Background

Porcine epidemic diarrhea (PED), characterized by vomiting, watery diarrhea, and dehydration, is a highly contagious swine enteric disease [1]. Pigs of all ages suffer from PED, while neonatal piglets are susceptible to PED with up to 100% mortality [2, 3]. PED was first reported in England in 1971 [4], and spread to other European and Asian countries [5]. Since 2010, PED re-emerges and becomes prevalent in China [6, 7]. In 2013, PED broke out in the USA and led to seven million deaths of pigs, corresponding to almost 10% of the national swine population [8, 9]. To date, PED is epidemic in numerous pig-raising countries and causes enormous economic losses to the global swine industry [10].

Vaccination remains as an effective strategy to prevent and control PED [10, 11]. However, due to their immature immune systems, neonatal suckling piglets acquire passive lactogenic immunity against infection by PED causative agent, PED virus (PEDV) [12]. Maternal antibodies in colostrum and milk play a critical role in lactogenic immunity. As immunoglobulin A (IgA) antibodies are dominant and persistent in milk throughout lactation [13, 14], their appropriate measurement is necessary to evaluate vaccination efficacies and neonate immune status against PEDV.

PEDV is an enveloped, single-stranded positive-sense RNA virus belonging to the order Nidovirales, family Coronaviridae, genus Alphacoronavirus. The PEDV genome is approximately 28 kb in length, and encodes four structural proteins, one accessory protein and 16 nonstructural proteins [15]. Among them, PEDV spike (S) glycoprotein is a class I virus fusion protein on virion surface [16], and can be further divided into the S1 subunit involved in receptor recognition and the S2 subunit mediating membrane fusion [17]. In addition, the S protein contains multiple antigenic epitopes, which induce immune responses and neutralizing antibodies [18, 19].

In order to establish a reliable diagnostic tool for PEDV IgA antibodies in milk, we firstly prepared monoclonal antibodies (mAbs) against recombinant PEDV S protein in the current study. Subsequently, we developed an enzyme-linked immunosorbent assay (ELISA) based on PEDV antigen capture by a specific anti-S mAb. The ELISA showed good sensitivity and repeatability in detecting PEDV IgA antibody positive and negative milk samples, and a high coincidence rate with a commercial ELISA kit for detecting clinical milk samples.

Results

Preparation of anti-PEDV mAbs

To prepare the mAbs specific for PEDV, we immunized mice with PEDV and screened via immunoperoxidase monolayer assay (IPMA). Positive hybridoma cell lines were subcloned, and four mAbs named as mAb9, mAb10, mAb17 and mAb18 were chosen for subsequent experiments. Figure 1 showed specific binding of these four mAbs to the cytopathic effect (CPE) regions in the PEDV-infected porcine intestine epithelial IPEC-J2 cells.

Fig. 1
figure 1

Screen of anti-PEDV mAbs by IPMA. The anti-PEDV mAbs strongly bound to the CPE regions in the PEDV-infected IPEC-J2 cells (brown). Representative images were shown for the mAbs 9, 10, 17 and 18, and PEDV positive (P) and negative (N) sera. Scale bars, 200 μm

Full size image

Identification and characterization of anti-S mAbs

Next, we determined whether the chosen anti-PEDV mAbs recognized PEDV S protein. Recombinant PEDV S1 subunit was prepared and obtained with a high purity (~ 110 kDa, Fig. 2A). As shown in Fig. 2B, the purified PEDV S1 subunit was specially bound by all the four mAbs in Western blot, while mock control or recombinant S protein from another Alphacoronavirus transmissible gastroenteritis virus (TGEV) wasn’t, indicating that these mAbs were capable of recognizing PEDV S protein in a specific manner.

Fig. 2
figure 2

Identification of anti-S mAbs by Western blot. A The recombinant PEDV S1 subunit was purified by HisTrap excel prepacked column and HiTrap Q HP prepacked column, and then analyzed by SDS-PAGE. B The purified PEDV S1 subunit was applied to SDS-PAGE, transferred to PVDF membranes, and detected by the mAbs 9, 10, 17 and 18. In parallel, mock control and recombinant TGEV S protein were applied. The results were visualized with HRP-conjugated goat anti-mouse IgG and Solarbio ECL plus reagent. The original gel and blots are presented in Additional file 2

Full size image

To further demonstrate their recognition capabilities, the PEDV-infected African green monkey kidney epithelial Vero cells were monitored by indirect immunofluorescence assay (IFA) using the anti-S mAbs. Figure 3 showed that the anti-S mAbs strongly bound to the CPE regions in the PEDV-infected Vero cells whereas not to the uninfected ones. These results revealed that the mAbs specially recognized PEDV S protein and virions during infection.

Fig. 3
figure 3

IFA analyses of PEDV-infected Vero cells recognized by the anti-S mAbs. The Vero cells with the 80% confluence were inoculated with PEDV until exhibiting obvious CPEs. After fixed and blocked, the cells were incubated with the anti-S mAbs and Alexa Fluor 488-donkey anti-mouse IgG. The cell nuclei were stained using DAPI and the cells were visualized by Leica fluorescence microscopy. Representative images were shown. Scale bars, 200 μm

Full size image

After characterization, the subtype of mAb9 heavy chain was identified as IgG2b, that of mAb10 and mAb18 was IgG2a, and that of mAb17 was IgG1 for further use.

Selection of the mAb9 for establishing ELISA

The anti-S mAbs were purified from ascites, and Fig. 4A showed that heavy (~ 50 kDa) and light chains (~ 25 kDa) were primarily visualized for all the four mAbs on sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), indicating their high purities.

Fig. 4
figure 4

Selection of the mAb9 for ELISA development. A The purified anti-S mAbs were analyzed by SDS-PAGE. The original gel is presented in Additional file 2. B Selection of the optimal anti-S mAb for developing ELISA. Each purified anti-S mAb was two-fold serially diluted as 8, 4, 2, 1, 0.5, 0.25, 0.125 and 0.0625 μg/mL, and coated on 96-well ELISA plates. Then, the plates were blocked and incubated with inactivated cell-cultivated PEDV supernatants (1 mg/mL). The plates were incubated with the rabbit anti-PEDV pAbs and HRP-conjugated goat anti-rabbit IgG, and reacted with TMB substrate solution. After stopped, OD450 values were measured on the BMG-Labtech microplate reader. The error bars indicate SD values

Full size image

To select the optimal one for establishing ELISA, each purified anti-S mAb was diluted, coated and tested for PEDV antigen capture. PEDV antigen capture by the mAb9 was shown with the highest value of optical density at 450 nm (OD450) at each coating concentration (Fig. 4B and Table S1) and therefore chosen to establish ELISA.

Development of ELISA

ELISA was established based on PEDV antigen capture by the mAb9 and optimized by checkerboard titration. As shown in Table S2, the optimal concentrations of the mAb9 for coating and inactivated cell-cultivated PEDV supernatants for viral antigen capture are 0.5 μg/mL and 0.5 mg/mL, respectively. The optimal coating condition of the mAb9 was at 4 °C for 12 h (Table S3). The optimal blocking solution was phosphate buffered saline (PBS) containing 5% bovine serum albumin (BSA), and the optimal blocking condition was at 37 °C for 1.5 h (Table S4). The optimal dilution and incubation time of milk samples were 1:40 and at 37 °C for 30 min (Tables S5 and S6). The optimal dilution and incubation time of horseradish peroxidase (HRP)-conjugated goat anti-pig IgA were 1:20,000 and at 37 °C for 30 min (Tables S7 and S8). The optimal reaction condition of 3, 3’, 5, 5’-tetramethylbenzidine (TMB) substrate solution was in dark at room temperature (RT) for 5 min (Table S9).

The cut-off value of the developed ELISA

Twenty PEDV IgA antibody negative milk samples were measured using the developed ELISA (Table S10). The mean OD450 value of these samples was 0.189 and the standard deviation (SD) value was 0.051. Based on the data, the cut-off value of the developed ELISA was 0.342, meaning that the milk samples with the OD450 values ≥ 0.342 were considered PEDV IgA antibody positive.

The sensitivity of the developed ELISA

The sensitivity of the developed ELISA was determined using the diluted milk samples. As shown in Fig. 5 and Table S11, the developed ELISA could detect 1:1280 or even lower diluted PEDV IgA antibody positive milk samples.

Fig. 5
figure 5

Determination of the sensitivity of the developed ELISA. Four PEDV IgA antibody positive and one negative milk samples were two-fold serially diluted as 1:40, 1:80, 1:160, 1:320, 1:640, 1:1280, 1:2560. The sensitivity of the developed ELISA was determined using the diluted milk samples. The error bars indicate SD values

Full size image

The repeatability of the developed ELISA

The repeatability of the developed ELISA was evaluated with five PEDV IgA antibody positive and one negative milk samples in triplicate on the same plate at the same time and on different plates at different times (Table S12). All the coefficient of variation (CV) values were between 3.60%-8.10%, which were within the allowable range of repeatability (< 10%) [20].

Clinical milk sample detection by the developed ELISA

In order to determine whether the developed ELISA is applicable to PEDV IgA antibody detection in clinical milk samples, a total of 92 unidentified clinical milk samples were simultaneously tested with the developed ELISA and a commercial ELISA kit (Anheal, Harbin, China). Table 1 showed that the number of the positive samples were 59 detected by the developed ELISA and 56 by the commercial ELISA kit, respectively. The sensitivity of the test was 98.21% (55/56) and the specificity of the test was 88.89% (32/36). Consequently, the developed ELISA highly agreed with the commercial Anheal ELISA kit for detecting PEDV IgA antibodies in clinical milk samples.

Table 1 Clinical milk sample detection by the developed ELISA and the commercial Anheal ELISA kit
Full size table

Discussion

PED has emerged or re-emerged as an economically critical enteric disease in pig farming worldwide [21]. The emergence of highly virulent PEDV variants with high morbidity and mortality in neonatal piglets brings about tremendous challenges to prevent and control PED in recent years [9, 22]. Vaccination is still an efficient manner to restrain PED outbreaks. Ideal vaccines for PEDV should be able to induce adequate maternal immunity in immunized sows and passively protect neonatal suckling piglets via colostrum and milk [12].

Among maternal antibodies, IgA antibodies are more resistant to proteolytic digestion in the gastrointestinal tract and have a higher neutralizing ability than IgG and IgM antibodies [23,24,25]. Current studies have shown that protection of neonatal suckling piglets by maternal antibodies against PEDV infection is mostly attributed to IgA antibodies in milk [13, 26]. As a consequence, PEDV IgA antibody detection in milk is imperative for evaluating vaccine efficacies and neonate immune status.

Although several ELISAs have been established for IgA antibody detection against PEDV in sera [20, 27,28,29], there are limited ELISAs for IgA antibody detection in colostrum or milk [30,31,32]. In detail, recombinant or truncated S proteins expressed in eukaryotic/prokaryotic system, nucleocapsid (N) protein expressed in prokaryotic system, or whole PEDV proteins in cell-cultivated virus supernatants were utilized as coating viral antigens in these reports [30,31,32]. PEDV N protein is inappropriate as coated viral antigen for IgA antibody detection because it shares highly conserved sequences with other Alphacoronaviruses and contains no neutralizing epitopes [33, 34], which may show false results. Whole PEDV proteins coated as viral antigens may also leads to false results due to nonspecific background [32]. In contrast, PEDV S protein is considered as a suitable viral antigen for IgA antibody detection [35]. Consequently, preparation of PEDV S protein with native conformation and high immunoreactivity is of great importance to reduce false results. On the one hand, prokaryotic expression system can’t render proper formation of disulfide bonds, glycosylation or even correct folding to such a membrane protein as PEDV S protein [36]. On the other hand, truncated S protein may not have a comparable immunoreactivity to the entire one [37].

In the current study, we prepared specific mAbs against PEDV S protein, and captured PEDV antigens in inactivated cell-cultivated virus supernatants by the optimal anti-S mAb9. We speculated that the captured PEDV antigens contained the whole virions and S proteins, which were specially and strongly bound by the mAb9 as indicated in Figs. 1, 2, 3 and 4, and therefore probably had a high immunoreactivity. We subsequently developed an ELISA based on captured PEDV antigens by the anti-S mAb. The developed ELISA showed high sensitivity and repeatability in detecting PEDV IgA antibody positive and negative milk samples. The developed ELISA also showed a high coincidence rate with the commercial Anheal ELISA kit for PEDV IgA antibody detection in clinical milk samples. Interestingly, the developed ELISA detected more positive samples than the commercial ELISA kit did. This discrepancy will be validated with other approaches in the future. In fact, we have further evaluated the stability of the developed ELISA and it showed decent detection performance within at least six months (data not shown).

Indeed, more milk samples are required to confirm the clinical performance of the developed ELISA, including the cut-off value and sample to positive ratio. In the meantime, reference IgA antibody positive and negative milk samples are unavailable to us, and as a result, we haven’t conducted comparative assays in the current study. In addition, unlike serum samples, IgA antibody positive milk samples for other swine viruses were difficult to collect, and therefore we haven’t determined the specificity and cross-reactivity of the developed ELISA. However, considering the high specificity and affinity of the anti-S antibody as well as the high immunoreactivity of the captured PEDV antigens, we can anticipate the superiority of the developed ELISA and will addressed this issue in the future. Moreover, the relationships amongst protection efficacy of suckling piglets, neutralizing ability of IgA antibodies in milk and IgA antibody level in milk detected by the developed ELISA should be explored in the future. It’s worth noting that the utilized anti-S mAb was non-neutralizing one (data not shown), and would neither shield the neutralizing epitopes of the captured PEDV antigens nor interfere with neutralization evaluation by IgA antibodies in milk.

Conclusion

Taken together, an ELISA was developed based on viral antigen capture by a specific anti-S mAb for detecting PEDV IgA antibodies in milk. The developed ELISA showed a high diagnostic sensitivity and repeatability, which provided a potent tool for PEDV surveillance and vaccination evaluation.

Methods

Mice, rabbits, cells, and virus

Six to eight-week-old and 16-week-old specific pathogen free female BALB/c mice were purchased from Henan Laboratory Animal Center (Zhengzhou, China). The experimental procedures for mice were authorized and supervised by the Ethical and Animal Welfare Committee of Key Laboratory of Animal Immunology of the Ministry of Agriculture of China (SYXK2021-0003).

Eight-week-old specific pathogen free female New Zealand white rabbits were purchased from Henan Chunying Biotechnology Co., Ltd (Zhengzhou, China). The experimental procedures for rabbits were authorized and supervised by the Ethical and Animal Welfare Committee of Key Laboratory of Animal Immunology of the Ministry of Agriculture of China (LLSC410113).

Vero cell line (CCL-81; kindly provided by Professor Zhanyong Wei of Henan Agricultural University, China, and kept in our laboratory) was cultivated in Dulbecco’s modified Eagle medium (Solarbio, Beijing, China) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco, Grand Island, USA) and antibiotics (100 U/mL penicillin, 100 μg/mL streptomycin; Solarbio) at 37 °C with 5% CO2. IPEC-J2 cell line (kindly provided by Professor Zhanyong Wei of Henan Agricultural University, China), SP2/0 myeloma cells (kept in our laboratory) and hybridoma cells (generated in the current study) were maintained in Roswell Park Memorial Institute-1640 medium (Solarbio) with 10% heat-inactivated FBS and antibiotics at 37 °C with 5% CO2. The Drosophila Schneider 2 (S2) cells stably expressing recombinant PEDV S1 subunit (residues 21–793, the numbering is according to GenBank accession number: KY496315.1) were kept in our laboratory and cultured in Schneider’s insect medium (Gibco) supplemented with 10% heat-inactivated FBS at 28 °C [38].

The prevalent PEDV strain CH/hubei/2016 (GenBank accession number: KY496315.1) was kept in our laboratory and used in the current study [39]. PEDV was propagated in Vero cells according to previous studies [40, 41]. When obvious CPEs were observed with characteristics of cell–cell fusion and syncytium formation at 18 h post infection, the cells were frozen and thawed three times, and then centrifuged at 9600 g at 4 °C for 10 min. The supernatant was filtrated through 0.22 μm filter membranes (Millipore-Merk, Darmstadt, Germany), and stored at -80 °C for further use or inactivated with formaldehyde for viral antigen capture.

Milk samples

Ten PEDV IgA antibody positive milk samples and 23 PEDV IgA antibody negative milk samples determined by the commercial Anheal ELISA kit (Harbin, China) along with 92 unidentified clinical milk samples were collected from 125 sows by our laboratory. All samples were centrifuged at 6200 g at 4 °C for 5 min. After the lipid layer removed, the middle layer was collected and stored at -20 °C for PEDV IgA antibody detection.

Preparation of mAbs targeting PEDV

MAbs targeting PEDV were prepared according to our previous study [39]. Briefly, PEDV CH/hubei/2016 virions were purified by tangential flow system and gel filtration chromatography, and utilized as immunogen to immunize six to eight-week-old female BALB/c mice. The spleen cells of the immunized mouse with the highest serum antibody titers were fused with SP2/0 myeloma cells. Supernatants of hybridoma cell cultures were screened by IPMA, and the positive ones were subcloned to obtain the hybridoma cell lines stably secreting anti-PEDV mAbs.

IPMA screen of anti-PEDV mAbs

PEDV was inoculated in the IPEC-J2 cells with 80% cell confluence on 96-well plates (Corning, Corning, USA). When obvious CPEs appeared, the cells were washed three times with PBS and then fixed with ice-cold methanol containing 3% H2O2 at -20 °C for 10 min. After fixation, the cells were washed three times with PBS containing 0.05% Tween-20 (PBST) and blocked with 5% skim milk at 37 °C for 1 h. Following washed with PBST, the cells were incubated with the supernatants of hybridoma cell cultures at 37 °C for 30 min. The cells were washed with PBST and added with 1:500 diluted HRP-conjugated goat anti-mouse IgG (Abbkine, Wuhan, China) and stained with 3-amino-9-ethylcarbazole solution (Solarbio) at RT for 5–10 min. The stained cells were examined by light microscopy (Leica, Weztlar, Germany). In parallel, the cells were incubated with PEDV positive and negative sera stored in our laboratory as positive and negative controls, respectively [38, 39].

Preparation of recombinant PEDV S1 subunit

Recombinant PEDV S1 subunit was prepared as previously described by us [38]. Briefly, recombinant PEDV S1 subunit was expressed by the Drosophila S2 cells in Sf-900 II serum-free medium (Invitrogen, Carlsbad, USA), and purified by HisTrap excel prepacked column (GE Healthcare, Fairfield, USA) and HiTrap Q HP prepacked column (GE Healthcare). The purified S1 subunit was applied to 15% SDS-PAGE for analysis.

Identification of anti-S mAbs by Western blot

The purified recombinant PEDV S1 subunit was subjected to SDS-PAGE and then transferred to polyvinylidene fluoride membranes (PVDFs; Millipore-Merk). The target protein was detected with the supernatants of positive hybridoma cell cultures, HRP-conjugated goat anti-mouse IgG and enhanced chemiluminescence (ECL) plus reagent (NCM Biotech, Suzhou, China). Mock control and recombinant TGEV S protein (kept in our laboratory) were subjected in parallel to assess the specificity of the mAbs.

IFA analyses of anti-S mAbs

When the confluence reached 80%, Vero cells were inoculated with PEDV and then cultivated at 37 °C until exhibiting obvious CPEs. The cells were fixed with ice-cold methanol and incubated with 3% BSA at 37 °C for 1 h. After washed with PBS, the cells were incubated with the supernatants of positive hybridoma cell cultures and Alexa Fluor 488-donkey anti-mouse IgG (Invitrogen). The cell nuclei were stained using 4’, 6-diamidino-2-phenylindole (DAPI; Solarbio) and then the cells were visualized by fluorescence microscopy (Leica).

Subtype identification of anti-S mAbs

Subtype identification of anti-S mAbs was conducted according to the manufacturer’s instructions (Proteintech, Wuhan, China).

Purification of anti-S mAbs

Ascites were prepared via in vivo induction in 16-week-old female BALB/c mice according to conventional procedures. Anti-S mAbs were purified from ascites through (NH4)2SO4 precipitation and Protein G Beads (Thermo Fisher Scientific, Waltham, USA). The purity of anti-S mAbs was analyzed by SDS-PAGE.

Generation of polyclonal antibodies (pAbs) targeting PEDV

The purified PEDV CH/hubei/2016 virions as described above were utilized as immunogen to immunize eight-week-old New Zealand white rabbits, and preparation of pAbs targeting PEDV was performed according to the procedures in our previous study [42]. A large amount of high-titer pAbs was obtained by heart blood collection method and purified by (NH4)2SO4 precipitation. The generated pAbs were identified using the purified recombinant PEDV S1 subunit by Western blot and further characterized in the IgG class (data not shown).

Selection of the optimal anti-S mAb for establishing ELISA

The four purified anti-S mAbs were examined to select the optimal one for viral antigen capture. In detail, each purified anti-S mAb (mAb9, mAb10, mAb17, or mAb18) diluted in Solarbio ELISA coating solution (8, 4, 2, 1, 0.5, 0.25, 0.125, and 0.0625 μg/mL; 100 μL/well) was coated on 96-well ELISA plates (NEST, Wuxi, China) at 4 °C overnight. Then, the plates were washed five times with PBST and blocked with 5% skim milk at 37 °C for 2 h. After washed, each coated well was incubated with 100 μL inactivated cell-cultivated PEDV supernatants (1 mg/mL) at 37 °C for 1 h to capture viral antigens. The plates were washed and incubated with the generated rabbit anti-PEDV pAbs at 37 °C for 1 h. Following washed, each well was incubated with 100 μL 1:1000 diluted HRP-conjugated goat anti-rabbit IgG (Abcam, Cambridge, England) as secondary antibody at 37 °C for 30 min. After washed, each well was reacted with 100 μL TMB substrate solution (Solarbio) in dark at RT for 5 min. After added with 100 μL/well of ELISA stop solution (Solarbio), OD450 values were immediately measured and recorded on a microplate reader (BMG-Labtech, Offenburg, Germany). The anti-S mAb with the highest OD450 values was chosen to establish ELISA. Three replicates of each sample were run, and each experiment was independently repeated three times.

Optimization of ELISA

The optimal concentrations of the anti-S mAb for coating and the inactivated cell-cultivated PEDV supernatants for viral antigen capture were determined by checkerboard titration [43]. Briefly, the anti-S mAb was two-fold serially diluted as 8, 4, 2, 1, 0.5 and 0.25 μg/mL, and the inactivated cell-cultivated PEDV supernatants were two-fold serially diluted as 4, 2, 1 and 0.5 mg/mL. The concentration was considered optimal when the OD450 ratio of PEDV IgA antibody positive to negative milk samples (P/N) was the highest and the OD450 value of PEDV IgA antibody positive ones was closest to 1.0. The positive and negative milk samples were diluted as 1:20. The coating condition, blocking solution and condition, dilution of milk samples and HRP-goat anti-pig IgA (Abcam) as secondary antibody, and incubation and reaction time periods were further optimized as described above. Three replicates of each sample were run, and each experiment was independently repeated three times. The mean OD450 values presented in Table S2, S3, S4, S5, S6, S7, S8 and S9 were calculated to obtain P/N values according to previous studies [44,45,46].

Determination of the cut-off value of the developed ELISA

The cut-off value between PEDV IgA antibody positive and negative milk samples for the developed ELISA was calculated by the mean OD450 value plus the three-fold SD value of 20 PEDV IgA antibody negative milk samples [20]. The milk samples with the OD450 values ≥ the cut-off value were considered PEDV IgA antibody positive.

Determination of the sensitivity of the developed ELISA

Four PEDV IgA antibody positive and one negative milk samples were two-fold serially diluted as 1:40, 1:80, 1:160, 1:320, 1:640, 1:1280, 1:2560. The sensitivity of the developed ELISA was determined using the diluted milk samples. Three replicates of each sample were run, and each experiment was independently repeated three times.

Determination of the repeatability of the developed ELISA

Five PEDV IgA antibody positive and one negative milk samples were applied to evaluate the repeatability of the developed ELISA. Each sample was measured in triplicate on the same plate at the same time and on different plates at different times. CV was calculated as the ratio of the SD value to the mean OD450 value [20]. When the CV value was less than 10%, it was considered an acceptable repeatability [20].

Clinical milk sample detection

A total of 92 unidentified clinical milk samples were simultaneously tested with the developed ELISA and the commercial Anheal ELISA kit to evaluate the detection performance of the developed ELISA. The sensitivity and specificity were calculated according to [the number of true positive samples/(the number of true positive samples + the number of false negative samples)] × 100% and [the number of true negative samples/(the number of true negative samples + the number of false positive samples)] × 100%, respectively [29].