Oral Immunization of Recombinant Lactococcus lactis and Enterococcus faecalis Expressing Dendritic Cell Targeting Peptide and Hexon Protein of Fowl Adenovirus 4 Induces Protective Immunity Against Homologous Infection

Hepatitis-hydropericardium syndrome (HPS) causes severe economic losses in the global poultry industry. The present study aims to explore oral immunization of recombinant Lactococcus lactis and Enterococcus faecalis expressing Hexon protein of fowl adenovirus 4 (FAdV-4). The bacteria L. lactis NZ9000 and E. faecalis MDXEF-1 were, respectively, modified as host strain to deliver truncated Hexon protein (ΔHexon) or ΔHexon protein fusing with dendritic cell (DC) targeting peptide (DC-ΔHexon) on the surface of bacteria. The expression of target protein in L. lactis NZ9000 and E. faecalis MDXEF-1 were detected by western blot. To evaluate the immune responses and protective efficacies provided by the live recombinant bacteria, chickens were immunized with the constructed ΔHexon-expressing bacteria three times at 2-week intervals, then experimentally challenged with hypervirulent FAdV-4/GX01. The results showed that oral immunizations with the four ΔHexon-expressing bacteria (NZ9000/ΔHexon-CWA, NZ9000/DC-ΔHexon-CWA, MDXEF-1/ΔHexon-CWA, and MDXEF-1/DC-ΔHexon-CWA), especially the two bacteria carrying DC-targeting peptide, stimulated higher levels of ΔHexon-specific sera IgG and secretory IgA (sIgA) in jejunal lavage fluid, higher proliferation of peripheral blood lymphocytes (PBLs) and higher levels of Th1/Th2-type cytokines, along with significantly decreased virus loads in liver and more offered protective efficacies against FAdV infection compared with PBS and empty vector control groups (p < 0.01). For chickens in the group MDXEF-1/DC-ΔHexon-CWA, the levels of aspartate transaminase (AST), alanine transaminase (ALT) and lactate dehydrogenase (LDH) in sera, and the virus loads in livers were significantly decreased vs. the other three ΔHexon-expressing bacteria (p < 0.01). The pathological changes in the hearts, livers, spleens and kidneys of chickens in MDXEF-1/DC-ΔHexon-CWA group were relatively slight compared to infection control group and other three ΔHexon-expressing bacteria groups. The rate of protection in MDXEF-1/DC-ΔHexon-CWA group was 90%. The present work demonstrated that cell surface-displayed target protein and immune enhancers in L. lactis and E. faecalis might be a promising approach to enhance immunity and immune efficacy against pathogen FAdV-4 infection.

Hepatitis-hydropericardium syndrome (HPS) causes severe economic losses in the global poultry industry. The present study aims to explore oral immunization of recombinant Lactococcus lactis and Enterococcus faecalis expressing Hexon protein of fowl adenovirus 4 (FAdV-4). The bacteria L. lactis NZ9000 and E. faecalis MDXEF-1 were, respectively, modified as host strain to deliver truncated Hexon protein ( Hexon) or Hexon protein fusing with dendritic cell (DC) targeting peptide (DC-Hexon) on the surface of bacteria. The expression of target protein in L. lactis NZ9000 and E. faecalis MDXEF-1 were detected by western blot. To evaluate the immune responses and protective efficacies provided by the live recombinant bacteria, chickens were immunized with the constructed Hexon-expressing bacteria three times at 2-week intervals, then experimentally challenged with hypervirulent FAdV-4/GX01. The results showed that oral immunizations with the four Hexon-expressing bacteria (NZ9000/ Hexon-CWA, NZ9000/DC-Hexon-CWA, MDXEF-1/ Hexon-CWA, and MDXEF-1/DC-Hexon-CWA), especially the two bacteria carrying DC-targeting peptide, stimulated higher levels of Hexon-specific sera IgG and secretory IgA (sIgA) in jejunal lavage fluid, higher proliferation of peripheral blood lymphocytes (PBLs) and higher levels of Th1/Th2-type cytokines, along with significantly decreased virus loads in liver and more offered protective efficacies against FAdV infection compared with PBS and empty vector control groups (p < 0.01). For chickens in the group MDXEF-1/DC-Hexon-CWA, the levels of aspartate transaminase (AST), alanine transaminase (ALT) and lactate dehydrogenase (LDH) in sera, and the virus loads in livers were significantly decreased vs. the other three Hexon-expressing bacteria (p < 0.01). The pathological changes in the hearts, livers, spleens and kidneys of chickens in MDXEF-1/DC-Hexon-CWA group were relatively slight compared to infection control group and other three INTRODUCTION Fowl adenoviruses (FAdVs) are non-enveloped, double-stranded DNA viruses belonging to the Aviadenovirus genus of the Adenoviridae family. FAdVs are currently divided into 5 species (A-E) and 12 serotypes (1-7, 8a, 8b, [9][10][11] (1,2). FAdV-4 usually infects 3-6 weeks old broilers showing a mortality rate of up to 80%, and causes accumulation of transparent or strawcolored fluid in the pericardial sac and hepatitis (3)(4)(5). Hepatitishydropericardium syndrome (HPS) is one of the typical gross pathological changes in FAdV-4 infected chickens, which has been widely reported in several countries and regions such as Middle East (6), Germany (7), Korea (8), Malaysia (9), and Japan, India and Pakistan (10). In China, HPS caused by novel hypervirulent FAdV-4 has been reported since 2015 (11), which leads to significant economic losses to the broiler poultry industry. Therefore, exploration of vaccine aiming at preventing HPS has become a research hot spot.
Currently, inactivated vaccines and live attenuated vaccines prepared based on the FAdV-4 ON1 strain were characterized to be immunogenic and effective against FAdV-4 infection (12). Considering the potential reversion to virulence for attenuated vaccines, and the potential tumorigenicity of adenoviruses, the novel genetic engineering vaccines are still necessary to be explored. It is generally accepted that mucosal immunity represents the first line of defense, and mucosal vaccination can evoke protective mucosal immune responses against highly contagious virus via the oral route (13). In recent years, lactic acid bacteria (LAB) has been widely used as vehicles to deliver important antigens of pathogens, such as the circumsporozoite protein of Plasmodium falciparum (14), spike protein of SARS-CoV-2 (15), the heavy-chain antigen of Clostridium botulinum serotype A neurotoxin and the Bacillus anthracis protective antigen (16). Previous studies have shown that the expression of FAdV-4 structural proteins, including Hexon, Penton, Fiber 1 and Fiber 2, in Escherichia coli and other expression systems could be used to develop subunit vaccines (17)(18)(19)(20)(21)(22). However, the protective efficiencies of live recombinant LAB delivering structural proteins of FAdV-4 against homologous challenge have not been assessed until now. We and others also reported that live recombinant Lactococcus lactis NZ9000 delivering Eimeria 3-1E protein (23) and avian hepatitis E virus (aHEV) ORF2 protein (24), Enterococcus faecalis displaying Eimeria 3-1E protein (25), Lactobacillus plantarum expressing Eimeria tenella MIC2 protein (26) to some extent provided protective efficacies against poultry disease. Moreover, previous studies have confirmed that oral vaccination with genetically modified bacteria delivering dentritic cell (DC) targeting peptides and immunogenic antigens of pathogens enhanced antigen-specific mucosal immunity against homologous pathogen infection (27)(28)(29)(30).
Based on the previous studies, we speculated that the effective mucosal and humoral immune responses evoked by oral immunization would be a promising means to prevent HPS caused by FAdV. In the present study, L. lactis NZ9000 and E. faecalis MDXEF-1 were used as host strains to express Hexon proteins of FAdV, respectively. Meanwhile, DC targeting peptide was introduced to fuse with anchored Hexon proteins to enhance the antigenic-specific immune responses. Then, the recombinant L. lactis and E. faecalis were used to immunize chickens via oral route, and the immune responses and protection against FAdV-4 challenge were evaluated.

Bacterial Strains, Plasmids, and Virus
Details of the strains and plasmids used in this experiment are listed in Table 1. Lactococcus lactis NZ9000, E. faecalis MDXEF-1 (25), and derivative strains were grown at 30 • C in GM17 broth (M17 containing 0.5% glucose, Luqiao, Beijing) without shaking or on GM17 culture plate with 1.5% agar. E. coli strains were cultured on Luria-Bertani (LB) medium (Hopebiol) at 37 • C. Chicken hepatoma cell line (LMH) (ATCC) was cultured in DMEM medium supplemented with 10% fetal bovine sera at 37 • C in 5% CO 2 . FAdV-4 strain GX01 (FAdV-4/GX01, GenBank no. MH229946.1) used in the present study was isolated from a natural case of HPS by Prof. Guangxing Li in our laboratory. The virus was purified and propagated in the LMH cells.

Animals
Two male New Zealand rabbits weighing about 2 kg were obtained from a rabbit farm. Specific-pathogen-free (SPF) White Leghorn chickens and embryos were purchased from the Harbin Veterinary Research Institute (Heilongjiang, China). 9-day-old SPF chicken embryos were infected with serial 10fold dilutions of FAdV-4/GX01 (0.2 mL) onto each embryo's chorioallantoic membranes and incubated at 37 • C for 10 days to calculate the 50% embryo lethal dose (ELD 50 ) according to the formula of Reed and Muench (32). In this study, animals were maintained on a 12-h light/12-h dark cycle, with unrestricted access to food and drinking water. Animal experiments were performed according to the regulations (SRM-12) of the Ethical Committee for animal sciences in Northeast Agricultural University, Heilongjiang Province, PR China.
pTX8048-SP-ORF2-CWA With fragment encoding signal peptide of secretion protein Usp45 (SP) and ORF2 protein in anchored form, no dendritic cell targeting peptides.
pTX8048-SP-Hexon-CWA With fragment encoding signal peptide of secretion protein Usp45 (SP) and Hexon protein in anchored form, no dendritic cell targeting peptides.
This study pTX8048-SP-DC-Hexon-CWA With fragment encoding signal peptide of secretion protein Usp45 (SP) and Hexon protein in anchored form, contains dendritic cell targeting peptides.

Preparation of Polyclonal Antisera Against Hexon
Virus DNA was extracted from semi-purified viral suspension using TaKaRa MiniBEST Viral RNA/DNA Extraction Kit Ver.5.0 (Takara, Beijing, China) according to the manufacturer's instructions. The target Hexon gene fragments were amplified using primers pair Hexon-F2 and Hexon-R2 ( were induced with the final concentration of 1 mM Isopropyl β-D-Thiogalactopyranoside (IPTG) (Solarbio, Beijing, China) for 7 h at 37 • C when optical density at 600 nm (OD 600 ) reached 0.5 or 0.6. The harvested cells were lysed by sonication, and the expression of Hexon protein was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).
The expression and purification of Hexon protein were carried out as previously described (35). Briefly, the target Hexon protein fused with a 6×His tag was purified using a His-tag Protein Purification Kit (Beyotine Biotechnology) according to the manufacturer's protocol. 2 mg of purified Hexon protein mixed with 2 ml of complete Freund's adjuvant (Sigma, USA) was used to immunize New Zealand white rabbits. The second, third, and fourth immunizations were boosted at 2-weeks intervals as the primary immunization except that the complete Freund's Chβ-actin-R GTAACACCATCACCAGAGTCCA / adjuvant was replaced by incomplete Freund's adjuvant (Sigma, USA). The specificity of antisera was detected by western blot, as previously described (35). The titer of the antisera was determined by enzyme linked immunosorbent assay (ELISA). In brief, 100 µL of purified Hexon protein (10 µg/mL) was added to each well in a 96-well plate and incubated overnight at 4 • C. The plate was washed three times with PBST (PBS containing 0.05% Tween 20), then blocked with 5% skimmed milk at 37 • C for 1.5 h. 100 µL of two-fold serially diluted sera was added to each well and incubated for 1.5 h at 37 • C. After washing, the plates were incubated with HRP-conjugated goat anti-rabbit IgG (Sigma-Aldrich) diluted at 1:5000 for 1 h at 37 • C. 1 mg/mL of o-phenylenediamine and 0.01% H 2 O 2 were added (100 µL per well), and the reaction was stopped by 2 M H 2 SO 4 . The absorbance was measured at 490 nm using a reader (Bio-Rad, USA).

Identification of Recombinant L. lactis and E. faecalis Expressing Hexon Protein
The Hexon gene fragment was amplified by primers pair Hexon-F2 and Hexon-R3 ( Table 2), and subcloned into the BamH I and Kpn I sites of pTX8048-SP-CWA (containing cell wall-anchored sequence) or pTX8048-SP-DC-CWA (containing cell wall-anchored sequence and dendritic cell-targeting peptide) to generate plasmids pTX8048-SP-Hexon-CWA and pTX8048-SP-DC-Hexon-CWA, respectively ( Figure 1A). The above two positive plasmids were confirmed by nucleotide sequence analysis, then were transformed into L. lactis NZ9000 and E. faecalis MDXEF-1 competent cells by electroporation, respectively. The characterized positive bacteria were named NZ9000/ Hexon-CWA, NZ9000/DC-Hexon-CWA, MDXEF-1/ Hexon-CWA, and MDXEF-1/DC-Hexon-CWA, respectively. Meanwhile, recombinant bacteria NZ9000/pTX8048 and MDXEF-1/pTX8048 were used as control. Recombinant positive L. lactis and E. faecalis were cultured to OD 600 values of 0.5, then induced by a final concentration of 5 ng/mL nisin (Sigma-Aldrich) for 4 h. The cell wall-anchored protein samples were prepared as previously described (31). In brief, cell pellets were washed and resuspended in TES (10 mM Tris-HCl pH 8.0, 1 mM EDTA, 25% sucrose). The buffer TES-LMR (TES containing 1 mg/ml lysozyme, 0.1 mg/ml mutanolysin, 0.1 mg/ml RNase) was applied to digest cell walls. After centrifugation, the cell walls were pelleted and removed, and the cell wall-anchored proteins in the supernatant were precipitated with final concentration of 16% trichloroacetic acid (TCA). The target cell wall-anchored protein was resuspended. The prepared protein samples were separated by SDS-PAGE, and electrophoretically transferred to nitrocellulose membranes. The membranes were incubated with rabbit anti-Hexon polyclonal antisera (1:1000), then reacted with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG antibody (1:2000) (Sigma, USA). The immunoreactive protein bands were visualized using an ECL chemiluminescence detection kit (BeyoECL Moon).
To further testify the target Hexon protein was displayed on the surface of recombinant L. lactis NZ9000 and E. faecalis MDXEF-1, indirect immunofluorescence assay (IFA) was performed. Briefly, the recombinant positive L. lactis and E. faecalis were cultured to OD 600 values of 0.5, and then induced by a final concentration of 5 ng/mL nisin (Sigma-Aldrich) for 4 h. The bacterial cultures were washed twice with sterile PBS (pH7.2), followed by centrifugation at 10,000 × g for 10 min. Then the pellets were incubated with rabbit polyclonal antisera against Hexon (1:200) (primary antibody) and goat anti-rabbit IgG conjugated to fluorescein isothiocyanate (FITC) (1:50, Solarbio) (secondary antibody). After washing, the fluorescence on the surface of recombinant bacteria were observed using fluorescence microscope (Leica DM2000).

Oral Immunization and Challenge Experiment
Animal grouping, immunizations, and challenges were displayed in Supplementary Table 1. A total of three immunizations were performed at 2-weeks intervals ( Figure 1B). Tissue samples of liver, spleen, and heart were collected, and sera and jejunal lavage fluid were prepared from five chickens randomly selected in each group at each time point ( Figure 1B). The jejunum tissue (10 cm in length) from chicken in each group was flushed twice with a total of 5 ml of cold PBS (pH 7.2) containg final concentration of 1 mM PMSF (Solarbio). Then the collected jejunal lavage fluid was centrifuged at 800 × g for 10 min at 4 • C, and the supernatants were harvested and stored at −80 • C until assays. On day 14 after the third immunizations, all chickens except those in PBS control group were challenged with FAdV-4/GX01 (200 µL, containing 10 5.2 ELD 50 ) (Supplementary Table 1).

Humoral Immune Responses
The levels of specific IgG in sera and sIgA in jejunal lavage fluid were detected by indirect ELISA as previously described (24). Briefly, Hexon protein (1 mg/mL) was coated on the 96-well plate (100 µL per well). The prepared sera diluted at 1:50 and jejunal lavage fluid diluted at 1:10 were added, respectively, and incubated at 37 • C for 1 h. HRP-conjugated goat anti-chicken IgG or goat anti-chicken IgA (Abcam, USA) were used as secondary antibodies, respectively. 100 µL of ophenylenediamine (1 mg/mL) and 0.01% H 2 O 2 were added to each well, and the reaction was stopped by 2 M H 2 SO 4 . The absorbance was measured at 490 nm using a reader (Bio-Rad, USA). Each sample was tested in triplicate.

Cytokine Levels in Spleen
On day 14 after the third immunization, the levels of chicken IL-2 (ChIL-2), chicken interferon-gamma (ChIFN-γ), chicken IL-4 (ChIL-4), and chicken IL-10 (ChIL-10) in spleens of all euthanized chickens were detected by quantitative real-time PCR (qPCR). Briefly, total RNA was extracted from spleen using Trizol reagent (Invitrogen, Carlsbad, CA). cDNA was synthesized from 1 µg of total RNA using Prime Script RT reagent Kit (TaKaRa Biotech Corp., Dalian, China) according to manufacturer's instructions. qPCR was carried out using SYBR R Premix Ex Taq TM II (Tli RNase H Plus) (TaKaRa Biotech Corp., Dalian, China). qPCR was performed following the minimum information for publication of quantitative realtime PCR experiments (MIQE) guidelines (36). β-actin was used as a reference gene for normalization. Primers pairs used in the present study are listed in Table 2. For each 100-fold diluted cDNA sample, amplification efficiencies of all target genes and the reference gene were similar, and the 2 − Ct method was used to analyze the relative quantification of the target gene (37).

Lymphocyte Proliferation in Peripheral Blood
To determine the proliferation of peripheral blood lymphocytes (PBLs) on day 14 after the final vaccination, peripheral blood of chickens in each group (n = 5) was collected (D) Enzyme linked immunosorbent assay (ELISA) was used to determine titers of rabbit anti-Hexon polyclonal antisera. 100 µL of recombinant Hexon protein (10 µg/mL) was coated. The prepared sera were diluted in 2-fold series and added into each well. The prepared sera from chickens before immunization was used as the negative control. The cut off value was 2.1 times of the OD 490 value of the negative control, the titer of prepared rabbit anti-Hexon polyclonal sera determined by ELISA was 1:2 16 . via wing vein, and PBLs were isolated using lymphocyte separation medium (1.077 g/mL) (Tianjin Haoyang Biological Manufacture, China). The isolated cells were modulated to a final concentration of 5 × 10 6 cells/mL. The proliferation of PBLs were determined with the Cell Counting Kit-8 solution (CCK-8, Bimake). Briefly, cell suspension containing RPMI 1640 medium supplemented with 10% fetal bovine sera was added into a 96-well plate (100 µL per well) with eight duplicates and incubated at 37 • C for 24 h in a 5% CO 2 incubator. Cells were stimulated with 5 µg/mL of recombinant Hexon protein for 48 h, then 10 µL of Cell Counting Kit-8 solution (CCK-8, Bimake) was added into each well to incubate at 37 • C for another 4 h. The CellTiter 96 R AQueous Non-Radioactive Cell Proliferation Assay (Promega, Fitchburg, WI, USA) was evaluated according to the manufacturer's instructions. The value of OD 490nm in each well was measured. Each sample was tested in triplicate. Cells stimulated with 5 µg/mL of concanavalin A (ConA, Sigma) or incubated with cell culture medium was used as a positive and negative control, respectively.

Detection of Hepatic Function and Viral Load in Liver
The levels of aspartate transaminase (AST), alanine transaminase (ALT), albumin (ALB), total protein (TP), and lactate dehydrogenase (LDH) in sera were determined using the corresponding detection kit (Jiancheng Biological Engineering Institute, Nanjing, Jiangsu) according to the manufacturer's protocol. Each sample was detected in triplicate. DNA of FAdV was extracted by TaKaRa MiniBEST Viral RNA/DNA Extraction Kit Ver.5.0 (Takara, Beijing, China), and the virus load in liver was quantified using qRT-PCR as previously described (33). The primer pair 52K-F/52K-R are listed in Table 2.

Necropsy and Histopathology
All the birds were euthanized and necropsied. Several organs, including hearts, livers, spleens, and kidneys, were collected for recording gross pathological changes on 5 days past infection (dpi). The above-collected samples were immersed in 10% neutral buffered formalin, embedded in paraffin, sectioned, and stained with hematoxylin-eosin (HE). The histopathological changes in the prepared slides were observed using a light microscope (Nikon, EX200).

Statistical Analysis
SPSS24 (SPSS/IBM, Chicago, IL, USA) and Prism7.0 (GraphPad Software, La Jolla, CA, USA) was used for one-way ANOVA and Duncan's multiple-comparison procedures of the data. All data are expressed as mean ± standard deviation (SD). A p-value of <0.05 is considered to be statistically significant, and a p-value of <0.01 is considered to be highly significant.

Production of Polyclonal Antisera Against Hexon Protein
Identification of recombinant plasmid pET30(a)-Hexon by PCR showed an expected fragment of 1230 bp (Figure 2A). The FAdV-4 Hexon protein (from 10 to 420 aa) was expressed in E. coli. BL21, and SDS-PAGE displayed a protein band of ∼50 kDa (Figure 2B), which showed positive immunoreaction with rabbit anti-Hexon sera in western blot assays ( Figure 2C). The titer of the prepared rabbit anti-Hexon protein polyclonal antisera was 1:2 16 (Figure 2D).

Lymphocytes Proliferation Responses
On days 14 after three immunizations, PBLs from chickens immunized with four Hexon-expressing bacteria showed significant specific responses to recombinant Hexon protein compared with that from chickens immunized with the NZ9000/pTX8048, MDXEF-1/pTX8048, and PBS (pH7.2) (p < 0.01). However, proliferation responses of PBLs to ConA between any other two groups showed no statistical difference (p >0.05; Figure 6C). Proliferation responses of PBLs to Hexon protein in MDXEF-1/DC-Hexon-CWA and NZ9000/DC-Hexon-CWA group were higher than that in MDXEF-1/ Hexon-CWA and NZ9000/ Hexon-CWA group, respectively (p < 0.01). Notably, proliferation responses of PBLs to Hexon protein in the two groups immunized with Hexon-expressing E. faecalis were higher than those in the two groups immunized with Hexon-expressing L. lactis (p < 0.01; Figure 6C).

Pathological Lesions
The typical gross lesions of HPS were observed after challenging with a lethal dose of FAdV-4/GX01, showing by typical gross lesions, including hydropericardium, splenomegaly, hepatomegaly, and nephritis during autopsy of challenged qRT-PCR was applied to detect copies of FAdV DNA in samples. Each value represents mean ± SD (n = 5). *p < 0.05, **p < 0.01. chickens in each Group (Supplementary Figure 1). Of note, chickens from the four groups immunized with live recombinant Hexon-expressing bacteria presented mild HPS, and the gross pathological lesions in hearts, spleens, livers, and renals were also moderate. The typical histopathological changes of HPS were presented by the myocardial fiber fracture and even dissolution, lymphocytic myocarditis, degeneration and even necrosis of renal tubular epithelium, and severe necrosis disintegration of splenic lymphocytes. However, chickens from the four groups immunized with Hexon-expressing bacteria, especially the group vaccinated with MDXEF-1/DC-Hexon-CWA, displayed mild histopathological changes (Figure 9).

DISCUSSION
The structural proteins of FAdV-4 contain Fiber 1, Fiber 2, Penton, and Hexon (39,40). The subunit vaccines based on Penton and Fiber 2 were recorded to induce immunoprotection FIGURE 9 | Histopathological changes in heart, liver, spleen, and kidney tissues from chickens at 5 dpi. The prepared slides were stained with hematoxylin and eosin (HE). The typical histopathological changes were recorded in the heart, liver, spleen, and kidney tissues of chickens in the challenged control group and empty vector control groups (NZ9000/pTX8048, and MDXEF-1/pTX8048). The hepatic cells displayed vacuolar degeneration and necrosis, and basophilic inclusion bodies were observed in hepatic cells. Myocardial fibers in heart tissues were rupture and necrosis. Renal tubular epithelial cells in kidney tissues were swelling and degenerating. Lymphocytes in the spleen tissues were severe necrosis. The relative slight histopathological changes in heart, liver, spleen, and kidney tissues were present in the groups immunized with Hexon-expressing bacteria (NZ9000/ Hexon-CWA, NZ9000/DC-Hexon-CWA, MDXEF-1/ Hexon-CWA, and MDXEF-1/DC-Hexon-CWA). The hearts, livers, spleens, and kidneys tissues of chickens from the PBS control group showed no histopathological changes. Scale bar = 80 µm. Solid arrows point to lesion areas.
against FAdV infection in chickens (22). However, the immune protection and immune mechanism of the vaccine based on Hexon protein was not reported until now. Hexon protein, one of the most important structural proteins on the surface of adenovirus, contains the type-, group-and subclass-specific antigenic determinants that can stimulate the body to produce efficient neutralizing antibodies. In the present study, the truncated Hexon protein ( Hexon) of the current epidemic FAdV-4 strain (isolated and stored in our lab) spanning amino acids from 10 to 420 was chosen, in which loop1, loop2, and P1 were contained. Prediction using molecular software DNAstar Lasergene 7.1 (DNAS Inc., Madison, WI, USA) showed that Hexon displays high antigenicity. Meanwhile, Hexon protein was fused to dentritic cell (DC) targeting peptide (DC-Hexon) (21,41) with the aim of enhancing Hexonspecific immune responses. Then fusion protein DC-Hexon was inserted into expressing vector (31) in which cell wallanchored (CWA) sequences were contained to display target fusion protein on the surface of recombinant L. lactis and E. faecalis (25,31).
The results of vaccinations suggested that all four recombinant Hexon-expressing bacteria elicited a higher level of sera IgG and sIgA in jejunal lavage compared to the empty control group (p < 0.01), demonstrating that humoral immune responses of immunized chicken were effectively evoked. Moreover, cell wall-anchored Hexon protein fused by DC targeting peptide (DCpep) induced higher sIgA levels in jejunal lavage than cell wall-anchored Hexon delivered by L. lactis and E. faecalis (p < 0.05), which indicating that introduction of DCpep on the surface of bacteria effectively targeted intestinal dendritic cells to enhance antigenic uptake and the subsequent delivery to immune cells. The above results were consistent with our previous report by Ma et al. (31). In addition, live recombinant bacteria MDXEF-1/ Hexon-CWA and MDXEF-1/DC-Hexon-CWA induced more robust immune responses and provided more protective effects than NZ9000/ Hexon-CWA and NZ9000/DC-Hexon-CWA, respectively (p < 0.05). The possible explanations for this result may be that E. faecalis was isolated from ceca and partially colonized in ceca (data not published), and live recombinant Hexonexpressing sustainably stimulated Hexon-specific immune responses. Meanwhile, Hexon-CWA, or DC-Hexon-CWA protein in recombinant Hexon-expressing E. faecalis could be detected without induction of nisin, which suggests that host bacteria E. faecalis MDXEF-1 probably produce nisin or nisin-like substance that continuously induced expression of the target protein. This prediction remains to be further studied and proved in our subsequent research. Previous study has demonstrated that adenovirus settles in the intestinal epithelium at 12 h post-infection via oral route, and the virus can be detected in blood as early as 24 h post-infection (42). The significantly higher levels of Hexon-specific IgG could bind to FAdV in sera to form complexes that could be more easily engulfed by macrophages. The higher levels of sIgA in jejunal lavage can bind to FAdV located at local intestinal epithelium and intercept virus invasion upon initial infection. Therefore, chickens vaccinated with the four recombinant bacteria expressing anchored Hexon protein, especially those carrying DCpep bacteria, induced more effective systemic humoral immune responses and protection against FAdV infection.
As for cellular immune responses, in the present study, the proliferation response of PBLs was more efficient in the groups immunized with four Hexon-expressing bacteria. Moreover, mRNA levels of ChIL-2, ChIFN-γ, and ChIL-4, ChIL-10 in spleens of chickens immunized with four Hexon-expressing bacteria were significantly upregulated, suggesting that both the Th1 and Th2 type responses were enhanced to resist virus infection. The present results could support the above analysis that on day 5 post-infection (dpi), the number of FAdV copies in livers of chickens from NZ9000/pTX8048 and MDXEF-1/pTX8048 group was significantly higher than that in the four Hexon-expressing bacteria groups, especially the two groups of Hexon-expressing E. faecalis.
The average AST, ALT, and LDH levels in sera of chickens in NZ9000/pTX8048, MDXEF-1/pTX8048 group, and infection control group were transiently and significantly elevated on 4 dpi compared to PBS control group and four Hexon-expressing bacteria groups, which indicating that severe pathological changes have occurred in liver (43). Meanwhile, the significant decrease in the levels of TP and ALB also suggested that protein synthesis was affected, which further demonstrated that the liver had been seriously damaged. This result is consistent with the previous research (38). Furthermore, the little-changed indexes of liver function in the four groups of chickens immunized with Hexonexpressing bacteria also proved the protective effects against FAdV infection.
In the present study, on 5 dpi, typical pathological and histopathological changes were accordingly observed in the hearts, livers, and kidneys of chickens in the challenged control group, NZ9000/pTX8048 and MDXEF-1/pTX8048 group, including hydropericardium, splenomegaly, hepatomegaly, and nephritis, all of which are also recorded in other reports (43,44). The pathological changes in hearts, livers, and kidneys of chickens in the four groups immunized with Hexonexpressing bacteria were relatively slight, which suggested that the live bacteria expressing Hexon-CWA protein protected against pathological injury of FAdV targeted organs. However, a 100% protection ratio was not observed in the present study, which may be explained by the fact that the virus strain FAdV-4/GX01 is highly virulent, and chickens were challenged by intramuscular injection way. In our preliminary test, chickens challenged with the lethal dose of FAdV-4/GX01 via intramuscular injection caused mortality of 90-100%, which is close to the reported highest clinical mortality rate (45). The present results showed that the constructed four recombinant bacteria, especially E. faecalis expressing Hexon protein, provided promising protection against the FAdV-4 infection.
Besides, it is worth noting that the viral load in the livers of the challenged chickens in the four groups immunizing with Hexon-expressing bacteria is still high, although it is significantly lower than that in the infection control group. Recently, several studies have shown that Fiber 2 plays a vital role during the invasion of FAdV into host cells (46,47). In some reports, the co-immunization of recombinant LAB expressing several target proteins is more promising (48, 49). Currently, another four kinds of live recombinant LAB expressing Fiber 2 protein have been constructed in our lab, and the subsequent co-immunization of live recombinant E. faecalis expressing Hexon and Fiber 2 would be further explored.
Overall, these results demonstrated that recombinant L. lactis and E. faecalis delivering Hexon protein could evoke strong systemic immune responses against FAdV infection, relieve pathological injury and functional damage in target organs, reduce virus load in liver to some extent, and prevent HPS caused by FAdV infection. The data presented in this report provide references for exploring potential vaccines for HPS.

DATA AVAILABILITY STATEMENT
All data generated for this study are original and included in this article and its Supplementary Material. Any further inquiries may be directed to the corresponding author Dexing Ma.

ETHICS STATEMENT
The protocol of animal experiments was approved by the Animal Ethics Committee of Northeast Agricultural University NEAU-2018-09-0232-12.

AUTHOR CONTRIBUTIONS
DM and ZJ designed the study. GL, CM, XP, XY, and ZJ prepared experimental materials. CM contributed to analytic tools. ZJ and CM analyzed the data and wrote the paper. ZJ, DM, CM, and GL revised the manuscript. All authors reviewed the results and approved the final version of the manuscript.

FUNDING
This study was funded by grants from the National Natural Science Foundation of China (30901061 and 31973003).