Study of the immunogenicity of recombinant protein consisting of multiple epitopes of porcine epidemic diarrhea virus highly expressed in Sf9 cells

Current inactivated and attenuated vaccines can not provide complete protection, resulting in the ongoing epidemic of porcine epidemic diarrhea virus (PEDV), which has caused signicant economic losses in the swine industry worldwide. To develop an alternative ecient and economical vaccine, a fusion protein named S1CD containing multiple neutralizing epitopes of PEDV was designed and expressed in Sf9 cells. The expression yield was optimized by production of baculoviral constructs harboring multiple S1CD-expressing cassettes, and the immunogenicity of the puried protein was evaluated. The results showed that high-level expression of the S1CD fusion protein was achieved by multi-copy gene co-expression. The recombinant virus carrying three copies of S1CD gene under the control of p10 promoter showed the highest expression level, with a yield of 462 mg/L, which is about 2.6 or 3.5 fold that of the recombinant viruses carrying only one copy. BALB/c mice were subcutaneously immunized with puried S1CD protein three times at 2 week intervals. After immunization, S1CD protein could induce the mice produce high level of specic IgG antibodies and neutralizing antibodies.


Conclusion
Fusion S1CD protein consisting of multiple epitopes of porcine epidemic diarrhea virus was highly expressed by multi-copy gene co-expression. Moreover, it stimulated mice to produce high level of speci c IgG antibodies and neutralizing antibodies. The S1CD protein expressed in this study could be used as a putative economic and e cient subunit vaccine against PEDV infection.

Background
Porcine epidemic diarrhea (PED) is a highly contagious intestinal disease, which is characterized by acute watery diarrhea and vomiting followed by severe dehydration leading to death, with a mortality rate of up to 90% in newborn piglets [1] . This disease was rst reported in England in 1971 and subsequently spread to other European countries and most of the Asian swine raising countries [2] . Since October 2010, outbreaks of highly virulent PED caused by new variant strains have been reported almost all over the world [3][4][5] . Currently, although pigs have been immunized with commercial PED vaccines, which mostly include inactivated and attenuated vaccines, PED still frequently broke out and caused signi cant economic losses to the global swine industry [6,7] . Therefore, it is important to develop an effective vaccine for the prevention of PED. Porcine epidemic diarrhea virus (PEDV), the causative agent of PED, belongs to the genera Alphacoronavirus, the family Coronaviridae, and the order Nidovirales [8] . It is a pleomorphic enveloped, single-stranded positive-sense RNA virus, with a genome size of about 28kb [9] . The virus possesses four structural proteins including spike protein (S), membrane protein (M), envelope protein (E), and nucleocapsid protein (N) [10] . Among the structure proteins, the S protein contains both the receptor binding domain, which mediates virus-host recognition and virus internalization, and the antigenic epitope, which is responsible for inducing neutralizing antibody (NA) [11] . Therefore, the S protein is the principle antigenic determinant and has the potential to be an ideal vaccine candidate [12] .
The S protein can be divided into two domains: S1 (1 ~ 789aa) and S2 (790 ~ 1383aa) [13] . Previous work has shown that the S1 domain contains the main neutralizing epitopes, including the CO-26K-equivalent epitope (COE), and the S1D domain [14] . The COE epitope (499 ~ 638aa), which corresponds to the collagenase-digested fragment (namely CO-26K) of transmissible gastroenteritis virus (TGEV), has been considered an ideal antigenic candidate for development of subunit vaccine [15,16] . In addition, previous studies have shown that the epitope region designated S1D (636 ~ 789aa) in the S1 domain has the capacity to induce neutralization antibodies against PEDV. Moreover, the immune serum against the two linear epitopes of S1D domain, namely S1D5 (744 ~ 759aa) and S1D6 (756 ~ 771aa), could react with the native S protein of PEDV [17] . Therefore, due to above-mentioned studies in the neutralizing antigenic epitopes of S1 domain, a multiepitope-based fusion protein (S1CD) containing COE, S1D5 and S1D6 epitope may be appropriate targets for developing subunit vaccine against PEDV. Baculovirus expression vector system (BEVS) has been widely utilized as a vaccine expression/delivery vector [18,19] . It shows many advantages over other expression systems, such as large cloning capacity, proper post-translation modi cations, high titers and biosafety, and strong induction innate immune response [20] . Therefore, the baculovirus expression system is a promising platform for the development of subunit vaccines in humans and animals. However, a main limitation of BEVS is the comparatively low protein yield compared with E.coli or yeast expression system. Here we report that application of multicopy gene co-expression in BEVS permits substantially improved S1CD protein production in Sf9 cells.
We also evaluated the immunogenicity of puri ed S1CD protein in BALB/c mice.

Generation and Characterization of recombinant baculoviruses
The recombinant bacteria carrying positive bacmids was used to infect Sf9 cells to produce recombinant baculovirus as described previously [21] . The expression of mCherry gene was used to trace proper viral infection. When examined by uorescence microscopy, it was found that obvious cytopathic effect (CPE) and red uorescence in virus-infected Sf9 cells. The presence of the S1CD gene in recombinant viruses was con rmed by direct PCR of the viral genomic DNA. As expected, PCR ampli cation of the recombinant viral genomic DNA revealed a speci c band of 600 bp. Non-speci c ampli cation was not observed in the negative control sample (Fig 1). These results indicated that S1CD gene was inserted into recombinant baculovirus.
Expression and identi cation of recombinant S1CD protein Sf9 cells were infected with eight recombinant baculoviruses, respectively, at a MOI of 10 for protein expression. At 72 h.p.i, the expression of S1CD protein in different virus-infected insect cells was rst determined by IFA. Since virus-infected Sf9 cells exhibited red uorescence, immuno uorescence in the infected Sf9 cells was monitored by incubation with FITC-labeled antibody. As shown in Fig 2A, the green uorescence indicated that eight recombinant baculoviruses expressed S1CD protein in Sf9 cells. Further analysis by western blot showed that a speci c band appeared at 23 kDa, which correlated with the expected size of S1CD protein ( Fig 2B). Thus, the experimental results suggest that the S1CD protein was expressed in the Sf9 cells. In addition, the S1CD protein was also detected clearly in the cell culture supernatant by the SDS-PAGE and western blot (Fig 2C and Fig 2D), which indicated that the S1CD protein could be released into the cell culture supernatant.

Gene expression yield analysis
Eight recombinant baculoviruses harboring from one to four repeats of S1CD-expressing cassettes under the control of the polh promoter, or the p10 promoter, or both promoters were constructed and used to infect Sf9 cells at a MOI of 5. At the transcriptional level, qPCR showed that the expression levels of S1CD mRNA increased with the increasing number of expression cassettes in baculovirus ( Fig 3A). However, the protein expression level of Ac-IM-p10-S1CD, Ac-IM-2p10-2S1CD, Ac-IM-3p10-3S1CD, Ac-IM-ph-S1CD, Ac-IM-2ph-2S1CD, Ac-IM-3ph-3S1CD, Ac-IM-php10-2S1CD, Ac-IM-2ph2p10-4S1CD were 176 mg/L, 318 mg/L, 462 mg/L, 131 mg/L, 245 mg/L, 345 mg/L, 294 mg/L, 327 mg/L, respectively ( Fig 3B). The maximum yield of S1CD protein in Sf9 cells were obtained from the recombinant baculovirus harboring three-copy tandem S1CD-expressing cassettes driven by the p10 promoter, Ac-IM-3p10-3S1CD. The yield reached 462 mg/L, which is 2.6-fold higher than that of Ac-IM-p10-S1CD, and 3.5-fold higher than than that of Ac-IM-ph-S1CD. Therefore, Ac-IM-3p10-3S1CD was used as seed virus to infect Sf9 cells in the following experiments.
Puri cation of recombinant protein S1CD 4.0 × 10 8 of Sf9 cells were infected with Ac-IM-3p10-3S1CD at a MOI of 5 for 96h. The puri cation of recombinant S1CD protein from the cell lysate was facilitated by the presence of a 6×His-tag at the Cterminal. The recombinant protein was puri ed by Ni 2+ -a nity chromatography, separated on 12% SDS-PAGE and stained with Coomassie blue to evaluate the purity of the expressed S1CD protein. As shown in Fig 4, a speci c band at 23 kDa was observed in the elution buffer, and the protein obtained was highly pure based on Coomassie blue staining.

Antibody responses against PEDV in immunized mice
Speci c IgG antibodies were detected in serum samples collected at different days post primary immunization using an indirect ELISA. As shown in Fig 5A, the S1CD protein group and the commercial PEDV inactivated vaccine group exhibited detectable speci c IgG antibody titers initially at 14 dpi and increased gradually over time. The speci c IgG antibody titers of the S1CD protein group and the commercial vaccine group were signi cantly higher than those of the PBS group at 14 dpi (P < 0.05), 28 dpi (P < 0.01) and 42 dpi (P < 0.01). Meanwhile, the speci c IgG antibody titer of the S1CD protein group was slightly lower than that of the commercial inactivated vaccine group at 14, 28, and 42 dpi, but there was no signi cant difference between them (P > 0.05).
The ability of the puri ed S1CD protein to induce PEDV-neutralizing antibodies was determined using a virus neutralization assay. As shown in Fig 5B, immunization with S1CD protein and commercial PEDV inactivated vaccine elicited neutralizing antibodies against PEDV in all immunization mice, and no signi cant difference was observed between them (P > 0.05). Meanwhile, no neutralizing antibody titers were detected in the mice of the PBS groups ( Fig 7B). Together, these results indicated that the puri ed S1CD protein could induce a strong humoral immune response in mice.

Discussion
In recent years, the ongoing epidemic of PEDV has caused signi cant economic losses to the swine industry in Asia as well as Europe and North America [22] . To date, there have been several PEDV commercial vaccines in the market, mainly including inactivated vaccines and attenuated vaccines.
However, for some reason unknown, none of these vaccines has provided complete protection in piglets [23] . Hence, it is of vital importance and a hot research topic to develop a new, e cient and economical vaccine to protect against PEDV infection.
Currently, several PEDV subunit vaccines, which mainly focus on the S1 domain, have been constructed and shown to induce protective immunity in piglets [24][25][26] . As the main neutralization epitope in the S1 domain, COE and S1D have the capacity to induce effective neutralization antibodies against PEDV, thus may be the ideal antigenic candidate for the development of subunit vaccine [15,27−29] . Moreover, previous studies implied that the immunogenicity of a single antigen epitope is weak and the protective effect is incomplete, and the combination of multiple antigen epitopes can enhance the immunogenicity [30] . For this reason, we evaluated synergistic effects between the COE and S1D through fusion of the proteins. Therefore, in this study, a fusion gene encoding S1CD protein containing multiple neutralizing epitope of S1 domain (COE, S1D5 and S1D6) was synthesized and high-yielding expressed, and its immunogenicity as a subunit vaccine candidate was evaluated in BALB/c mice.
Previous studies reported that full length or truncated S1 protein of PEDV could be produced in various expression systems, including E. coli, Lactobacillus, yeast, mammalian cell and baculovirus [26,31− 34] . Among these expression systems, baculovirus expression system is a powerful platform for commercial production of subunit vaccines (Cervarix®, GlaxoSmithKline; Ingelvac CircoFLEX®, Boehringer Ingelheim) and has also been used for the production of antigen proteins of various viruses [35] . In addition, the S1 protein has abundant potential glycosylation sites which is conducive to maintain its biological function [36] . Thus, BEVS was chosen to express the S1CD protein that may require complex folding machinery and post-translational modi cations to maintain its naturally occurring conformation. IFA and western blot results showed that S1CD protein was expressed in the Sf9 cells. Surprising, the S1CD protein was also detected in the cell culture supernatant, indicating that the S1CD protein could be secreted into the medium. However, the prediction result of SignaIP online predictive sites shows that there is no signal peptide sequence in the S1CD protein. The detailed mechanism of S1CD protein secretion into the medium is still unclear, which needs to be further explored.
The protein yield using conventional BEVS is usually relatively low, which represents a bottleneck to commercial applications [37] . To improve the yield of S1CD protein, we designed a panel of baculoviral constructs which containing from one to four repeats of S1CD-expressing cassettes under the control of the polh promoter, or the p10 promoter, or both promoters. The results suggested that the number of expression cassettes has a positive effect on the mRNA level, however, an increase in the number of expression cassettes did not always enhance its protein expression level, which indicated that increasing the number of expression cassettes can improve the transcription of mRNA but not always enhanced the translation level. It is still unclear as to why such phenomenon occurred here, however, some studies have reported similar conclusion in Pichia pastoris [38] . We speculated that possibly high copy number may exert severe stress on the endoplasmic reticulum (ER) and disturb the metabolic balance, thereby affecting exogenous protein expression [39] . In addition to the number of expression cassette, promoter also affect the protein yield, as the expression level of S1CD protein driven by the p10 promoter was higher than that of polh promoter, when the number of expression cassettes was equal. These results indicated that appropriate promoter and expression cassette number could contribute to increase the protein production in Sf9 cells. In this study, the highest yield obtained was 462 mg/L, which is an acceptable amount of recombinant protein in terms of the possibility of producing a novel vaccine.
The PEDV-speci c IgG antibody and neutralizing antibody (NA) in the serum of mice immunized with puri ed S1CD fusion protein were evaluated compare to that in mice of inactivated vaccine-immunized group and PBS-immunized group, which indicates humoral immune response. The serum isolated from mice immunized with puri ed S1CD protein showed signi cant higher PEDV-speci c IgG antibodies and neutralizing antibodies compared to that in the PBS-immunized mice (P < 0.05), and slightly lower than that in inactivated vaccine-immunized mice (P > 0.05), indicating that S1CD protein induced a high level of immune response. These results were consistent with the previous studies, which reported that epitope 499-600 (COE), 744-759 (S1D5) and 756-771 (S1D6) might represent the immunodominant neutralizing epitope of PEDV [40] . Considering that subunit vaccine can solve some problems experienced with inactivated vaccine, which cannot distinguish between infected and vaccinated animals and has biosafety concerns, the puri ed S1CD protein have been suggested as an alternative to develop a subunit vaccine against PEDV. However, as the BALB/c mice is not a susceptible animal model for PEDV, we cannot use it to carry out challenge experiments to evaluate the protection e cacy of S1CD protein. On the basis of this study, our further studies will include usage of pigs to investigate the host immunity and protection induced by the S1CD protein to develop effective subunit vaccines for the prevention of PED.

Conclusions
In this study, the S1CD fusion gene containing multiple neutralizing epitopes of PEDV (COE, S1D5 and S1D6) was synthetized, and its expression yield in baculovirus expression system was optimized by multi-copy gene expression. The mice immunized with puri ed S1CD protein elicited high level of speci c IgG antibodies and neutralizing antibodies, suggesting the possibility of a subunit vaccine against PEDV infection.

Methods
Virus, Cell line, bacterial strains and plasmid PEDV HeN170821 (GenBank accession No: MK862249.1) was isolated from the intestines of piglets with severe diarrhea in Henan, China, and was maintained in our laboratory. Vero cells were cultured in DMEM (Gibco, USA) in a culture bottle at 37℃ and 5% CO 2 . Sf9 cells were grown and maintained at 27℃ in Sf-

Construction of transfer vectors
The nucleotide sequence of S protein of PEDV CV777 strain (GenBank accession No. JN599150.1) was used as a reference to synthesize COE (nt 1495-1914) gene, and D56 (nt 2230-2313) gene. In this study, we fused D56 gene along with a 6×His-tag to the C terminus of COE gene with the exible linker (GGGGS) 3 to generate a novel fusion gene, denoted as S1CD (Fig 6). The S1CD gene was synthesized, and optimized for expression in Sf9 cells by Sangon Biotech (Shanghai, China). The codon-optimized S1CD gene was cloned into plasmid pYBDM-IM, thereby producing a series of recombinant transfer vectors, which comprising S1CD gene under the control of different promoter or promoter combinations (Fig 7). All of the constructed plasmids were veri ed by nucleotide sequencing.

Generation of recombinant baculoviruses
The recombinant transfer vectors were transformed into E. coli AcMultiBacmid/ rSW106/asd -/inv + competent cells through mini-Tn7 transposition to generate recombinant bacmids. Then, E. coli AcMultiBacmid/rSW106/asd -/inv + cells with recombinant S1CD gene were cultured in LB broth until OD 600 = 0.5 -1. The culture were collected by centrifugation (3000 g) and resuspended in Sf-900 TM SFM medium. The bacterial suspension was adjusted to different densities (10 5 -10 8 cells/mL). Sf9 cells at medium, 500 μL bacterial cells at different concentrations were added to each well to infect Sf9 cells [21] .
After culturing at 27°C for 4 -5 h, each well was washed three times. Finally, 500 μL fresh Sf-900 TM SFM medium was added and incubated for 3-5 days. The transfected cells were lysed by 3 freeze-thaw cycles and the supernatant containing viruses was collected for further infection. After 2 passages in Sf9 cells, the generated viruses were harvested and detected with PCR using S1CD gene-speci c primers (S1CD F: 5'-CTCGAGATGA CTTTGCCATCATTCA -3'; S1CD R: 5'-GCTAGCTTAGTGGTGGTGGTGATGGT -3'). The titers of the various viruses constructed were measured by plaque analysis.

Immuno uorescence identi cation
An indirect immuno uorescence assay (IFA) was performed to determine whether the desired proteins were expressed in the Sf9 cells infected with different recombinant baculoviruses. Brie y, Sf9 cells were seeded in 24-well cell culture plates (2.0 × 10 5 cells/well) and infected with recombinant baculoviruses. Cells were xed with 4 % (v/v) paraformaldehyde at 48 h post infection (h.p.i) and washed three times with PBS containing 0.05% Tween (PBST). Immunostaining was performed with mouse anti-His antibody (Boshide, Wuhan, China) as primary antibody and FITC-labeled goat anti mouse IgG antibody (Invitrogen, China) as secondary antibody. Cells were washed ve times with PBST after incubation with each antibody. Digital images were captured using uorescence microscopy.

ELISA and Western blot analysis
To further evaluate S1CD protein expression level, Sf9 cells were cultured in 50 mL Sf-900 TM SFM medium at 27 °C with shaking (125 rpm) and then inoculated with different recombinant baculoviruses at a multiplicity of infection (MOI) of 5, respectively. After 96 h.p.i, the culture was separated into cells and supernatant at 12000 g for 10 min. The total protein concentration of the S1CD protein in the cells and supernatant was determined by the His Tag ELISA Detection Kit (GenScript, Nanjing, China). For western blot analysis, the cells and supernatant samples were separated on 12% SDS-PAGE gels, respectively, then transferred to polyvinylidene uoride (PVDF) membranes (Merck Millipore, USA) and blocked with 5% (w/v) nonfat milk in PBST for 1 h at 37 °C. The membranes were incubated overnight with mouse anti-His monoclonal antibody at 4°C, rinsed with PBST, and incubated with HRP-conjugated goat anti mouse IgG antibody (Boshide, Wuhan, China). After ve washes with PBST for 5 min each time, detection was performed with DAB (3, 3'-diaminobenzidine) solution (Boshide, Wuhan, China).

Quantitative real-time PCR analysis
To test the differences in S1CD transcripts between Sf9 cells infected with different viruses, quantitative real-time PCR (qPCR) analysis was performed. Total RNA was extracted from virus infected-Sf9 cells and quanti ed by spectrophotometry. 500 ng of total RNA from each sample was used as a template to synthesize cDNA according to the PrimeScript TM II 1st Strand cDNA Synthesis Kit's instructions (TaKaRa).
The synthesized cDNA was used for qPCR analysis of S1CD transcript levels. In qPCR, the primer pairs used for a 145 bp amplicon of S1CD gene were S1F: 5'-GTCATAGTGGTGCCAACCT-3' and S1R: 5'-TTACTATCCTGTGACTTAGA-3'. An internal system control with the primer pair of 28S F: 5'-CTGGCTTGATCCAGA TGTTCAG-3'and 28S R: 5'-GGATCGATAGGCCGTGCTT-3' for the Sf9 cell 28S rRNA was used to normalize the reactions. The relative transcript level of each gene was calculated according to the 2 -Ct , for unnormalized genes, and the 2 -ΔΔCt method, for the genes normalized to 28S. The relative transcript levels in the Ac-IM-p10-S1CD infection sample was considered to be 1 and the relative transcript levels of other samples was calculated accordingly. Each of the ampli cations was run in triplicates to calculate experimental variance for statistical analysis by the Student's T-test.
Puri cation of S1CD protein 4.0 × 10 8 of Sf9 cells were grown in 100 mL volumes in 500 mL polycarbonate Erlenmeyer asks and incubated in an orbital shaker incubator at 125 rpm and 27°C. The initial cell density was 2.0 × 10 5 cells/mL. The Sf9 cells were infected with the recombinant baculoviruses, Ac-IM-3p10-3S1CD, at a MOI of 5. At 96 h.p.i, the cells were harvested by centrifugation (1000 g, 10 mins, 4°C) and lysed by sonication. The lysed Sf9 cells were centrifuged at 12, 000 g for 10 min, the supernatant was puri ed using Ni-NTA agarose according to the manufacturer's instructions.

Immunization of mice
Six-week-old female speci c pathogen-free (SPF) BALB/c mice were purchased from Autobio Diagnostics Co., Ltd in Henan, China, and kept under SPF conditions with free access to standard water and diet. All animal procedures were approved by the Ethical Committee for Animal Experiments (Nanyang Normal University, China). The mice were randomly divided into three groups (G1-G3) each with 10 mice. Group 1 were immunized subcutaneously with 100 μg of puri ed S1CD protein in the presence of Freund's complete adjuvant (Sigma, USA) and boosted twice in the presence of Freund's incomplete adjuvant at 2week intervals. Group 2 were immunized subcutaneously with equivalent volumes of adjuvant as a negative control. Group 3 were immunized subcutaneously with equivalent volumes of the commercial PEDV inactivated vaccine (inactivated virus of PEDV CV777 strain; virus titer before inactivation ≥ 10 7 TCID 50 ). Serum samples from all groups were collected at 0, 14, 28, and 42 days post primary immunization (dpi) for the antibody detection and serum neutralization assay. At the end of experiment, animals were euthanized by the intravenous administration of an overdose of sodium pentobarbital.

Analysis of antibody levels
An indirect ELISA was performed to determine the levels of IgG antibodies in the serum. Brie y, 100 μL of puri ed S1 protein (1 μg/mL, Genscript, China), which expressed as a recombinant protein in E. coli, was added to each well of the 96-well plate and incubated overnight at 4 °C. After blocking with PBS containing 1% bovine serum albumin, serum samples were serially diluted, added to the corresponding wells in triplicate and incubated at 37 °C for 1 h. Then, HRP conjugated goat anti-mouse IgG antibodies were added to each well and incubated for 1 h at 37 °C. Plates were incubated with TMB substrate solution (Sigma, USA) for 15 min in the dark. The reactions were terminated with 2 M H 2 SO 4 and the wells read at 450 nm. Serum samples from Balb/C mouse received PBS as blank control (background). Endpoint titers were de ned as the highest dilution that gave an absorbance two times higher than the background.

Neutralization assay
To determine whether mice generated PEDV-neutralizing antibody, sera (1: 100 dilution, 50 μL) from mice of each group were mixed with an equal volume of PEDV HeN170821 (10 5 pfu/mL) at 37 °C. After 1 h incubation, the treated viruses were used to infect Vero cells in 96-well plates. The plates were incubated at 37 °C in a 5% CO 2 atmosphere and examined daily for 3 days for PEDV-speci c cytopathic effects (CPE). The Reed -Muench statistical method was used to calculate the results.

Statistical analysis
Data were expressed as the means ± SEM of three replicates per test in a single experiment. The differences between each group were compared by one-way analysis of variance (ANOVA), combined with Tukey's post hoc test. P < 0.05 was considered statistically signi cant and P < 0.01 was considered highly signi cant. This procedure was approved by the Nanyang Normal University (Belongs to Chinese Association for Laboratory Animal Sciences, CALAS). All animals were handled strictly in accordance with the good animal practice in order to minimize animal sufferings during blood sampling. Finally, the mice were euthanized by the intravenous administration of an overdose of sodium pentobarbital.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Figure 3
Analysis of the expression level of S1CD gene in Sf9 cells. (A) Analysisof the transcriptional level (mRNA) by qPCR. (B) Analysis of the protein expression level by ELISA detection kit. Data are shown as the mean ± standard error value from three independent experiments. One-way ANOVA followed by Tukey's post hoc test was used to determine signi cant difference between groups. When two sets of data are labeled with superscripts of different letters, it indicates that these sets of data are statistically different (P < 0.05). 1 ~ 8: Samples from the Sf9 cells infected with the recombinant baculovirus Ac-IM-p10-S1CD, Ac-IM-2p10-2S1CD, Ac-IM-3p10-3S1CD, Ac-IM-ph-S1CD, Ac-IM-2ph-2S1CD, Ac-IM-3ph-3S1CD, Ac-IM-php10-2S1CD, and Ac-IM-2ph2p10-4S1CD, respectively.   Immunogenicity of puri ed S1CD protein in mice. BALB/c mice were vaccinated with puri ed S1CD protein, commercial PEDV inactivated vaccine or PBS three times on days 0, 14, and 28. (A) Anti-PEDV speci c IgG antibody levels in the sera. End-point titer (log2) were de ned as the highest dilution yielding an absorbance that was two times higher than the background for the serum samples. (B) Virus NA levels in the sera. Results represent the mean ± standard error value of triplicates. One-way ANOVA followed by Tukey's post hoc test was used to determine signi cant difference between groups (*p < 0.05 and **p < 0.01).

Figure 6
Design of S1CD Fusion protein. (A) Diagram of S1CD containing COE gene of PEDV, linker (GGGGS)3, S1D5 and S1D6 gene of PEDV, and 6 × His tag. S1CD has 205 amino acids with a PI of 6.90 and MW of 23.0 kDa. (B) Amino acid sequence of S1CD; sequence of COE is marked in red; sequence of the linker is marked in blue; sequence of the S1D56 protein is marked in green; and sequence of 6 × His tag is marked in orange.