key: cord-267228-g2tf1jz6
authors: Huang, Ke-Yan; Yang, Gui-Lian; Jin, Yu-Bei; Liu, Jing; Chen, Hong-Liang; Wang, Peng-Bo; Jiang, Yan-Long; Shi, Chun-Wei; Huang, Hai-Bin; Wang, Jian-Zhong; Wang, Guan; Kang, Yuan-Huan; Yang, Wen-Tao; Wang, Chun-Feng
title: Construction and immunogenicity analysis of Lactobacillus plantarum expressing a porcine epidemic diarrhea virus S gene fused to a DC-targeting peptide
date: 2018-03-02
journal: Virus Res
DOI: 10.1016/j.virusres.2017.12.011
sha: 
doc_id: 267228
cord_uid: g2tf1jz6

Porcine epidemic diarrhea virus (PEDV) is one of the most important causative pathogens of swine diarrhea, which is widely prevalent throughout the world and is responsible for significant economic losses in the commercial pig industry, both domestic and abroad. The spike (S) protein in the PEDV capsid structure can carry the major B lymphocyte epitope, which induces production of neutralizing antibodies and provides immunoprotective effects. Moreover, the conserved region encoded by the S gene can be considered a target for establishing a new diagnostic method and is a new candidate for vaccine design. In this study, use of anchorin pgsA' allowed the fusion protein of S-DCpep to express on the surface of recombinant Lactobacillus plantarum (NC8-pSIP409-pgsA'-S-DCpep) NC8 strain. Mice were immunized by lavage administration of the recombinant NC8-pSIP409-pgsA'-S-DCpep, which was observed to induce DC activation and high production of sIgA and IgG antibodies in experimental animals, while also eliciting production of significantly more IgA(+)B220(+) B cells. More importantly, secretion of cytokines IFN-γ, IL-4 and IL-17 in mice that were vaccinated with NC8-pSIP409-pgsA'-S-DCpep was remarkably increased. The results of our study suggest that NC8-pSIP409-pgsA'-S-DCpep potently triggers cellular and humoral immune responses. The obtained experimental results can provide a theoretical basis that lays the foundation for production of a novel oral vaccine against PED.

Porcine epidemic diarrhea (PED) is a highly contagious disease with typical symptoms in piglets that include severe diarrhea, vomiting and intestinal dehydration (Ko et al., 2017) . The morbidity and mortality rates of suckling piglets caused by PEDV are more than 50%, and it is the primary cause of death in piglets and weight loss in fattening pigs (Choi et al., 2009) . This viral disease was first reported in Britain in 1971 and then became an epidemic in other European countries (Yu et al., 2017) . In 1976, PEDV was detected for the first time in China, and since October 2010, PEDV outbreaks worldwide have become more frequent (Song et al., 2015) . For instance, PED broke out in Minnesota and Iowa in the United States in 2013, resulting in huge losses to the pig industry (Diep et al., 2017; Lee, 2015) . As a coronavirus, PEDV has a typical structure that consists of four structural proteins, including the nucleocapsid protein (N), a membrane protein (M), a glycosylated spike protein (S) and a envelope protein (E) (Kocherhans et al., 2001) . The PEDV spike protein-encoding gene is an immune dominant gene that encodes a protein with a very dense hydrophobic region of amino acids that harbors multiple glycosylation sites and has a high antigenic index . Several studies have reported that the S protein is a structural surface protein of PEDV, containing both the receptor binding domain, which mediates invasion of the virus into the host cell, and the antigenic epitope, which mediates production of the neutralizing virus antibody . Thus, the S-glycoprotein plays a crucial role in viral infection, pathogenicity and adsorption. It has been shown that the S-glycoprotein is immunogenic and has the potential to be a good vaccine candidate (Fan et al., 2015) .

Lactic acid bacteria (LAB) are diverse species that are widely distributed in nature and produce large amounts of lactic acid by fermenting carbohydrates (Riaz Rajoka et al., 2017) . LAB, which are generally recognized as safe bacteria, colonize the intestines of humans or animals and have potentially beneficial effects (Landete et al., 2015) . The study of LAB has many advantages, including that they are easy to culture, convenient to manipulate and highly safe, which is why LAB can be regarded as a heterologous antigen protein delivery system (Wanker et al., 1995) . L. plantarum has been manipulated to deliver a number of viral antigens in previous studies (Li et al., 2017b; Yang et al., 2017c) . The L. plantarum strain NC8 was isolated from ensilages and has a better antireversion characteristic compared with LAB isolated from healthy animal intestines (Caggianiello et al., 2016; Park et al., 2017) . Thus, L. plantarum NC8 is more suitable to act as a mucosal delivery shuttle vector to express protective antigens.

Recently, poly-γ-glutamic acid synthetase A (pgsA) was identified as being a constitutively expressed protein of the polyglutamate acid (PGA) synthetase system of Bacillus subtilis. PgsA can stably anchor γ-PGA synthetase on the cell membrane, acting as a transport carrier of γ-PGA and playing a crucial role in the formation of γ-PGA (Sung et al., 2005) . The pgsA protein has a transmembrane region near its N-terminus (26-42 amino acid residues), providing the necessary characteristics for the establishment of a pgsA transmembrane expression system. The pgsA gene was combined with the target gene for fusion expression, after which L. plantarum was used as a mucosal vehicle to deliver an exogenous antigen. This experiment provided a theoretical basis for the construction of a novel, orally administered, genetic engineered vaccine to immobilize exogenous medicinal proteins on the cell wall surface of L. plantarum (Cai et al., 2016; Narita et al., 2006) .

Dendritic cells (DCs) are the most powerful functional antigen presenting cells (APCs) and are the exclusive APCs that are capable of inducing initial T and B cell activation . DCs can elicit T cell polarization and differentiation in T cells subsets, which also trigger secretion of IFN-γ (Subramaniam et al., 2017) . DCs are involved in regulating the function of B cells and promoting the differentiation of B cells into plasma cells to produce IgG . In the germinal center, DCs can boost B cell antibody affinity maturation and help to form and maintain memory B cells. In addition, sIgA can play an important role in maintaining intestinal integrity and intestinal symbioses (Shaw et al., 2000) . DCs can restrict penetration of microorganisms into the mucosa and promote uptake of antigens in the intestinal mucosal system (Owen et al., 2013) . It has been reported that NC8-pSIP409-pgsA'-S-DCpep can specifically conjugate dendritic celltargeting peptide (DCpep) and induce an immune response (Apostolico et al., 2017) . To make DCs recognize the S-antigen more effectively in the intestinal mucosa, recombinant NC8-pSIP409-pgsA'-S-DCpep was constructed using standard molecular biology techniques, and its immunogenicity was further investigated. After BALB/c mice were immunized by lavage administration of our recombinant L. plantarum, the results indicated that the in vivo antibody levels were greatly improved.

The Escherichia coli-Lactobacillus shuttle vector pSIP409 (Sorvig et al., 2005) and L. plantarum NC8 were kindly provided by Professor A. Kolandaswamy (Madurai Kamaraj University, India). The pM18-T-S-DCpep (DC peptide, FYPSYHSTPQRP) was stored at Jilin Provincial Engineering Research Center of Animal Probiotics in Jilin Agricultural University. The pM18-T-S-DCpep plasmid was cut at XbaI and HindIII sites within the primers using the appropriate restriction enzymes at 37°C for 3-4 h. Next, the digested 2240 bp S gene fragment (Spike Protein Gene GenBank: JQ257007.1 (source: 75-2313)/Spike Protein GenBank: AEW24858.1 (source: 26-771)) and DCpep (S-DCpep) were ligated into pSIP409-pgsA' using T4 DNA ligase at 16°C overnight. The recombinant pSIP409-pgsA'-S-DCpep plasmid was detected and purified using a gel extraction kit (Axygen). The recombinant plasmid was introduced into the L. plantarum NC8 strain by electroporation and erythromycin-resistant bacteria were selected for, after which the plasmid was sequenced. In addition, to meet the rigorous nature of the experimental study, a NC8-pSIP409-pgsA'-S-Ctrlpep (control peptide, EPIHPETTFTNN) was constructed similar to NC8-pSIP409-pgsA'-S-DCpep and used as the experimental control group.

To examine the expression of the S-protein on the surface of recombinant NC8-pSIP409-pgsA', NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA'-S-DCpep was cultured in MRS with 10 μg/ml erythromycin. As described in previous studies, we added Sakacin P (SppIP) to the culture medium when the (OD 600 ) of the medium was between 0.3 and 0.4. After induction at 37°C for 10 h, cells from each strain were washed with PBS containing 1% bovine serum albumin. Next, cells were resuspended in PBS and 1 × 10 5 CFU of bacteria were incubated with 1 ml of primary antibody (1/2000 anti-S-produced in rat) for 1 h at 37°C with shaking at 60 rpm. The pellets were washed with PBS containing 0.2% Tween-20 (PBS-T), after which secondary antibodies (FITC, goat anti-rat IgG (H + L), CST) were added and the mixture was incubated for one hour at 37°C with shaking at 60 rpm in the dark. Finally, samples were washed with PBS-T and resuspended in PBS. The cells were examined under a laser scanning confocal microscope (LSM710; Carl Zeiss, Germany) following the manufacturer's instructions.

To detect the S-protein expression levels, we induced its expression in recombinant LAB as described previously. Strains were subsequently resuspended in 500 μl of PBS and were lysed by ultrasonic crushing as previously described (Shi et al., 2014) . Samples were assessed by 10% SDS-PAGE and transferred to PVDF membranes (Millipore, USA). The S-Ctrlpep and S-DCpep proteins were detected using an anti-S polyclonal primary antibody (polyclonal antibody was obtained from serum of mice which were immunized with the purified S protein) and a HRPconjugated goat anti-rat IgG secondary antibody (A21040; Abbkine, USA). The bands were visualized using enhanced chemiluminescence reagents in an ECL Plus detection kit (Thermo Scientific) according to the manufacturer's instructions.

For experiments involving animals, thirty-six female BALB/c mice without specific pathogen (SPF) were bought from Beijing HFK Bioscience Co., Ltd., China. According to the provisions of the Animal Care and Ethics Committees of Jilin Agriculture University. All animals were strictly raised with commercialized germ-free feed and sterilized water in the Jilin Provincial Engineering Research Center of Animal Probiotics, which provided the pathogen-free environment facilities.

Six-week-old mice with no specific pathogens were randomly divided into four groups, with 6 mice in each group (n = 9). Briefly, using oral gavage, four groups of mice were immunized with NC8-pSIP409-pgsA', NC8-pSIP409-pgsA'-S-Ctrlpep, NC8-pSIP409-pgsA'-S-DCpep or saline. The entire immunization procedure was divided into three phases: prime immunization (days 1-3), booster immunization (days 8-10), and the last immunization (days 29-31), with 10 8 CFU/mouse administered by oral gavage.

Generally, experimental mice were sacrificed by cervical dislocation 30 days after being immunized. Lamina propria (LP) cells of the small intestine, PPs and MLNs were acquired via a previously described method (Kikuchi et al., 2014 ) on a clean sterile clean bench. Small intestine samples were washed with PBS and cut into small pieces. Next, samples were digested with lymphocyte separation medium, after which LP cell digestive juice was added to obtain single LP cells from the small intestine in an isotonic percoll solution. The cell suspension was filtered through a 70 μm sterile filter-film and the filtrate was centrifuged at 500g for 5 min at 4°C. Subsequently, the cells were washed with 4°C PBS and were eventually resuspended in complete RPMI 1640 (containing 10% FBS, 1% penicillin and streptomycin).

The method used for the analysis of S protein expression in recombinant L. plantarum was consistent with the immunofluorescence assay, with cells analyzed by flow cytometry (BD FACS LSR Fortessa™ Flow Cytometer). Data were analyzed via FlowJo 7.6.1 software. To test the activation of DC surface costimulatory molecules and B cells, flow cytometry was conducted using a previously published protocol . In brief, single LP cells from the small intestine suspensions obtained from immunized mice were diluted to 1 × 10 6 cells/ 100 μl and incubated with 10 μl anti-CD11c (clone HL3), CD80 (clone 16-10A1), CD86 (clone GL1), CD40 (clone 3/23) and MHCII (clone 3/ 23) antibodies to assess the DC cells. To analyze B lymphocyte cells activation, 10 μl of anti-B220 (clone RA3-6B2) and IgA (clone C10-3) antibodies was added to stain B lymphocyte cells for 20 min at room temperature in the dark. Afterwards, cell suspensions were washed twice with FACS buffer. After staining, samples were examined using flow cytometry (BD FACS LSR Fortessa™ Flow Cytometer). Data were analyzed using FlowJo 7.6.1 software.

Blood samples were obtained on days 14, 35 and 42 after immunization and were centrifuged at 4000 rpm for 15 min at 4°C to obtain the serum, which was stored at −80°C. The fecal samples were prepared according to our previously published method, followed by centrifugation at 5000 rpm for 20 min at 4°C. The supernatants were acquired and stored at −80°C for further analysis. ELISA was performed to determine the levels of specific IgG and sIgA antibodies (Yang et al., 2017d) . In brief, 100 μl of purified S protein (4 μg/ml) was added to each well of the 96-well plate and incubated overnight at 4°C. On the second day, the blocking solution was added and the plate was incubated overnight at 4°C. On the third day, the fecal samples (1:10) and serum samples (1:100) were diluted and added separately, after incubating at 37°C for 2 h, the wells were rinsed. Subsequently, goat anti-mouese IgG (H + L)-Biotin and goat anti-mouse IgA-Biotin (Southern Biotechnology, Birmingham, AL) was added separately, and the plate was incubated at 37°C for one hour, after which Streptavidin-HRP (Southern Biotechnology, Birmingham, AL) secondary antibody was added and the plate was incubated at 37°C for 30 min. Finally, TMB substrate solution and stop solution was added, after which the color intensity was measured at 492 nm. The End-point titers (log2) were defined as the highest dilution yielding an absorbance that was two and three times higher than the background for the fecal and serum samples, respectively .

The MLN cells were stimulated by S protein (5 μg/ml) and the supernatant was collected 3 days later and examined for specific cytokines, including IFN-γ, IL-4 and IL-17A using an ELISA kit (LIUHEBIO, Wuhan, China) according to the manufacturer's instructions.

For determining the PEDV neutralizing activity of the serum antibodies we detected sample obtained from mice which were orally immunized with recombinant L. plantarum by antibody neutralization assay (microdetermination) as described before (Tian et al., 2014) with minor modifications. Under aseptic conditions serum and fecal samples were collected at 14, 35 and 42 days respectively after immunization, quantitative virus and serial diluted serum were added to Vero cells to evaluate serum neutralizing antibody titers, the neutralizing antibody titer PD50 (Median protective dose refers to the dose of serum or vaccine that protects half (50%) of the experimental animals) was determined by Karber's method at 72 h.

All results were performed by Two-tailed t-tests and One-way analyses of variance (ANOVA) (GraphPad Prism 5.0, GraphPad Software) were used to analyze the significance of the difference between means. All data were obtained from at least three independent experiments, and the geometric mean titers were expressed as averages. All data were expressed as the means ± SEM.

The pSIP409-pgsA'-S-DCpep and pSIP409-pgsA' plasmids were introduced separately into L. plantarum NC8 strain by electroporation (Fig. 1A) . Using flow cytometry, S protein expression levels of recombinant L. plantarum NC8-pSIP409-pgsA'-S-DCpep and NC8-pSIP409-pgsA'-S-Ctrlpep were determined (Fig. 1B) . The green fluorescence marked by white arrows indicated that recombinant Lactobacillus of NC8-pSIP409-pgsA'-S-DCpep and NC8-pSIP409-pgsA'-S-Ctrlpep expressed S protein on the surface (Fig. 1C ) than NC8-pSIP409-pgsA', which was barely observed with green fluorescence under a confocal microscope. Western blot analysis showed that a specific band appeared at 80 kDa, which demonstrated that the PEDV S protein had been expressed. Thus, the experimental results suggest that NC8-pSIP409-pgsA'-S-DCpep and NC8-pSIP409-pgsA'-S-Ctrlpep were expressed in L. plantarum NC8.

To determine the effect of NC8-pSIP409-pgsA'-S-DCpep on DCs, flow cytometry was performed to test the level of costimulatory molecules activated by DCs, which were induced by NC8-pSIP409-pgsA'-S-DCpep gavage immunization 3 days after the last immunization ( Fig. 2A) . Compared with the saline group, the expression level of CD11c + CD40 + of DCs surface molecules in the LP cells of the small intestine was significantly increased in the NC8-pSIP409-pgsA'-S-DCpep group (P < 0.01) and NC8-pSIP409-pgsA'-S-Ctrlpep group (P < 0.05) experimental groups (Fig. 2B) . Moreover, our experimental results indicated that NC8-pSIP409-pgsA'-S-DCpep can promote greater expression of CD80 + in LP DCs than the NC8-pSIP409-pgsA'-S-Ctrlpep (P < 0.05), NC8-pSIP409-pgsA' (P < 0.05) or saline (P < 0.01) groups (Fig. 2C) . However, no obvious differences were observed in CD11c + CD86 + expression of DCs surface molecules in the LP cells of the small intestine.

The number of IgA + B220 + B cells in PP was determined by flow cytometry 3 days after the last immunization (Fig. 3A) . The results showed that there was a remarkable increase in the percentage of IgA + B220 + B cells within the B cell populations of BALB/c mice having received lavage administration with NC8-pSIP409-pgsA'-S-DCpep compared with the NC8-pSIP409-pgsA', NC8-pSIP409-pgsA'-S-Ctrlpep and saline groups. In addition, NC8-pSIP409-pgsA'-S-DCpep in PP significantly increased the production of IgA + B220 + B cells compared to the NC8-pSIP409-pgsA' and saline group control (P < 0.01) (Fig. 3B) , and the NC8-pSIP409-pgsA'-S-DCpep was significantly higher than the NC8-pSIP409-pgsA'-S-Ctrlpep groups in PP (P < 0.05) (Fig. 3B) . Collectively these results demonstrated that vaccination with the NC8-pSIP409-pgsA'-S-DCpep can activate the B cell response in mice.

The specific sIgA levels in fecal intestines were measured at days 14, 35 and 42 after the first immunization using ELISA (Fig. 4) . However, the concentration of sIgA in the excrement of mice treated with NC8-pSIP409-pgsA'-S-DCpep was higher than that of the NC8-pSIP409-pgsA' (P < 0.05) and saline (P < 0.05) groups at days 14 (Fig. 4) , and NC8-pSIP409-pgsA'-S-DCpep was higher than that of the NC8-pSIP409-pgsA' (P < 0.05) and saline (P < 0.01) groups at day 35 (Fig. 4) . In contrast, no obvious difference was observed with respect to SlgA was observed at days 14 and 35 in the NC8-pSIP409-pgsA'-S-DCpep and NC8-pSIP409-pgsA'-S-Ctrlpep groups (Fig. 4) . With respect to the 42 day lavage administration, the NC8-pSIP409-pgsA'-S-DCpep group was higher than that of the NC8-pSIP409-pgsA' (P < 0.01) and saline (P < 0.01) groups. Higher expression of sIgA was also observed in NC8-pSIP409-pgsA'-S-DCpep group compared with NC8-pSIP409-pgsA'-S-Ctrlpep group (P < 0.05) at day 42 (Fig. 4) . These data suggested that lavage administration of NC8-pSIP409-pgsA'-S-DCpep could stimulate the mucosal immune system in mice.

The specific IgG-antibodies level after lavage administration of NC8-pSIP409-pgsA'-S-DCpep in mice was notably increased compared with the NC8-pSIP409-pgsA ', NC8-pSIP409-pgsA'-S-Ctrlpep and saline groups at days 14, 35 and 42 after the first immunization. The results showed that the concentration of specific IgG in the serum of mice treated with NC8-pSIP409-pgsA'-S-DCpep was significantly increased compared to the NC8-pSIP409-pgsA' (P < 0.05) and saline (P < 0.01) groups at days 14 and 35 (Fig. 5 ). In addition, only NC8-pSIP409-pgsA'-S-DCpep showed higher levels than the NC8-pSIP409-pgsA'-S-Ctrlpep (P < 0.05), NC8-pSIP409-pgsA' (P < 0.01) and saline (P < 0.001) groups at day 42 (Fig. 5) , with an increasing trend from days 28 to 35 (Fig. 5) . These data suggested that lavage administration of NC8-pSIP409-pgsA'-S-DCpep could increase IgA antibody titers and produce humoral immune responses in mice.

Specific cytokines in the MLN cell secretions were measured 3 days after the last immunization using ELISA. The results showed that the concentration of IL-4 in the supernatants of MLN cells cultured from the NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA'-S-DCpep-treated mice were significantly increased compared to those from the saline group (P < 0.01) (Fig. 6A) . Higher level of IL-17 was observed in the group of mice orally immunized with recombinant NC8-pSIP409-pgsA'-S-DCpep compared to the group of mice orally administered with saline (P < 0.01), NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA' groups (P < 0.05) (Fig. 6C) . Unexpectedly, the level of IFN-γ in the supernatant of MLN cells cultured with the strains expressing S-DCpep was significantly higher in the group of mice orally immunized with recombinant NC8-pSIP409-pgsA'-S-Dcpep compare to the group of mice orally administered with saline (P < 0.01), NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA' groups (P < 0.05) (Fig. 6B ). Fig. 3 . Detection of activated B cells in PP of immunized mice using flow cytometry. The number of total cells for (A) the quadrants and (B) the bar graphs refer to IgA + B220 + cells. The degree of activation of B cells was analyzed by counting the percentage of IgA + B220 + expression. Significant differences relative to the saline and NC8-pSIP409-pgsA' controls (P < 0.05, P < 0.01). Significant differences are denoted by an asterisk (*) between saline, NC8-pSIP409-pgsA', NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA'-S-DCpep controls (*p < 0.05, **p < 0.01, ***p < 0.001).

K.-Y. Huang et al. Virus Research 247 (2018) 84-93 3.7. NC8-pSIP409-pgsA'-S-DCpep induced production of serum/fecalneutralizing antibodies

Specific PEDV antibodies titers in serum and faeces were measured by cell neutralization assays at different time after the first immunization. Results showed that serum and fecal antibodies of mice orally immunized with recombinant NC8-pSIP409-pgsA'-S-DCpep was higher than that from mice orally administered with and Saline at days 14 (P < 0.01) (Fig. 7) . The serum antibody of mice orally immunized with recombinant NC8-pSIP409-pgsA'-S-DCpep was significantly increased compared with NC8-pSIP409-pgsA' at days 35 (P < 0.01) (Fig. 7A) , however, the fecal antibodies of NC8-pSIP409-pgsA'-S-DCpep was no significant difference at days 35 compared with saline and NC8-pSIP409-pgsA' at days 14 (Fig. 7B ). Significant differences were observed in both serum and fecal antibodies of mice orally immunized with recombinant NC8-pSIP409-pgsA'-S-DCpep compared with the other three group at days 42 (Fig. 7) , particularly, difference were also observed in serum and fecal antibodies between the group of mice orally immunized with recombinant NC8-pSIP409-pgsA'-S-DCpep and NC8-pSIP409-pgsA'-S-Ctrlpep (P < 0.05). The results suggesting that recombinant L. plantarum NC8-pSIP409-pgsA'-S-DCpep has good PEDV virus-neutralizing activity in humoral and mucosal immunity.

Porcine epidemic diarrhea (PED) caused by PEDV is a contagious disease which seriously threaten pig production industry (Huang et al., 2013) . PED characterized by high morbidity and mortality in piglets was reported in southern China in fall of 2010 (Huang et al., 2013 . Furthermore, in April 2013 PED has also spread throughout and to the highest degree destroyed the American pork industry, the homology between the genome of popular strain in American and China is as high as 97%-99%, which has caused immeasurable economic losses (Huang et al., 2013) . The continuing outbreak of PED suggests that it is able elude current immunization strategies (Curry et al., 2017) . Currently, there is no effective drug for PEDV treatment, although inactivated and live vaccines are the two most commonly used vaccines, Although live and inactivated vaccines are already in widespread use, development of effective, safe and economical oral vaccines are still the trend of future. And researchs of developing oral vaccine to prevent PEDV have achieved effective protection.

L. plantarum is an intestinal predominant flora with immunomodulatory effects that enhance the chemotaxis and phagocytosis of neutrophils, prevent viral infection and maintain intestinal microbial homeostasis (Li et al., 2017a,b) . Therefore, L. plantarum with probiotic properties has become good exogenous protein vector platform. The anchorin expressed by pgsA gene can help exogenous proteins to locate to the surface of bacterium and display their biological function. Other and our team have achieved good protection using pgsA to display exogenous antigens, hence, using pgsA to display exogenous antigens during vaccine development is a viable strategy.

The feature of localized intestinal infection of PEDV determined that the immunization and prevention is difficult, hence, intestinal mucosal immunity plays an important role in pigs except suckling piglets (Temeeyasen et al., 2017) . Resistance to PEDV infection often relies on IgA secreted by mucosa, which requires the development of vaccines that can effectively increase intestinal immunity. Contemporary strategies in vaccination primarily focus on using different vectors to express the identified antigenic regions. Antigens delivered by different routes might stimulate the intestinal mucosal immune system to achieve immunoprophylaxis effect. The use of probiotics to express antigens and proliferate them in the intestinal of host can promote the digestion of livestock and poultry; meanwhile, the protective antigen expressed can stimulate the intestinal mucosal immune system to produce protective effect.

As a favorable mucosal immunization vaccine vector, L. plantarum can also be used as a natural adjuvant in the mucosal microenvironment and attenuate inflammatory response to enhance the body immunity (Yu et al., 2015) . Morever, vaccines constructed by DCpep could provide effective protection (Yang et al., 2017a,b) . Consequently, we constructed fusion protein by the immunogenic S-protein of PEDV and the induced DCpep and expressed it on the surface of L. plantarum, which could activate DCs to initiate mucosal and humoral immune responses against microbial infection via secreting antibodies to clear the antigens. Results showed that the DCpep fusion group had better immune protection compared with other groups.

DCs are the foremost antigen presenting cells (APCs) and play an important role in inherent and adaptive immunity (Sato et al., 2017) . CD11c + is a surface marker of DCs, and CD80 + and CD40 + are subsets of CD11c + . There was a difference in expression of CD11c + CD40 + between the group of mice orally immunized with recombinant NC8-pSIP409-pgsA'-S-DCpep and saline group but not NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA' groups, however, the MEI of group NC8-pSIP409-pgsA'-S-DCpep was still higher than the above two groups ( Fig. 2A) . The expression of the CD11c + CD80 + in NC8-pSIP409-pgsA'-S-DCpep group was significantly higher than the control group and the NC8-pSIP409-pgsA' group in the LP of small intestine (Fig. 2B) . In this study, the expression of CD11c + CD86 + in DCs was no significant difference between NC8-pSIP409-pgsA'-S-DCpep and saline group, which should also increased according to previous studies. This difference in results might be due to different host strains and different immunization programs. However, CD11c + CD86 + did not affect the biological function of CD11c + CD80 + and CD11c + CD40 + DCs, which could also identify CD40 + , CD80 + , CD86 + and other co-stimulatory molecules, Fig. 4 . S-specific sIgA titers levels in mouse feces examined by ELISA se. The S-specific sIgA Ab titers were detected after prime immunization, booster immunization, and the last immunization. End-point titer (log 2 ) were defined as the highest dilution yielding an absorbance that was two and three times higher than the background for the fecal and serum samples, respectively. Significant differences are denoted by an asterisk (*) between Stroke-physiological saline solution, NC8-pSIP409-pgsA', NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA'-S-DCpep controls (*p < 0.05, **p < 0.01, ***p < 0.001).

S-specific IgG titers levels in sera of mice detected by ELISA. The S-specific IgG Ab titers were detected after prime immunization, booster immunization, and the last immunization. End-point titer (log 2 ) were defined as the highest dilution yielding an absorbance that was two and three times higher than the background for the fecal and serum samples, respectively Significant differences are denoted by an asterisk (*) between Stroke-physiological saline solution, NC8-pSIP409-pgsA', NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA'-S-DCpep controls (*p < 0.05, **p < 0.01, ***p < 0.001). present antigen to activated T cells or effector T cells. Meanwhile, CD80 + and CD86 + of activated B cells could combine with CD28 expressed on the surface of T cells, which involved in co-stimulatory signal transduction. In addition, the results of S-specific IgG in serum and S-specific sIgA in feces showed that L. plantarum strains expressing S protein could improve the secretion of IgG and sIgA and further enhance immune function, particularly for NC8-pSIP409-pgsA .

Several studies showed that IgA secreted by mucosal lamina propria could combine with mucus and cover on the surface of epithelial cells which inhibit the adhesion of microorganisms to epithelial cells and neutralization of toxins produced by microorganisms or enzymes. SIgA agglutinates and incapacitates pathogens by which it inhibits pathogen adhesion to the mucosal surface and easily cleared in the secretions. Therefore, sIgA disenables pathogens interact with epithelial cell receptors and inhibits the assembly of viral particles within the host cell cytoplasm (Kurashima and Kiyono, 2017; Motegi et al., 2000) . Our study suggesting that the anti-S mucosal sIgA responses of recombinant Lactobacillus NC8-pSIP409-pgsA'-S-Dcpep group significantly increased. Following four indicators are used to evaluate the effectiveness of a vaccine in mucosal system: connecting closely with lymphoid tissue in mucous membrane; the existence of scattered lymphoid tissues and lymphoid organs with a certain structure; DCs use a special mechanism to uptake antigen; and the presence of many activated and memory lymphocytes in the absence of an infection (Chung et al., 2017; Faria and Reis, 2017) . Currently, many investigators are focused on the development of mucosal vaccines. Our study demonstrated that the recombinant L. plantarum NC8-pSIP409-pgsA'-S-DCpep could increase the activation of B cells and further improve the level of humoral immunity (Fig. 3) . Orally administration of recombinant NC8-pSIP409-pgsA'-S-DCpep in mice boosted the number and frequency of IgA + B220 + B cells compared with the NC8-pSIP409-pgsA' and NC8-pSIP409-pgsA'-S-Ctrlpep groups. Similarly, we also detected anti-S systemic IgG in the blood, The results showed that S-DCpep could effectively initiate a mucosal immune response and humoral immunity. Subsequently, L. plantarum expressed S-DCpep fusion protein could enhance the secretion of mucosal antibodies against pathogens (Mohamadzadeh et al., 2009; Shi et al., 2016; Yang et al., 2017a Yang et al., ,b, 2016 . In the construction of the recombinant strain NC8-pSIP409-pgsA'-S-Dcpep, pgsA has been used as an anchorin for its stability and safety in clinical use. Besides, pgsA has a wide range of applications for surface expression on some types of Lactobacillus compared with other anchorins. Therefore, we used pgsA as a surface display element for applying to surfaced of L. plantarum. A recombinant vaccine constructed by S-protein expressed on the surface of L. plantarum and DCpep fusion protein can provide protection since PEDV is primarily replicated in the small intestine. The results of western blot, FACS and confocal microscopy analyses suggested that S protein was expressed on the surface of L. plantarum.

Compared with traditional intramuscular injection, gavage administration allows antigens to be presented through the gastrointestinal mucosa and obtain better immunological response, which is a simple, convenient and ideal immunization strategy. Several genetically engineered bacteria (Salmonella, Escherichia coli, Lactobacillus) could colonize in the intestines after inoculation that were regarded as promising oral vaccine vectors (Garcia-Fruitos, 2012 ). In addition, peptidoglycan expressed on the surface of L. plantarum, could function as a natural immune adjuvant and be utilized to design a novel vaccine (Dieye et al., 2001) . Previous studies have demonstrated that L. plantarum could act as a mucosal delivery system to induce immune response and tolerance (Tauer et al., 2014) . The distinctive feature of the immune response triggered by oral mucosal immunization is the production of specific secretory IgA. Antigens were taken up by M cells after oral inoculation, then dendritic cells and T cell subsets were activated, cytokines and chemokine were released and MHCI and MHCII were expressed (Kurashima and Kiyono, 2017) . Finally the activation of B cells, the expression of specific integrin, phenotype transformation (especially to IgA type conversion) are observed (Perdigon et al., 2002) .

Th1 and Th2 responses induced by recombinant L. plantarum were evaluated via cell-specific cytokines analysis. Th1 cell response can helpcytotoxic T cell differentiation mediate cellular immune responses, secrete IFN-γ and activate macrophages, which enhance the ability of cells to kill phagocytized pathogens (Maldonado-Lopez and Moser, 2001). Besides, the primary function of Th2 cells is to stimulate B cell proliferation and produce antibodies, which is related to humoral immunity (Mosmann and Coffman, 1989) . Secretion of IL-4 increased the ability of monocytes differentiate into macrophages . IL-17 can increase the ability of costimulatory molecules, MHC, antigens and T lymphocytes to stimulate antigen peptide delivery and promote DC maturation . Therefore, we analyzed cytokine system IFN-γ, IL-4, and IL-17 of Th1/Th2/Th17 cells secreted by the MLN of the immunized animals after treatment with recombinant bacteria. Our results showed that NC8-pSIP409-pgsA'-S-DCpep could significantly increase cytokine secretion and stimulate Th1 and Th2 cells and Th17 cell response, maintain the balance of Th1, Th2 immune response. The focus of our study was Th1 and Th2 immune responses induced by NC8-pSIP409-pgsA'-S-DCpep, which also stimulated the activation of Th17 cells (Fig. 6) . Taken together, recombinant L. plantarum expressed S-DCpep could improve Th1/Th2/Th17 immune response in BALB/c mice and induce immune responses, such as the differentiation of B cells and DCs maturation.

In this study, we detected PEDV-specific neutralizing antibody in serum and fecal obtained from the group of mice orally administrated with L. plantarum expressed PEDV-S protein on the surface, which could significantly inhibit the formation of plaque on Vero cells. The results indicated that systemic immune response induced by recombinant Lactobacillus provided an immunoprotective effect against the virus as shown in (Fig. 7) . The constructed recombinant L. plantarum might contribute to the development of novel vaccines against PEDV Fig. 7 . NC8-pSIP409-pgsA'-S-DCpep induced mice to produce serum(A)/feces(B)-neutralizing antibodies against PEDV. PD50 (Median protective dose refers to the dose of serum or vaccine that protects half (50%) of the experimental animals).Significant differences are denoted by an asterisk (*) between Stroke-physiological saline solution, NC8-pSIP409-pgsA', NC8-pSIP409-pgsA'-S-Ctrlpep and NC8-pSIP409-pgsA'-S-DCpep controls (*p < 0.05, **p < 0.01, ***p < 0.001).

challenge.

In conclusion, our study demonstrates that NC8-pSIP409-pgsA'-S-DCpep can modulate mucosal and humoral immunity and produce specific antibodies against S-protein. Next, we will study the protective effect of recombinant bacteria against PEDV challenge in piglet.

Exopolysaccharides produced by lactic acid bacteria: from health-promoting benefits to stress tolerance mechanisms

Surface-displayed IL-10 by recombinant Lactobacillus plantarum reduces Th1 responses of RAW264.7 cells stimulated with poly (I:C) or LPS

Antiviral activity of quercetin 7-rhamnoside against porcine epidemic diarrhea virus

Sinulariolide suppresses LPS-induced phenotypic and functional maturation of dendritic cells

Appearance of US-like porcine epidemic diarrhoea virus (PEDV) strains before US outbreaks and genetic heterogeneity of PEDVs collected in Northern Vietnam during

Design of a protein-targeting system for lactic acid bacteria

Development of an enzymelinked immunosorbent assay for the monitoring and surveillance of antibodies to porcine epidemic diarrhea virus based on a recombinant membrane protein

Tissue adaptation: implications for gut immunity and tolerance

Lactic acid bacteria: a promising alternative for recombinant protein production

Oral administration of Lactobacillus plantarum strain AYA enhances IgA secretion and provides survival protection against influenza virus infection in mice

Rapamycin-induced autophagy restricts porcine epidemic diarrhea virus infectivity in porcine intestinal epithelial cells

Completion of the porcine epidemic diarrhoea coronavirus (PEDV) genome sequence

Mucosal ecological network of epithelium and immune cells for gut homeostasis and tissue healing

Use of anaerobic green fluorescent protein versus green fluorescent protein as reporter in lactic acid bacteria

Developmental plasticity of Th17 and Treg cells

Porcine epidemic diarrhea virus: an emerging and re-emerging epizootic swine virus

IFN-lambda preferably inhibits PEDV infection of porcine intestinal epithelial cells compared with IFN-alpha

The regulatory effects of L. plantarum peptidoglycan microspheres on innate and humoral immunity in mouse

Dendritic cell subsets and the regulation of Th1/ Th2 responses

Dendritic cell targeting of Bacillus anthracis protective antigen expressed by Lactobacillus acidophilus protects mice from lethal challenge

TH1 and TH2 cells: different patterns of lymphokine secretion lead to different functional properties

Display of alpha-amylase on the surface of Lactobacillus casei cells by use of the PgsA anchor protein, and production of lactic acid from starch

New generation of oral mucosal vaccines targeting dendritic cells

Lactobacillus plantarum HAC01 regulates gut microbiota and adipose tissue accumulation in a diet-induced obesity murine model

Interaction of lactic acid bacteria with the gut immune system

Capacity of lactic acid bacteria in immunity enhancement and cancer prevention

Regulatory dendritic cells

Engineering the microflora to vaccinate the mucosa: serum immunoglobulin G responses and activated draining cervical lymph nodes following mucosal application of tetanus toxin fragment C-expressing lactobacilli

Immunoprotection against influenza virus H9N2 by the oral administration of recombinant Lactobacillus plantarum NC8 expressing hemagglutinin in BALB/c mice

Lactobacillus plantarum vaccine vector expressing hemagglutinin provides protection against H9N2 challenge infection

Porcine epidemic diarrhea: a review of current epidemiology and available vaccines

High-level, inducible gene expression in Lactobacillus sakei and Lactobacillus plantarum using versatile expression vectors

Efficient priming of CD4T cells by Langerin-expressing dendritic cells targeted with porcine epidemic diarrhea virus spike protein domains in pigs

Natural and edible biopolymer poly-gamma-glutamic acid: synthesis, production, and applications

Tuning constitutive recombinant gene expression in Lactobacillus plantarum

Cloning, expression and antiviral bioactivity of red-crowned crane interferon-alpha

Orally administrated whole yeast vaccine against porcine epidemic diarrhea virus induced high levels of IgA response in mice and piglets

Expression of Bacillus subtilis levanase gene in Lactobacillus plantarum and Lactobacillus casei

Crossprotective efficacy of dendritic cells targeting conserved influenza virus antigen expressed by Lactobacillus plantarum

Alleviation of enterotoxigenic Escherichia coli challenge by recombinant Lactobacillus plantarum expressing a FaeG-and DC-targeting peptide fusion protein

Construction and immunological evaluation of recombinant Lactobacillus plantarum expressing SO7 of Eimeria tenella fusion DC-targeting peptide

Protection of chickens against H9N2 avian influenza virus challenge with recombinant Lactobacillus plantarum expressing conserved antigens

Protective efficacy of Fc targeting conserved influenza virus M2e antigen expressed by Lactobacillus plantarum

Lactobacillus protects the integrity of intestinal epithelial barrier damaged by pathogenic bacteria

Advances in reverse genetics to treat porcine epidemic diarrhea virus

Construction of a bivalent DNA vaccine co-expressing S genes of transmissible gastroenteritis virus and porcine epidemic diarrhea virus delivered by attenuated Salmonella typhimurium