strategies. However, these mechanisms that affect or trigger differentially the host immune response needs to be further elucidated.

Materials and methods

Virus isolates and cells. Two PEDV isolates including EAS1 and CBR1, were used in the study. EAS1 and CBR1 are belonged to G1 and G2, respectively. These two PEDV isolates and their genetic characterization were described in a previous study²⁹. Both PEDV variants were isolated from 2- to 3-day-old piglets displaying acute watery diarrhea from swine farms in Thailand using the continuous Vero cell line (ATCC, CCL-81), as a previously described³⁰. The full-length genome sequencing data of these two PEDV isolates was deposited in GenBank with the following accession number; EAS1 (KR610991) and CBR1 (KR610993).

Experimental design. All animal procedures were conducted in accordance with the Care and Use of Laboratory Animals of the National Research Council of Thailand according to protocols reviewed and approved Faculty of Veterinary Science, Mahidol University-Institute Animal Care and Use Committee (FVS-MU-IACUC; animal use license number U1-01281-2558). The study is reported in accordance with the ARRIVE guidelines (https://arriveguidelines.org).

Forty-five, 3-week-old, castrated-male-pigs from a swine commercial farm with no history of PEDV were used for this study. PEDV-free status was confirmed by ELISA and PCR on serum and fecal swabs^19,31. Upon arrival, pigs were weighed and randomly allocated into 3 treatment groups of 15 pigs each based on weight stratification. Group 1 and group 2 were inoculated with PEDV, while group 3 served as negative control. Three pigs from group 3 were euthanized at 0 day post infection (dpi) to serve as baseline. Mesenteric lymph nodes (MLNs) were collected from each pig, and lymphocytes were separated and assayed for presence of interferon-gamma producing cells (IFN-γ PC). The number of IFN-γ PC from these negative pigs were used as a baseline reference. After one day of acclimatization, pigs in group 1 and group 2 were orally first inoculated with 10 mL of either 1 × 10³ TCID₅₀/mL of PEDV isolate EAS1 (G1-inoculation; (G1I)), or PEDV isolate CBR1 (G2-inoculation; (G2I)), respectively. In group 3, pigs were inoculated with 10 mL of cell culture supernatant as control group or mock (Table 2). Three pigs from G1I, G2I and control were euthanized at 3, 7 and 14 dpi. MLN were collected for mononuclear cells (MNC) isolation and were assayed for the presence of IFN-γ PC and IgA ASC. At 14 dpi, the 6 remaining pigs in G1I and G2I groups were divided into 2 subgroups of 3 pigs each and subjected for homologous and heterologous reinfection with either G1 or G2 variant. Three pigs from G1I group were orally reinoculated with 10 mL of EAS1 at 1 × 10³ TCID₅₀/mL (G1 homologous-reinfection (G1HomoRI)) and 3 pigs with 10 mL of PEDV isolate CBR1 at 1 × 10³ TCID₅₀/mL (G1 heterologous-reinfection (G1HeteRI)). Three pigs from G2I group were orally reinoculated with 10 mL of CBR1 at 1 × 10³ TCID₅₀/mL (G2 homologous-reinfection (G2HomoRI)) and 3 pigs with 10 mL of PEDV isolate EAS1 at 1 × 10³ TCID₅₀/mL (G2 heterologous-reinfection (G2HeteRI)). Three pigs in the control group were reinoculated with 10 mL of cell culture supernatant. Seven days post reinfection (dpri), all pigs were euthanized at 21 dpi for MLN collection.

Isolation of mononuclear cells (MNC) from mesenteric lymph nodes (MLN). MNC from MLN were isolated as described previously²⁸. Briefly, small sections of approximately 1 gram of MLN were grinded and pressed through stainless cell strainers (70 micron) (Corning^® cell strainer, Corning, NY, USA). Cell suspension was centrifuged, and MNCs were removed from the pellet by gradient centrifugation using Lymphosep^® (Biowest, France). Isolated MNCs were washed twice with 1X of phosphate buffer saline (PBS; pH 7.4) and resuspended in advance RPMI media (Gibco^®, CA, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco^®, CA, USA). Cell viability was confirmed by Trypan blue exclusion.

Phenotyping of IFN-γ producing cells. The numbers of IFN-γ PC together with lymphocyte surface marker, CD4 and CD8, were determined in MNC using flow cytometry analysis. Briefly, 200 µl of cell suspension containing 1 × 10⁶ MNC were stimulated in 96-well plates with 0.01 multiplicity of infection (MOI) of PEDV isolate either EAS1 or CBR1. Mock-infected Vero cell lysates suspension served as internal control for this assay. Homologous recall antigen of G1I and G2I groups was performed by stimulation with EAS1 and CBR1 strains respectively, while heterologous recall antigen was performed by stimulation of G1I with CBR1 and G2I with EAS1 strains, respectively. MNC were stimulated for 72 h at 37 ^oC and 5% C0₂ with BD Golgistop™ protein transport inhibitor (BD Biosciences, CA, USA), that was added according to the manufacturer’s instruction. Swine lymphocyte surface markers CD4 and CD8 were evaluated using 1:50 FITC-conjugated mouse anti-porcine CD4 (74-12-4, IgG2b) monoclonal antibody (mAb) (Southern Biotech, AL, USA) and 1:50 of SPRD-conjugated mAb mouse anti-porcine CD8α (76-2-11, IgG2a) (Southern Biotech, AL, USA) in FACS buffer. Antibodies were incubated in the dark for 30 min at 4^oC. Cells were then fixed and permeabilized by Reagent A of Leucoperm™ (AbD Serotech^®, UK) for 15 min in dark chamber at room temperature (RT). Biotinylated-conjugate mAb mouse anti-porcine IFN-γ (P2C11, IgG2a) (BD Pharmingen™, CA, USA) was diluted 1:100 in Reagent B of Leucoperm™ and used for staining for 45 min in a dark chamber at RT, followed by a 1:500 dilution of PE/Cy-7 conjugated streptavidin (Invitrogen™, MD, USA) in FACS buffer for 30 min at 4^oC. Cells stained with the isotype control antibody (biotinylated IgG2a antibody, BD Pharmingen™, CA, USA) were included to establish background cut-off. MNC subpopulations were sorted by flow cytometry and analysis was based on 100,000 events using a FC500 MPL cytometer (Beckman Coulter^®, CA, USA). The fluorescence minus one (FMO) staining controls were performed during the establishment and validation of assays. The results are presented as the percentage of the PEDV-specific stimulated IFN-γ PC from each infected group calculated as the percentage of IFN-γ⁺ PC from the EAS1 and CBR1 PEDV recall minus the percentage of IFN-γ⁺ PC from the mock suspension recall. Baseline of each cell phenotype was obtained after stimulation of MNC isolated from unexposed pigs.

PEDV-specific IgA antibody secreting cells by ELISPOT. The detection of PEDV-specific IgA ASC was performed by ELISPOT assay in accordance with previously described protocol²⁸. Briefly, confluent monolayer of Vero cell in 96 well plates were infected with 0.1 MOI of PEDV isolate either CBR1 or EAS1, and incubated for 12 h at 37ºC, 5% CO₂. MEM supplement with 5% FBS was used as control. PEDV and mock-inoculated plates were then fixed with 80% acetone and stored at -20 ^oC until used. Three different concentrations (5 × 10³, 5 × 10⁴ and 5 × 10⁵) of MNC obtained from each treatment group (G1I, G2I, G1HomoRI, G1HeteRI, G2HomoRI, G2HeteRI) were incubated in duplicate for 18 h at 37 ^oC in Vero cells previously infected with PEDV. A primary biotinylated-conjugate goat anti-porcine IgA polyclonal antibody (1:1000), was incubated with MNC overnight at 4 ^oC followed by a horseradish peroxidase-conjugated streptavidin (1:2500) for 2 h at RT. Five washes with 0.5% Tween-20 in PBS (PBST) were performed between steps. Spots were developed by addition of 3-amino-9-ethylcarbazole (AEC; Sigma Aldrich, MO, USA) and counted under an optic microscope. Results represent the mean (± SEM) of the total number of spots per treatment calculated from duplicated wells relative to 5 × 10⁵ MNCs.

Statistical analysis. Data were tested for normality by Kolmogorov-Smirnov. The significance between two treatment groups was determined by non-parametric statistical analysis by Mann-Whitney test. Significance was assessed at P < 0.05. Data analysis was performed using GraphPad Prism^® (Graphpad software Inc, CA, USA).

[See PDF for image]

Fig. 2

Number of CD8⁺IFN-γ⁺ cells in mesenteric lymph nodes (MLN) of pigs infected with different PEDV variants. PEDV-specific CD8⁺IFN-γ⁺ stimulation effect of MNC isolated from MLN at 3, 7, 14 dpi with PEDV infected pigs, and homologous and heterologous PEDV reinfection at 21 dpi (7dpri) with EAS1 (A), classical variant PEDV recall antigen activation and CBR1 (B), pandemic variant PEDV recall antigen activation. The red dash line demonstrated the baseline of PEDV unexposed pigs (negative controls at 0 dpi). G1I = EAS1-infected group; G2I = CBR1-infected group; G1HomoRI = EAS1-infected group followed by EAS1 reinfection at 14 dpi; G1HeteRI = EAS1-infected group followed by CBR1 reinfection at 14 dpi; G2HomoRI = CBR1-infected group followed by CBR1 reinfection at 14 dpi; G2HeteRI = CBR1-infected group followed by EAS1 reinfection at 14 dpi.

[See PDF for image]

Fig. 3

Number of CD4⁺CD8⁺IFN-γ⁺ cells in mesenteric lymph nodes (MLN) of pigs infected with different PEDV variants. PEDV-specific CD4⁺CD8⁺IFN-γ⁺ stimulation effect at 3, 7, 14 dpi on MNC isolated from MLN of PEDV infected pigs with homologous and heterologous PEDV reinfection at 21 dpi (7dpri) with EAS1 (A), classical variant PEDV recall antigen activation and CBR1 (B), pandemic variant PEDV recall antigen activation. The red line demonstrated the baseline of PEDV unexposed pigs (negative control at 0 dpi). G1I = EAS1-infected group; G2I = CBR1-infected group; G1HomoRI = EAS1-infected group following by EAS1 reinfection at 14 dpi; G1HeteRI = EAS1-infected group followed by CBR1 reinfection at 14 dpi; G2HomoRI = CBR1-infected group followed by CBR1 reinfection at 14 dpi; G2HeteRI = CBR1-infected group followed by EAS1 reinfection at 14 dpi.

Table 1. Mean (± SEM) of PEDV specific IgA ASC detected in mesenteric lymph node (MLN) of oral inoculated pigs in Immuno plates coated with PEDV G1 (EAS1) variant and PEDV G2 (CBR1) variant.

Treatment group/ days post infection (dpi)	MLN IgA ASC EAS1 coated plate	MLN IgA ASC CBR1 coated plate
G1 3 dpi 7 dpi 14 dpi 21 dpi/ 7 dpri (G1HomoRI) 21 dpi/ 7 dpri (G1HeteRI)	0.00 0.00 2.00 ± 1.00 3.33 ± 3.33 1.33 ± 1.33	0.00 0.00 0.33 ± 0.33 2.67 ± 2.19 18.33 ± 17.34
G2 3 dpi 7 dpi 14 dpi 21 dpi/ 7 dpri (G2HomoRI) 21 dpi/ 7 dpri (G2HeteRI)	0.00 0.00 3.67 ± 2.73 0.00 3.00 ± 2.52	0.00 0.00 1.00 ± 1.00 10.00 ± 4.51 8.00 ± 8.00
Negative group 0 dpi	0.00	0.00
3 dpi	0.00	0.00
7 dpi	0.00	0.00
14 dpi	0.00	0.00
21 dpi/ 7 dpri	0.00	0.00

Data are means ± SEM.

Table 2. Three-week-old pigs were separated to oral inoculation with PEDV G1 (EAS1) variant and PEDV G2 (CBR1) variant according to the group, G1I and G2I, at 0 dpi. Afterwards, remaining pigs in each group were separated to oral reinoculation with homologous (G1HomoRI and G2HomoRI) and heterologus (G1HeteRI and G2HeteRI) PEDV variant at 14 dpi.

Group	No. pigs	1st inoc. variant (1 dpi)	2nd inoc. variant (14 dpi)	No. pigs necropsied
				0 dpi	3 dpi	7 dpi	14 dpi	21 dpi (or 7 dpri)
G1	15	EAS1	EAS1 (G1HomoRI)		3	3	3	3
			CBR1 (G1HeteRI)					3
G2	15	CBR1	CBR1(G2HomoRI)		3	3	3	3
			EAS1 (G2HeteRI)					3
G3	15	Mock (cell culture supernatant)	Mock (cell culture supernatant)	3	3	3	3	3

Acknowledgements

This research was funded by the Thailand Science Research and Innovation Fund, Chulalongkorn University, and Swine viral evolution and vaccine development Research Unit (SVEVR), Chulalongkorn University. Additional funding was provided by the Agricultural Research Development Agency (ARDA, public organization), the National Research Council of Thailand (NRCT, grant number N41A640366).

Author contributions

Conceptualization D.N., P.P., H.L., K.K. and A.T.; methodology D.N. and G.T.; software G.T.; validation P.J., K.S. and P.J.; formal analysis P.J.; investigation, G.T.; writing-original draft preparation G.T., K.S., P.J. and D.N.; writing-review and editing D.N., H.L. and A.T.; visualization D.N. and K.K.; project administration D.N.; funding acquisition D.N. All authors have read and agreed to the published version of the manuscript.

Data availability

The full-length genome sequencing data of these two PEDV isolates have been deposited in GenBank under the accession numbers EAS1 (KR610991) and CBR1 (KR610993). The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Wood, E. N. An apparently new syndrome of Porcine epidemic diarrhoea. Vet. Rec; 1977; 100, pp. 243-244.1:STN:280:DyaE2s3jsleksQ%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/888300]

2. Pensaert, MB; de Bouck, P. A new coronavirus-like particle associated with diarrhea in swine. Arch. Virol.; 1978; 58, pp. 243-247.1:STN:280:DyaE1M%2FotFKqtg%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/83132][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7086830]

3. Martelli, P et al. Epidemic of diarrhoea caused by Porcine epidemic diarrhoea virus in Italy. Vet. Rec; 2008; 162, pp. 307-310.1:STN:280:DC%2BD1c7mvFOguw%3D%3D [DOI: https://dx.doi.org/10.1136/vr.162.10.307] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/18326842]

4. Chae, C et al. Prevalence of Porcine epidemic diarrhoea virus and transmissible gastroenteritis virus infection in Korean pigs. Vet. Rec; 2000; 147, pp. 606-608.1:STN:280:DC%2BD3M3hsFWhtw%3D%3D [DOI: https://dx.doi.org/10.1136/vr.147.21.606] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11110482]

5. Pritchard, GC; Paton, DJ; Wibberley, G; Ibata, G. Transmissible gastroenteritis and Porcine epidemic diarrhoea in Britain. Vet. Rec; 1999; 144, pp. 616-618.1:STN:280:DyaK1MzhvFaktw%3D%3D [DOI: https://dx.doi.org/10.1136/vr.144.22.616] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/10390802]

6. Pijpers, A; van Nieuwstadt, AP; Terpstra, C; Verheijden, JH. Porcine epidemic diarrhoea virus as a cause of persistent diarrhoea in a herd of breeding and finishing pigs. Vet. Rec; 1993; 132, pp. 129-131.1:STN:280:DyaK3s7ptVCmsQ%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/8383370]

7. Cima, G; PED virus reinfecting, U.S. herds. Virus estimated to have killed 7 million-plus pigs. J. Am. Vet. Med. Assoc.; 2014; 245, pp. 166-167. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25115019]

8. Puranaveja, S et al. Chinese-like strain of Porcine epidemic diarrhea virus, Thailand. Emerg. Infect. Dis.; 2009; 15, pp. 1112-1115. [DOI: https://dx.doi.org/10.3201/eid1507.081256] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19624933][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2744260]

9. Sun, RQ et al. Outbreak of Porcine epidemic diarrhea in suckling piglets, China. Emerg. Infect. Dis.; 2012; 18, pp. 161-163. [DOI: https://dx.doi.org/10.3201/eid1801.111259] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22261231][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3381683]

10. Theuns, S. et al. Complete genome sequence of a porcine epidemic diarrhea virus from a novel outbreak in belgium, january 2015. Genome announcements 3 (2015). https://doi.org/10.1128/genomeA.00506-15

11. Ojkic, D et al. The first case of Porcine epidemic diarrhea in Canada. Can. Vet. J.; 2015; 56, pp. 149-152. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25694663][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4298265]

12. Lin, CN et al. US-like strain of Porcine epidemic diarrhea virus outbreaks in taiwan, 2013–2014. J. Vet. Med. Sci.; 2014; 76, pp. 1297-1299. [DOI: https://dx.doi.org/10.1292/jvms.14-0098] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24898162][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4197162]

13. Lee, S; Lee, C. Outbreak-related Porcine epidemic diarrhea virus strains similar to US strains, South korea, 2013. Emerg. Infect. Dis.; 2014; 20, pp. 1223-1226. [DOI: https://dx.doi.org/10.3201/eid2007.140294] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24960370][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4073847]

14. Vui, DT et al. Complete genome characterization of Porcine epidemic diarrhea virus in Vietnam. Arch. Virol.; 2015; 160, pp. 1931-1938.1:CAS:528:DC%2BC2MXptlOrs70%3D [DOI: https://dx.doi.org/10.1007/s00705-015-2463-6] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26026958]

15. Chen, Q et al. Isolation and characterization of Porcine epidemic diarrhea viruses associated with the 2013 disease outbreak among swine in the united States. J. Clin. Microbiol.; 2014; 52, pp. 234-243.1:CAS:528:DC%2BC2MXjtlarsr8%3D [DOI: https://dx.doi.org/10.1128/JCM.02820-13] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24197882][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3911415]

16. Hanke, D et al. Comparison of Porcine epidemic diarrhea viruses from Germany and the united states, 2014. Emerg. Infect. Dis.; 2015; 21, pp. 493-496.1:CAS:528:DC%2BC28XitVOisbnL [DOI: https://dx.doi.org/10.3201/eid2103.141165] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25695311][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4344272]

17. Masuda, T et al. New Porcine epidemic diarrhoea virus variant with a large deletion in the Spike gene identified in domestic pigs. Arch. Virol.; 2015; 160, pp. 2565-2568.1:CAS:528:DC%2BC2MXhtFKrsr3I [DOI: https://dx.doi.org/10.1007/s00705-015-2522-z] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26162305][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7087250]

18. Li, W et al. New variants of Porcine epidemic diarrhea virus, china, 2011. Emerg. Infect. Dis.; 2012; 18, pp. 1350-1353. [DOI: https://dx.doi.org/10.3201/eid1808.120002] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22840964][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3414035]

19. Temeeyasen, G et al. Genetic diversity of ORF3 and Spike genes of Porcine epidemic diarrhea virus in Thailand. Infect. Genet. Evol.; 2014; 21, pp. 205-213.1:CAS:528:DC%2BC2cXisFertb4%3D [DOI: https://dx.doi.org/10.1016/j.meegid.2013.11.001] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24291225]

20. Cheun-Arom, T et al. Full-length genome analysis of two genetically distinct variants of Porcine epidemic diarrhea virus in Thailand. Infect. Genet. Evol.; 2016; 44, pp. 114-121.1:CAS:528:DC%2BC28XhtFeqsrjP [DOI: https://dx.doi.org/10.1016/j.meegid.2016.06.046] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27353184][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7185804]

21. Chen, Q et al. Pathogenesis comparison between the united States Porcine epidemic diarrhoea virus prototype and S-INDEL-variant strains in conventional neonatal piglets. J. Gen. Virol.; 2016; 97, pp. 1107-1121.1:CAS:528:DC%2BC28XhvVGks7vI [DOI: https://dx.doi.org/10.1099/jgv.0.000419] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26841768]

22. Krishna, VD et al. Immune responses to Porcine epidemic diarrhea virus (PEDV) in swine and protection against subsequent infection. PLoS One; 2020; 15, e0231723.1:CAS:528:DC%2BB3cXosVGjt7s%3D [DOI: https://dx.doi.org/10.1371/journal.pone.0231723] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32343704][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7188253]

23. Cao, L et al. Porcine epidemic diarrhea virus infection induces NF-κB activation through the TLR2, TLR3 and TLR9 pathways in Porcine intestinal epithelial cells. J. Gen. Virol.; 2015; 96, pp. 1757-1767.1:CAS:528:DC%2BC2MXhsFaiurvP [DOI: https://dx.doi.org/10.1099/vir.0.000133] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25814121]

24. Ding, Z et al. Porcine epidemic diarrhea virus nucleocapsid protein antagonizes Beta interferon production by sequestering the interaction between IRF3 and TBK1. J. Virol.; 2014; 88, pp. 8936-8945. [DOI: https://dx.doi.org/10.1128/jvi.00700-14] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24872591][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4136253]

25. Cao, L et al. Porcine epidemic diarrhea virus inhibits dsRNA-induced interferon-beta production in Porcine intestinal epithelial cells by Blockade of the RIG-I-mediated pathway. Virol. J.; 2015; 12, 127.1:CAS:528:DC%2BC28XivVamsr4%3D [DOI: https://dx.doi.org/10.1186/s12985-015-0345-x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26283628][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4539884]

26. de Arriba, ML; Carvajal, A; Pozo, J; Rubio, P. Lymphoproliferative responses and protection in conventional piglets inoculated orally with virulent or attenuated Porcine epidemic diarrhoea virus. J. Virol. Methods; 2002; 105, pp. 37-47. [DOI: https://dx.doi.org/10.1016/s0166-0934(02)00063-0] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/12176140][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7119748]

27. Ward, LA; Yuan, L; Rosen, BI; To, TL; Saif, LJ. Development of mucosal and systemic lymphoproliferative responses and protective immunity to human group A rotaviruses in a gnotobiotic pig model. Clin. Diagn. Lab. Immunol.; 1996; 3, pp. 342-350.1:STN:280:DyaK28zjs1Sqtg%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/8705681][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC170344]

28. de Arriba, ML; Carvajal, A; Pozo, J; Rubio, P. Mucosal and systemic isotype-specific antibody responses and protection in conventional pigs exposed to virulent or attenuated Porcine epidemic diarrhoea virus. Vet. Immunol. Immunopathol.; 2002; 85, pp. 85-97. [DOI: https://dx.doi.org/10.1016/s0165-2427(01)00417-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11867170]

29. Cheun-Arom, T. et al. Complete genome sequences of two genetically distinct variants of Porcine epidemic diarrhea virus in the Eastern region of Thailand. Genome Announcements. 3https://doi.org/10.1128/genomeA.00634-15 (2015).

30. Hofmann, M; Wyler, R. Propagation of the virus of Porcine epidemic diarrhea in cell culture. J. Clin. Microbiol.; 1988; 26, pp. 2235-2239.1:STN:280:DyaL1M7nsFeltw%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/2853174][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC266866]

31. Srijangwad, A et al. Immune response of gilts to single and double infection with Porcine epidemic diarrhea virus. Arch. Virol.; 2017; 162, pp. 2029-2034.1:CAS:528:DC%2BC2sXjvVyisr0%3D [DOI: https://dx.doi.org/10.1007/s00705-017-3307-3] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28271162]