Construction of a Lactobacillus plantarum Strain Expressing the Capsid Protein of Porcine Circovirus Type 2d (PCV2d) as an Oral Vaccine

Abstract

Porcine circovirus type 2 (PCV2) is a pathogenic virus that causes high rates of porcine death, resulting in severe economic losses to the swine industry. In recent years, the prevalence of PCV2d genotype infection in pigs has increased, but most commercially available vaccines were developed against the PCV2a strain and do not ensure complete protection from PCV2d. Here, we first constructed an expression vector for the antigenic ORF2-encoded capsid protein of PCV2d (pLp3050-His₆-tag-capsid). We then utilized Lactobacillus plantarum to express the protein at mucosal sites in orally vaccinated mice. After transducing L. plantarum with pLp3050-His₆-tag-capsid, the expressed protein could be found in cell wall and cell-free supernatant fractions by Western blotting. Using flow cytometry, we found that L. plantarum cells with surface-displayed capsid protein increased with time after SppIP induction. Finally, mice that were orally immunized 18 times with capsid-expressing L. plantarum showed increased levels of capsid-specific sIgA and virus neutralizing activity at mucosal sites, suggesting mucosal immunity had been stimulated by the vaccine. Overall, our findings demonstrate the feasibility and utility of a PCV2d-based vaccine, which may be of great value in porcine agriculture.

Electronic supplementary material

The online version of this article (10.1007/s12088-019-00827-9) contains supplementary material, which is available to authorized users.

Introduction

Porcine circovirus (PCV) is a small, non-enveloped virus with a diameter of 17 nm and a 1.76-kb circular single-stranded DNA genome. The virus has two major types, PCV1 and PCV2, and belongs to the genus Circovirus in the family Circoviridae. PCV1 is considered to be non-pathogenic, whereas PCV2 is associated with postweaning multisystemic wasting syndrome (PMWS) in piglets, which has symptoms that include dyspnea, gauntness, icterus, lymphoid follicle depletion, macrophage infiltration and slow weight gain [1, 2]. PCV2 has five open reading frames (ORFs), indicated as ORF1-ORF5, each encoding a separate functional protein [3–5]. Among these proteins, ORF2-encoded capsid protein is known to readily trigger an immune response in porcine hosts [3, 6].

Four major genotypes of PCV2 (PCV2a–d) can be distinguished by phylogeny [7, 8], which is largely influenced by the characteristically high evolutionary rate of ORF2. As such, overall differences in genome length and sequence among the different genotypes can be attributed to ORF2 [9, 10]. A high rate of genotype shift in PCV2 was recently demonstrated in an epidemiological study conducted in China. This study revealed that from 2002 to 2008, PCV2b was the major genotype in swine, while in 2014, the dominant genotype in swine infections was PCV2d [11]. Similar global genetic shifts from PCV2b to PCV2d have been reported in other regions as well [7, 12]. Importantly, several ORF2-encoded amino acids that appear to be antibody recognition sites differ between PCV2a/b and PDV2d, such as F53I, R59 K and A68 N [10, 13]. Most commercial vaccines, including CIRCOVAC^®, PCV^®/Circumvent^® and Suvaxyn PCV2 One Dose^®, were designed based on the PCV2a genotype [14, 15], and although a PCV2a-based vaccine may partially reduce the serum viral load in PCV2d-infected pigs, complete inhibition can only be achieved with a PCV2d-based vaccine [16, 17]. Compared to PCV2a, PCV2d has additional lysine residues in ORF2, which may interfere with recognition by neutralizing antibodies in PCV2a vaccine-immunized pigs [18]. In addition, PCV2d-challenged pigs appeared to have more severe pathological lesions and viremia symptoms compared to those challenged with PCV2a, suggesting that the virulence of PCV2d is higher than that of PCV2a [19]. Based on all these considerations, an urgent need exists for a vaccine specifically against PCV2d.

Probiotics are live microbial flora that react with non-digestible carbohydrates (i.e., prebiotics) to enhance individual health, including through immunomodulation, nutrient absorption and pathogen inhibition [20, 21]. Based on their broad effects, probiotics have many potential applications. For example, exopolysaccharides have been widely used as additives to modulate rheological properties of food and pharmaceutical products [22, 23]. Also, enzymes produced by probiotics have been used to degrade environment toxicants [24]. Through the advancement of technology, selection or engineering of probiotics with desired properties has become increasingly achievable [25]. Lactobacillus, a species of lactic acid bacteria (LAB) that is categorized as probiotic, is often employed for this purpose. In one particularly striking example, genome engineering of Lactobacillus was applied to change metabolites of gut microbiota to regulate the immune system of the host [25]. Due to its characteristic safety, ease of administration, immune-enhancing function, prevention of metabolic disease and ease of genetic engineering, many studies have successfully utilized recombinant protein-expressing Lactobacillus strains as oral vaccines to prevent various diseases [26–30].

Owing to the increasing prevalence of PCV2d and the incomplete protection provided by PCV2a vaccines in swine, a PCV2d-based vaccine is needed to improve protection against PMWS. In this study, we generated a Lactobacillus plantarum strain that expresses the capsid protein of PCV2d for use as an oral vaccine. After administering the vaccine to mice, the mucosal immune response was evaluated.

Materials and Methods

Plasmids, Bacterial Strains, and Culture Conditions

The plasmids and bacterial strains used in this study are listed in Table 1. Escherichia coli DH5α cells and BL21(DE3) pLysS competent cells were grown aerobically by shaking in Luria–Bertani (LB) broth at 37 °C (Difco Laboratories, USA). L. plantarum was cultivated in de Man, Rogosa, and Sharpe (MRS) broth (Difco Laboratories) or on plates (1.5% agar) containing erythromycin (5 µg/ml) under anaerobic conditions at 37 °C. Ampicillin was added to a final concentration of 100 µg/ml in LB broth for the selection of recombinant E. coli, and erythromycin was added to a final concentration of 200 µg/ml in MRS broth or 5 µg/ml in MRS agar for the selection of recombinant L. plantarum.

Table 1.

Plasmids and strains used in this study

Material	Characteristics	Source or reference
Strains
E. coli DH5α	Competent cells of DNA amplification	Invitrogen
L. plantarum	Wild type strain	[31]
L. plantarum pLp3050	L. plantarum harboring pLp3050 plasmid	This work
L. plantarum pLp3050-capsid	L. plantarum harboring pLp3050-capsid plasmid	This work
Plasmids
pLp3050	Emr, E. coli/L. plantarum shuttle vector based on pSIP401	[32]
pLp3050-capsid	Emr, E. coli/L. plantarum shuttle vector with capsid gene	This work

Cloning

The sequences of PCV2d used in this study are provided in Fig. 1S. The capsid gene was obtained from PCV2d viral DNA by PCR amplification. The His₆-tag-capsid gene was amplified by PCR with the following primers, forward primer: 5′-AGATGTCGACATGCGGGGTTCTCAT-3′ and reverse primer: 5′-GGCCCGGGTCAGTGGTGGTGGT-3′. The primers contained restriction sites for SalI and XmaI, respectively, to create a His₆-tag-capsid gene flanked by these two sites (SalI-His₆-tag-capsid-XmaI). To construct pLp3050-His₆-tag-capsid, the amplified DNA fragment (648 bp) was digested with SalI and XmaI and then ligated to pLp3050 that had also been digested with SalI and XmaI. Under the inducible sakacin P promoter (P_sspA) in pLp3050, protein expression is induced by the addition of Sakacin P (SppIP, 50 ng/ml) (Kelowna International Scientific, Taiwan). A signal peptide gene (Lp_3050) derived from L. plantarum is located downstream of P_sspA in the plasmid. After confirming the sequence of the construct, the plasmid was electroporated into L. plantarum as described previously [33]. L. plantarum transformants were incubated in MRS broth at 37 °C for 3 h, then grown on MRS agar (5 µg/ml erythromycin) at 37 °C for 24 h until colonies were apparent. The correctness of L. plantarum transformants was further confirmed by sequencing.

Detection of His₆-tag-capsid Protein by Western Blot, Flow Cytometry and Confocal Microscopy

The recombinant L. plantarum strain was cultivated overnight, diluted with fresh MRS broth to an OD₆₀₀ of 0.1 and cultured. When the OD₆₀₀ reached 0.3, the culture was induced by the addition of 50 ng/ml SppIP for 3 h [32]. The cell-free supernatant (culture broth) and cell pellet were harvested by centrifugation at 5000× g for 5 min at 4 °C. Proteins in the cell-free supernatant and cell wall were extracted as described previously [34]. The proteins were separated by 10% SDS-PAGE, transferred to polyvinylidene difluoride membranes (Bio-Rad, USA), and then detected by antibodies, including a mouse monoclonal anti-His-tag antibody (Santa Cruz Biotechnology, USA) and a HRP-conjugated anti-mouse IgG (Abcam, Cambridge, USA). The resulting bands were then visualized with a UVP BioSpectrum instrument (Thermo Scientific, USA) using enhanced chemiluminescence (Millipore, USA).

For flow cytometric analysis, the bacteria were incubated with a fluorescein isothiocyanate (FITC)-conjugated anti-His-tag-antibody (Invitrogen, Waltham, MA, USA) at room temperature for 1 h, and then washed with PBS three times. The immunostained bacteria were then analyzed by a CytomicsFC500 Flow Cytometry System (Beckman Coulter, USA). For confocal immunofluorescence microscopy, the immunostained bacterial cells were resuspended in 1 × PBS on a glass slide. Fluorescence was analyzed with a Zeiss LSM 780 spectral confocal microscope (Carl Zeiss, Germany).

Measurement of Capsid Protein Level

Lactobacillus plantarum pLp3050-His₆-tag-capsid was cultured overnight in MRS broth (5 µg/ml erythromycin) at 37 °C. The overnight culture was diluted tenfold with fresh MRS broth to an OD₆₀₀ of 0.1 and incubated at 37 °C until the OD₆₀₀ reached 0.3. Expression of capsid protein was then induced with 50 ng/ml SppIP. After 12 h of SppIP induction, the cultured broth and pellets were collected by centrifugation at 5000× g for 5 min at 4 °C. One hundred microliters of bacterial cells were cultivated on agar plate in 5 µg/ml erythromycin, and the number (CFU) of L. plantarum pLp3050-His₆-tag-capsid was determined by counting colonies. Proteins in cell-free supernatant and the cell wall of recombinant L. plantarum were extracted from 5 ml culture broth and 8 × 10⁸ CFU, respectively, as previously described [24]. Protein extracts were dissolved in 200 µl 1 × sample buffer and boiled. The protein extracts (50 µl) were separated on SDS-PAGE (10% acrylamide) and transferred onto a PVDF membrane at 70 V for 90 min. Proteins were detected by mouse monoclonal anti-His-tag antibody and HRP-conjugated anti-mouse IgG. The bands were visualized on a UVP BioSpectrum. Purified His₆-tag capsid protein was used as a standard. Protein signals were quantified by UVP software, and the amount of capsid protein was calculated by comparing signals from known amounts of purified His₆-tag capsid protein to those from samples.

Immunization and Sampling Protocol

Female 6-week-old BALB/c mice were housed under standard conditions with free access to water and standard diet. Cultures of L. plantarum pLp3050-His₆-tag-capsid and the control, L. plantarum pLp3050, were separately induced with SppIP (50 ng/ml) for 12 h. The cells were harvested by centrifugation, and adjusted to a final concentration of 2 × 10¹¹ CFU/ml in PBS for administration to mice. Mice were divided into five groups (n = 5 each group): (1) PBS, received 0.2 ml of PBS orally; (2) pLp3050, immunized orally with 0.2 ml of L. plantarum pLp3050 (10¹⁰ cells); (3) pLp3050-His₆-tag-capsid, immunized orally with 0.2 ml of L. plantarum pLp3050-His₆-tag-capsid (10¹⁰ cells); and (4) positive control, administered one intraperitoneal injection of inactivated virus vaccine (Lab of Dr. Tsun-Yung Kuo, Ilan University). Mice were immunized by the oral route with the recombinant strains or PBS three times a week, and booster immunizations were administered at two-week intervals until week 12. Fresh fecal samples (200 mg) were collected from mice in each group after the last immunization, and subsequently resuspended in 1 ml of PBS containing protease inhibitor (Thermo Scientific) at 4 °C overnight followed by vigorous vortexing. Supernatant fractions of the fecal samples were collected from each group by centrifugation at 20,000× g, 4 °C for 10 min and then stored at − 80 °C.

Detection of capsid-specific IgA Antibodies in Feces by ELISA

The wells of 96-well microplates were coated overnight at 4 °C with 300 ng purified PCV2d. Supernatants of the fecal samples were added to each well and incubated at room temperature for 1 h, then washed with wash buffer (0.9% NaCl and 0.1% Tween-20). After removing the wash buffer, the wells were incubated with a HRP-conjugated anti-mouse IgA antibody (Novus, USA) at room temperature for 1 h and then washed. The wells were incubated with 100 µl of substrate solution (Kirkegaard & Perry Laboratories, USA) for 15 min and then reacted with 100 µl of 2 N H₂SO₄. Then, the OD₄₅₀ of the plates was measured with a multi-well spectrophotometer (Thermo Scientific).

Virus Neutralization Test

The supernatants from fecal samples were twofold serially diluted in MEM. Diluents from twofold to 512-fold dilutions were incubated with 500 TCID₅₀ of PCV2d for 1 h at 37 °C, and then feces-virus mixtures containing 100 TCID₅₀ PCV2d were incubated with 80% confluent PK-15 (porcine kidney) cells for 72 h at 37 °C. The cells were fixed with 80% acetone at 4 °C for 1 h and washed with PBS. Cells were then incubated with rabbit anti-PCV2 antibody, and subsequently reacted with FITC-conjugated goat anti-rabbit IgG (Invitrogen). After detection using a fluorescent microscope, the neutralizing titers were calculated as the highest dilution of feces capable of reducing 50% virus infection in PK-15 cells.

Statistical Analyses

Results are shown as mean ± standard deviation, and all statistical analyses were performed using GraphPad Prism 6 software (GraphPad Software, USA). Statistical differences between groups were determined using Student’s t test and considered significant at P-values less than 0.05.

Results

Construction of a Recombinant Lactobacillus plantarum Strain for Capsid Antigen Production

To construct a capsid protein-expressing plasmid, we inserted the capsid gene from PCV2d between the SalI and XmaI restriction sites of the E. coli and L. plantarum shuttle vector, pLp3050 [35]; the resulting construct, pLp3050-His₆-tag-capsid, expresses capsid protein with a hexa-histidine-tag (Fig. 1a). Notably, the SalI site (GTCGAC) in this plasmid encodes for valine and aspartic acid and serves as a linker for the fusion of a signal peptide (Lp_3050) and the His₆-tag-capsid. The signal peptide can be cleaved at the cleavage site by a L. plantarum signal peptidase and then secreted from the cells [36]. PCR, utilizing a pair of primers flanking the inserted His₆-tag-capsid gene, showed a band of the expected size of ~ 600 nt, indicating the successful cloning of the His₆-tag-capsid in pLp3050 (Fig. 1b, Lane 1). We further confirmed the pLp3050-His₆-tag-capsid construct by enzymatic digestion with SalI and XmaI to retrieve the target insert (~ 600 bp) from the pLp3050 vector (Fig. 1b, Lane 2).

Fig. 1.

Construction and characterization of recombinant L. plantarum pLp3050 fused with the His₆-tag-capsid insert. a Schematic of the pLp3050-His₆-tag-capsid plasmid. The His₆-tagged capsid was fused to the C-terminus of a signal peptide (Lp3050) via the indicated restriction digestion sites (SalI-XmaI). P_sspA: Sakacin-P inducible promoter; RBS: ribosomal binding site. b Confirmation of the L. plantarum pLp3050-His₆-tag-capsid construct using restriction enzyme digestion. Lane M: 1 kb DNA ladder; Lane 1: capsid gene PCR product; Lane 2: pLp3050-His₆-tag-capsid digested with SalI and XmaI. The arrow (~ 600 bp) indicates the capsid gene fragment

Production of His₆-tag-capsid Protein in Recombinant L. plantarum

To investigate the expression of His₆-tag-capsid protein in L. plantarum containing the pLp3050-His₆-tag-capsid, we induced protein expression in the recombinant strain with SppIP (final concentration, 50 ng/ml) and subsequently detected capsid protein expression in different bacterial fractions by Western blotting. As expected, one major band at ~ 25 kDa was observed in the protein fraction collected from the culture broth, consistent with the predicted size of the His₆-tag-capsid protein (24.5 kDa). This band was not observed in the culture broth of the control L. plantarum pLp3050 strain or the non-induced L. plantarum strain containing the pLp3050-His₆-tag-capsid (Fig. 2a). However, the cell wall fractions showed a single band at ~ 28 kDa, consistent with the size of the Lp3050 signal peptide (4.5 kDa) plus His₆-tag-capsid protein (24 kDa), suggesting that the His₆-tag capsid protein translocates to the cell wall with the aid of the Lp3050 signal peptide (Fig. 2b).

Fig. 2.

Capsid expression in L. plantarum pLp3050-His₆-tag-capsid cells. Western blot analysis of His₆-tag-capsid protein a on the cell wall of L. plantarum pLp3050-His₆-tag-capsid cells and b in the cell-free supernatant. The “+” and “−” indicate the presence or absence of induction using 50 ng/ml SppIP, respectively. c Growth curves of L. plantarum with pLp3050 or pLp3050-His₆-tag-capsid in the presence (+) or absence (−) of 50 ng/ml SppIP at 37 °C

The overexpression of heterologous protein often slows bacterial growth [37]. Consistent with this notion, we found that the non-induced L. plantarum pLp3050-His₆-tag-capsid exhibited a similar growth rate to the SppIP-induced pLp3050 control strain, however, the growth rate of SppIP-induced L. plantarum pLp3050-His₆-tag-capsid was significantly decreased (Fig. 2c).

Detection of His₆-tag-Capsid PROTEIN on the SURFACE of L. plantarum Containing pLp3050-His₆-tag-capsid

Western blot analysis revealed that the His₆-tag-capsid protein was not only expressed in cultured broth but was also found on the cell wall of L. plantarum carrying pLp3050-His₆-tag-capsid (Fig. 2). In order to further confirm this observation, we used confocal fluorescence microscopy to detect the accumulated capsid protein on the surface of the cell wall. After induction with SppIP for 3 h, His₆-tag-capsid protein was detected on the surface of L. plantarum harboring pLp3050-His₆-tag-capsid (Fig. 3a), confirming that induced capsid proteins were translocated and docked on surface of the cell wall. We then examined the portion of recombinant L. plantarum pLp3050 cells expressing His₆-tag-capsid cells on the surface after SppIP induction for 3 h, 6 h, 9 h and 12 h using flow cytometry. The data showed that the proportion of His₆-tag-capsid-expressing cells gradually increased in a time dependent manner after SppIP induction (Fig. 3b). These results suggested that secretion of the capsid protein might be related to the translocation and docking of the capsid on the cell wall of L. plantarum. In order to confirm the identity of the capsid protein on the surface of L. plantarum pLp3050-His₆-tag-capsid after SppIP induction, we used a mouse monoclonal anti-capsid antibody to detect recombinant protein expressed from L. plantarum pLp3050-His₆-tag-capsid on the cell surface. The results suggest that L. plantarum pLp3050-His₆-tag-capsid successfully produced capsid protein on the cell surface (Fig. S2).

Fig. 3.

Examination of His₆-tag-capsid protein expression on L. plantarum cell surface. a Confocal immunofluorescence microscopy examination of L. plantarum cells harboring pLp3050 or pLp3050-His₆-tag-capsid after 50 ng/ml SppIP induction for 3 h. Cells were labeled with mouse monoclonal anti-His₆-tag antibody, followed by FITC-conjugated anti-mouse IgG antibody. b Number of His₆-tag-capsid-expressing L. plantarum cells harboring pLp3050 or pLp3050-His₆-tag-capsid after SppIP induction for 3 h, 6 h, 9 h, and 12 h in total 10,000 cells from each sample. All the recombinant bacterial strains were initially labeled with a monoclonal mouse anti-His-tag antibody and a FITC-conjugated anti-mouse IgG antibody. Then strains were analyzed by flow cytometry. A total of 10,000 cells from each sample were analyzed. Data represent results from two independent experiments

Immune Responses at Mucosal Sites in Mice After Oral Immunization with Recombinant His₆-tag-capsid-Expressing L. plantarum

Since we found maximal surface expression of capsid protein on L. plantarum pLp3050-His₆-tag-capsid after 50 ng/ml SppIP induction at 37 °C for 12 h (Fig. 3b), we applied these induction conditions in an in vivo study. We used 10¹⁰ CFU of L. plantarum pLp3050-His₆-tag-capsid to test mucosal immunogenicity of capsid-expressing L. plantarum in mice [26]. Sufficient yield of capsid protein from the induced bacteria was confirmed before immunizing mice. As shown in Fig. 4a and b, the production of capsid protein was higher on the cell surface than in the cell-free supernatant fraction (culture broth) of L. plantarum pLp3050-His₆-tag-capsid after induction. The amount of capsid protein from 10¹⁰ CFU of L. plantarum pLp3050-His₆-tag-capsid was estimated to be 4.8 μg on the surface of cells, while 1.6 μg capsid was measured in culture broth (cell-free supernatant). Thus, the pellets of 10¹⁰ CFU His₆-tag-capsid-expressing L. plantarum present approximately 4.8 μg capsid for a single immunization. The procedure for oral immunization of mice is illustrated in Fig. 5a. Briefly, 12-h-SppIP-induced L. plantarum pLp3050-His₆-tag-capsid were administered 18 times orally to the mice over a 12-week (84 days) period; PBS (0.2 ml) and L. plantarum pLp3050 (10¹⁰ CFU in 0.2 ml) were used as controls. Since secretory immunoglobulin A (sIgA) is the major antibody in the mucosal tissue of the intestinal tract [38], we collected the feces of mice and analyzed the levels of capsid-specific sIgA. On Day 84, a significant increase in the levels of capsid-specific sIgA was detected in the feces of mice immunized with L. plantarum pLp3050-His₆-tag-capsid (P < 0.001) as compared to the control groups (Fig. 5b).

Fig. 4.

Production of capsid protein in Lactobacillus with pLp3050-His₆-tag-capsid after 50 ng/ml induction for 12 h. a Capsid protein was purified, and different amounts were applied to SDS-PAGE. (lane 1) 31.25 ng capsid; (lane 2) 62.5 ng capsid; (lane 3) 125 ng capsid; (lane 4) 250 ng capsid; (lane 5) protein in cell-free supernatant of L. plantarum containing pLp3050-His₆-tag-capsid, extracted from 5 ml of culture broth; (lane 6) protein in cell-wall extracts from 8 × 10⁸ CFU of L. plantarum with pLp3050-His₆-tag-capsid; (lane 7) protein in cell-free supernatant of L. plantarum harboring pLp3050, extracted from 5 ml of culture broth; (lane 8) protein in cell-wall extracts from 8 × 10⁸ CFU of L. plantarum with pLp3050. b The amount of capsid protein produced by 10¹⁰ CFU of L. plantarum with pLp3050-His₆-tag-capsid was estimated

Fig. 5.

Immunogenicity of His₆-tag-capsid-expressing L. plantarum cells in mice. a A schematic diagram depicting the oral immunization (12 h-SppIP-induced L. plantarum pLp3050-His₆-tag-capsid cells) process in mice. Feces were collected and pooled for the evaluation of capsid-specific IgA levels and for performing the virus neutralization assay. b Detection of capsid-specific sIgA antibodies in fecal samples from indicated groups of mice by ELISA. Fecal samples from mice immunized with one shot of inactivated vaccine via intraperitoneal injection were considered as the positive control. c Neutralizing titers in fecal samples of immunized mice. Neutralizing titers are reciprocals of the highest dilution of feces that resulted in a 50% reduction of virus-infected cells. The assays were performed in triplicate and the data are presented as the mean ± SD; ***P < 0.001; n.s. represents no significant difference between the groups

To further address the protective effect of induced capsid-specific sIgA in L. plantarum pLp3050-His₆-tag-capsid immunized mice, we next evaluated the PCV2d-neutralizing activities. As expected, no neutralizing antibody activity was detected in the feces of mice treated with either PBS or L. plantarum pLp3050. On the other hand, the fecal solution from mice orally vaccinated with L. plantarum pLp3050-His₆-tag-capsid showed a 50% reduced PCV2d infectivity at a dilution factor of 1:55, whereas the positive control (inactivated vaccine injection) showed a 50% reduced PCV2d infectivity at a dilution factor of 1:90 (Fig. 5c). Together, these results suggest that the oral vaccine of capsid-expressing L. plantarum has significant immunogenicity in mice; however, this vaccine may be less effective than traditional vaccines.

Discussion

Most commercially available inactivated PCV2 vaccines and capsid protein-based vaccines have been developed against the PCV2a genotype and exhibit poor neutralizing activity toward PCV2d. Thus, it is necessary to develop a PCV2d-based vaccine to meet the challenge of ever-increasing PCV2d prevalence. Commercially available PCV2 vaccines are mostly delivered by injection. However, there are some drawbacks to the use of injectable vaccines, which require trained personnel and may cause pain or increase risk of transmitting infections compared to oral vaccines [39]. The pLp3050 vector can be used to express and secrete heterologous proteins in L. plantarum, and as such, it has been utilized in LAB oral vaccines [40]. In this study, we found that the antigenic PCV2d capsid expressed by pLp3050-His₆-tag-capsid is partitioned to both the extracellular fraction and the cell wall. Furthermore, mice receiving oral immunization of PCV2d capsid-expressing L. plantarum exhibited antigen-specific immune responses.

Although Lactococcus lactis has been used as vehicle to express PCV2b antigen for oral vaccination in a previous study [41], there is still no PCV2d-specific LAB oral vaccine. In this study, L. plantarum was chosen to express PCV2d antigen because it can reside in the gastrointestinal tract for 7 days, which is much longer than the residence of L. lactis, which only lasts a few hours [42].

The secreted form of PCV2d of capsid protein was expressed on the surface of L. plantarum pLp3050-His₆-tag-capsid and gradually accumulated over time after SppIP-induction (Fig. 3b). Because L. plantarum has only a single cytoplasmic cell membrane and thick cell wall, the heterologous protein only needs to pass through one cell membrane, whereas the relatively thick cell wall enhances opportunities for protein docking [43]. In addition, the accumulation of capsid protein on the cell wall indicates that signal peptidase cleavage activity was limited. As a result of this limited cleavage, the cells may be subject to secretion stress and consequent slow growth, as we observed (Fig. 2c). Therefore, our result fits the idea that overexpression of heterologous proteins in Gram-positive hosts may cause stress and retard the growth of the cells [44, 45].

sIgA plays an important role in immune defense at mucosal surfaces. Our results revealed that oral vaccination of capsid-expressing L. plantarum was able to induce capsid-specific sIgA in the gastrointestinal tract of mice (Fig. 5b). The induced sIgA further showed anti-virus immunogenicity (Fig. 5c), which could be beneficial for protection against PCV2 infection [46]. Since the capsid protein might be degraded under the harsh conditions of the digestive tract (bile salts and low pH [47, 48], the cell wall expression of the His₆-tag-capsid could beneficially limit proteolytic degradation [43].) In a previous study, pLp3050 was used to express antigen from Mycobacterium tuberculosis, which docked on the external bacteria cell wall. This construct could elicit M. tuberculosis antigen-specific sIgA production in mice after oral immunization of recombinant L. plantarum [49]. In our study, L. plantarum expressing PCV2d capsid protein on the external cell wall at an equivalent dosage of 4.8 μg (Fig. 4b) was administrated to mice 18 times orally. Similar to the M. tuberculosis study, our results showed that oral immunization of L. plantarum expressing PCV2d capsid protein could elicit capsid-specific sIgA in mice (Fig. 5b) [49].

The mucosal immunity and anti-PCV2d immunogenicity in mice that was induced by capsid-expressing L. plantarum were less than those derived from a traditional inactivated virus vaccine (Fig. 5b, c). This observation is consistent with the notion that oral immunization typically leads to a modest response, which is lower those that attained by other routes of parenteral administration [50]. Moreover, we found that the level of capsid-specific IgG antibody in serum of orally immunized mice was not different than controls (Fig. S3), indicating that the mucosal immune response did not trigger systemic protection [51]. To strengthen the immune response, improvement of antigen expression level in L. plantarum may be important [52]. In this study, the inducible promoter (PsspA) in pLp3050 did not express capsid protein in the absence of induction. Replacement of the inducible promoter with a constitutive promoter might enhance mucosal immunity, since capsid protein would then be continuously expressed and dock on the surface of L. plantarum cells, increasing the presentation of antigen to intestinal mucosa during the colonization period. While improvements in expression level may increase the utility of our oral vaccine, the L. plantarum pLp3050-His₆-tag-capsid system we developed can still provide a safe and non-invasive mucosal immunization platform for protection from PCV2d.

Conclusion

This study shows that recombinant L. plantarum can express the PCV2d capsid protein on the bacterial cell surface and secrete the protein into the extracellular environment. Moreover, oral vaccination using L. plantarum expressing His₆-tag-capsid could elicit a mucosal immune response in mice, suggesting the potential of this strain as an enhancer of mucosal immunity for the prevention of PCV2d infection. While our in vivo study did not show that a systemic immune response was elicited in mice after oral immunization, additional modifications of this recombinant strain to continuously express capsid protein of PCV2d might be beneficial. Furthermore, combining adjuvants or prebiotics with recombinant Lactobacillus strains might be able to enhance systemic immune response as well.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Compliance with Ethical Standards

Conflict of interest

The authors declare no conflict of interest associated with this research.

Ethical Approval

This animal experiment was approved by the by the Institutional Animal Care and Use Committee of the National Taiwan University (NTU-IACUC/protocol 125/2013). All procedures performed in studies involving animals were in accordance with the ethical standards of the institution or practice at which the study was conducted.